Computer and Information Science

Computer and Information Science February, 2009
1
Contents
A Framework for Selection of Information Systems Development Methodologies 3
Masoud Yaghini, Atieh Bourouni & Roozbeh Hesam Amiri
Model Checking the Inconsistency and Circularity in Rule-Based Expert Systems 12
Desheng Xu, Kejian Xia, Dezheng Zhang & Huangsheng Zhang
End-User Computing Satisfaction (EUCS) in Computerised Accounting System (CAS): Which the Critical
Factors? A Case in Malaysia
18
Azleen Ilias, Mohd Zulkeflee Abd Razak, Rahida Abdul Rahman & Mohd Rushdan Yasoa'
Layer Model and Algorithm of Organization Authorization Based on Position Network 25
Feng Ding, Yanzhang Wang & Xin Ye
Chi-Square Test for Anomaly Detection in XML Documents Using Negative Association Rules 35
K. Premalatha & A.M. Natarajan
Research of Sales Contract Management System Based on WEB 43
Hualun Lai, Liangwei Zhong & Jing Ni
Complexity Analysis in Heterogeneous System 48
Kuldeep Sharma & Deepak Garg
Pixel Based Temporal Analysis Using Chromatic Property for Removing Rain from Videos 53
Peng Liu, Jing Xu, Jiafeng Liu & Xianglong Tang
Fault Recovery Mechanisms in Utility Accrual Real Time Scheduling Algorithm 61
Idawaty Ahmad & Muhammad Fauzan Othman
The Fusion of Remote Sensing Images Based on Lifting Wavelet Transformation 69
Wei Zhang, Jinzhong Yang, Xiaohong Wang, Qinghua Yang, Qin Huang & Zhenkun Wang
Individual Differences, Perceived Ease of Use, and Perceived Usefulness in the E-Library Usage 76
Yusliza Mohd.Yusoff, Zikri Muhammad, Mohd Salehuddin Mohd Zahari, Ermy Syaifuddin Pasah &
Emmaliana Robert
Research and Application of the 3D Virtual Community Based on WEBVR and RIA 84
Meng Zhang, Zhihan Lv, Xiaolei Zhang, Ge Chen & Ke Zhang
Framework for Conference Management System 90
Abu Bakar Md Sultan, Abdul Azim Abdul Ghani & Lim Su Fang
Design of the Closed Loop Speed Control System for DC Motor 95
Guoliang Zhong & Liangzhong Jiang
Simulation Research of Extension Control Based on Crane-Double Pendulum System 103
Gang Yang, Weiwei Zhang, Ying Huang & Yongquan Yu
An Analysis of Improving Memory Performance Based on EEG Alpha and Theta Oscillations 108
Tianbao Zhuang, Hong Zhao & Zheng Tang
A Study on Query Optimization for Federated Database Systems 115
Xinhua Xu
Study on the Optimization Method of Select Query Sentence in Standard SQL Language 121
Yuqing Geng & Chunsheng Zhang
Vol. 2, No. 1 Computer and Information Science
2
Contents
Tracking High Speed Skater by Using Multiple Model 126
Guojun Liu & Xianglong Tang
Digital Divide: Issues Facing Adult Learners 132
Ramesh Rao
To Construct Search Engine Analyzer for Electrical Enterprises Based on Lucene 137
Kehe Wu, Xia He, Tingshun Li & Hongyu Tao
The Characteristic and Success Factors of an Organizational Memory Information System 142
Saidatul Rahah Hamidi & Kamaruzaman Jusoff
The Research of the Model of Communication System Depending on the Queuing Theory 152
Min Zhao, Jiyin Sun & Xiaochun Che
Estimate 3D Arm Motion with Hierarchical Limb Model 156
Xuesong Yu & Jiafeng Liu
Study on the Software Development Based on AOP Technology 164
Wei Wang & Wenju Liu
Analysis of BitTorrent on an Open Source VPN Technology to End User 168
Farok Azmat, Md Ikhwan Hamdan & Kamaruzaman Jusoff
Generating a Simple Fractal Graphics in Computer 184
Changhao Shuai & Yong Li
Stochastic Programming Models and Hybrid Intelligent Algorithm for Unbalanced Bidding Problem 188
Xingzi Liu, Liang Lin & Dongran Zang
People Portraits and Cartoon Algorithm Study 195
Weihua Gai & Dakun Zhang
A Summary Sentence Decomposition Algorithm for Summarizing Strategies Identification 200
Norisma Idris, Sapiyan Baba & Rukaini Abdullah
The Design of Intelligent Home Alarming System Based on Wireless Sensor Network Technology 211
Xin Chen & Weizhi Huang
3
A Framework for Selection of Information Systems Development

Methodologies
Masoud Yaghini
Department of Railway Engineering, Iran University of Science and Technology (IUST)
Narmak, PO box 1684613114, Tehran, Iran
Tel: 98-21-7749-1238 E-mail: yaghini@iust.ac.ir
Atieh Bourouni (Corresponding author)
Department of Industrial Engineering, Iran University of Science and Technology (IUST)
Tel: 98-912-5007-402 E-mail: bourouni@ind.iust.ac.ir
Roozbeh Hesam Amiri
Department of Industrial Engineering, Iran University of Science and Technology (IUST)
Tel: 98-912-5356-487 E-mail: amiri_r@ind.iust.ac.ir
Abstract
Information systems are increasingly becoming regarded as crucial to an organisation's success. A development
methodology for an information system is a framework to organize, program and supervise the process of developing an
information system. There many are different methodologies for information systems development. Obviously, no
methodology can claim that it can be applied to any organisation. Therefore, organisations should have an evaluation
framework for selecting an appropriate and efficient methodology. In this paper, we propose an evaluation framework
for selecting information systems development methodology in order to facilitate the development process of
information systems in organisations.
In this work, we will describe and compare different hard and soft systems approaches at first, then in the following
sections, one kind of methodologys classifications is pointed and one methodology from each class is introduced.
Finally, a comparison framework is proposed and methodologies are compared by this framework. The value of this
framework is that, with use of it, organisations can evaluate their development methodology with respect to the key
features of it before implementing any methodology as well as expending extra costs. This framework is theoretical in
nature, and is build based on a review of related literature.
Keywords: Systems methodologies, Information systems, Methodology selection, Methodology comparison
framework
1. Introduction
Methodology is defined by Britain computer society (BCS): (Avison, 2006)
Systems development methodology is a set of recommended means for information systems development or part of it
which is based on logics and specific philosophy. The recommended means often contain of definition of phases,
procedures, activities, rules, techniques, documentations, tools and guidance. It may also include suggestions which
notice on management and organisation, approach and determining and training participations.
In last decades, it is clear for anyone that information systems and their development are important. Information
systems development is a complicated and long process in which different elements collaborate from beginning to end.
It is evident that programming and managing these projects are very hard and even impossible without utilizing a
methodology (Inman, 2006). Many researchers and companies have designed and introduced methodologies in
academic and commercial levels. Each of these methodologies uses a set of tools, models, and specific concepts;
therefore they are successful and applicable in some specific backgrounds. Whereas these methodologies are very
4
numerous, selecting appropriate methodology for applying in a specific system is a very important and difficult process.
Two different approaches can generally be adopted in the process of this methodology selection: (Niaki, 2001)
Select a known methodology and accomplish its steps carefully;
Combine two or many comparatively suitable methodology and adjust them according to specific situation of project.
Methodology comparison is done with the following purposes: (Avison, 2006)
Academic purpose: for better understanding the nature of methodologies (goals, philosophy, characteristics and so on.)
in order to do some classifications and improving coming information systems development
Practical purpose: to select one or more methodologies for specific applications or the whole organisation.
In this paper, at first, hard and soft systems approaches are considered and compared, then one kind of methodologys
classifications is pointed and one methodology from each class is introduced. At the end, comparison frameworks were
proposed and methodologies are compared by this framework.
2. Hard and Soft Systems Approaches
Hard systems thinkers observe the world systematically. They believe that systems have a specific and defined goal and
determined boundary. In other words, there is no search about systems goal and boundary. Hard systems analysis in
mechanical cases or simple biophysical problems would be useful. Hard systems thinkers observe social phenomena as
stable, predictable, and repeatable problems.
On the other hand, soft systems thinkers believe that problems are happening when hard systems thinking is applied in
problem situations which in behavior, action, and human aspects are effective factors. They know phenomena as
dynamic, unpredictable, full of chaos, and immethodical. They do not always observe the world systematically but
sometimes observing the world systematically is known useful. (Checkland, 1981)
In order to distinguishing that the information systems development problem is hard or soft, a framework which is
represented by Harry is used (Harry. 1997):
Definition: Is there any knowing and agreeing about what the problem is?
Boundary: Does the definition of problem enable the clarification of what the problem is, and what it is not?
Separation: Can issues be dismissed as being not part of the problem as a result of drawing the boundary?
Responsibility: Does the definition and separation of the problem make it clear who should be involved in its solution
and who not?
Information: If there is clarity on what the problem is and whose responsibility it is, do we know what information is
required for its solution?
Description: Is it known what a solution would, and would not, look like if all the previous criteria have been satisfied?
If the answer to these questions based on all the above criteria is yes, then the problem can be classified as a hard
problem. If at least one of these answers is no, we are facing a soft or hard uncertainty problem.
According to Checkland, the most important difference between the two concepts is the fact that in hard systems
thinking, the end result would be to implement the designed system, while in soft systems thinking, one would
implement the agreed changes. (Watkins, 2003)
3. Hard Systems Methodologies in Information Systems Development
Different classifications are represented for hard methodologies. In this paper, we gain this taxonomy by combining
some of them: (Harry, 1997; Scott, 2005; Course of information systems development, 2005)
Structured methodologies
Data-oriented methodologies
Process-oriented methodologies
Blended methodologies
Object-oriented methodologies
In the next chapter, each of these classifications is introduced as well as a methodology of each classification.
3.1 Data-Oriented Methodologies
In this group of methodologies, the focus is on the data aspect of system to information systems development and data
flows are very important. Information engineering methodology (IEM) is an example of this group.
The basic framework of information engineering methodology is developed by Clive Finkelestien and James Martin. Its
5
source is the methods which Finklestien suggested for data modeling in the end of 1970s in Australia. In 1981, they
published a book with a title, information engineering, which described principles and its application of methodology.
On the other hand, in the end of 1970s, another group in England innovated a new methodology, system development in
a shared data environment (D2S2). It is effective on IE methodology development. For the moment, different narratives
of these methodologies are represented by numerous companies and organisations which insert on general IE
framework but each of them are different to each other according to circumstances or focuses on parts of methodology.
Martin said regarding this subject: It should not consider that IE is a completely specific methodology. Identical
software engineering, IE is an echelon of methodologies. (Avison, 2006)
Different steps of IE are as follows: (Course of information systems development, 2005)
1) Information strategic planning which is contained existing situation analysis, needs analysis, architecture definition
and information strategic plan collect.
2) Domain analysis which is earning attributes analysis, interaction analysis, existing systems analysis, designing
confirmation and planning.
3) Design and programming system:
Logical designing
Technical designing
4) Build and implementation:
Build: producing and evaluating system
Implementation
Protecting
3.2 Process-Oriented Methodologies
This group, unlike data-oriented methodologies, focuses on existing processes in information system and tries to apply
these processes into information systems development. Yourdon systems method (YSM) is one of these methodologies.
This method is one of the most comprehensive systems development methodologies whose last version is represented in
1993. Yourdon innovated this method in order to eliminate inexpressivenesss which these methods have indicated in
practice. This method has three phases which each of them have some steps. Its steps and phases are as follows:
(Yourdon Inc., 1993)
Phase 1: Feasibility
Interviewing with users
Building DFD
Building ERD
Phase 2: Main model building
Building Enterprise Essential Model (EEM)
Building System Essential Model (SEM)
Phase 3: Performance model building
Building Enterprise Performance Model
Building System Performance Model
3.3 Blended Methodologies
In this group, the attention in information systems development process is on both data and process aspect of the system.
The most important methodology in this group is Structured Systems Analysis and Design Methodology (SSADM).
At first, SSADM was represented to standardize development methods and built systems in governmental centers by
CCTA in 1981. By now, it has changed very much. On of the results would be technologys changes and methodology
week points. Its third edition introduced new techniques to contact users with system in 1986. In version 4.0,
prototyping technique was mooted at first time. Its last version, 4.3, was represented in 1996. (Avison, 2006)
This methodology steps are expressed in five phases and seven steps which are as follows: (Poon, 2005)
Phase 1: feasibility study
Stage 0: feasibility
Phase 2: requirement analysis
6
Stage 1: investigation of current environment
Stage 2: business system options (BSO)
Phase 3: requirement specification
Stage 3: definition of requirement
Phase 4: logical system specification
Stage 4: technical system options
Stage 5: logical design
Phase 5: physical design
Stage 6: physical design
3.4 Object-Oriented Methodologies
By encapsulating processes and data together, defining objects, and capturing the essential building blocks of the system,
Object-Oriented methodologies modularize the system, allowing objects to be defined based on boundaries existing in
the real world system. Such methodologies encourage the analyst to look at the what of a system rather than the how,
holding off the design phase until all objects within the system have been analyzed to their fullest. This category
encompasses such methodologies as: Rational Unified Process (RUP). (Kruchten, 2003)
The roots of rational processes go back to the original spiral model of Barry Boehm. The Rational Approach was
developed at Rational Software in the 1980s and 1990s. The Rational Unified Process (RUP) is based on the integrated
work of three primary methodologists, Ivar Jacobson, Grady Booch and, James Rumbaugh. The Unified Process is
consists of cycles that may repeat over the long-term life of a system. A cycle is consists of four phases: Inception,
Elaboration, Construction, and Transition. Each cycle is concluded with a release, there are also releases within a cycle.
RUP is based on a set of building blocks, or content elements, describing what is to be produced, the necessary skills
required and the step-by-step explanation describing how specific development goals are achieved. (Formal RUP, 2008)
The main building blocks, or content elements, are as follows:
Roles (who) A Role defines a set of related skills, competences, and responsibilities.
Work Products (what) A Work Product represents something resulting from a task, including all the documents and
models produced while working through the process.
Tasks (how) A Task describes a unit of work assigned to a Role that provides a meaningful result.
The process can be described in two dimensions, or along two axes: (Scott, 2005)
The horizontal axis represents time and shows the dynamic aspect of the process as it is enacted, and it is expressed in
terms of cycles, phases, iterations, and milestones.
The vertical axis represents the static aspect of the process: how it is described in terms of activities, artifacts, workers
and workflows.
4. Soft Systems Methodologies in Information Systems Development
These methodologies are focused on soft (human) aspect of systems and supposed that people in organisation are
influenced in systems development process. We classify soft methodologies into two groups: ETHICS methodology and
SSM.
4.1 ETHICS Methodologies
ETHICS is a shortened form of Effective Technical and Human Implementation of Computer- based Systems. The
name of this method means methodology establishment which is designed by Enid Mumford. In order to make system
effective, technology should have many adoptions with social and organisational factors. Therefore, the main goal of
methodology from system design process is high quality of work life and extensive user job satisfaction. (Avison, 2006)
The steps of this methodology are as bellow: (Course of information systems development, 2005)
1) Why change? 2) System boundaries, 3) Description of existing system, 4, 5, and 6) Definition of key objectives and
tasks, 7) Diagnosis of efficiency needs, 8) Diagnosis off job satisfaction needs, 9) Future analysis, 10) Specifying and
weighing efficiency and job satisfaction needs, 11) The organisational design of the new system, 12) Technical options,
13) The preparation of a detailed work design, 14) Implementation, and 15) Evaluation.
4.2 SSM Methodologies
Checkland try to apply systems theory to an applicable methodology. He demonstrated that system analyzers apply their
skills in problems which are not well-defined. These soft, fuzzy, or ill-structured problem situations which are
7
complicated are ordinary in organisations. Checklands methodology was developed in Lancaster University. He
represented soft systems methodology description in 1981. SSM has seven stages which are shown in figure2.
Those stages which above dash line (1, 2, 5, 6 and 7 stages) are real world activities that contain people in problem
situation. Those stages which below dash line (3 and 4 stages) are activities that think about problem situations.
(Checkland & Scholes, 1990)
5. Existing Frameworks Deliberation for Hard and Soft Systems Methodologies Comparison
There are different approaches for information systems development methodologies which we consider some of them:
5.1 Andersen Framework
Andersen identifies a checklist which includes criteria relating to values and society. This checklist is consists of the
following questions:
1) What research paradigms/perspectives form the foundation for the methodology?
2) What are the underlying value systems?
3) What is the context where a methodology is useful?
4) To what extend is modification enhanced or even possible?
5) Does communication and documentation operate in the users dialect, either expert or not?
6) Does transferability exist?
7) Is the societal environment dealt with, including the possible conflicts?
8) Is users participation really encouraged or supported?
This checklist is useful because focuses on those wider issues that are ignored. Furthermore, its a subjective list and
makes some assumptions. (Avison, 2006)
5.2 NIMSAD Approach
NIMSAD is the abbreviated of Normative Information Model- based Systems Analysis and Design and based on the
models and epistemology of systems thinking and to a large degree evaluation and measures a methodology against
these criteria. Three elements of this evaluation are:
The problem situation (the methodology context)
The intended problem solver (the methodology user)
The problem-solving process (the methodology)
The evaluation of elements has a wide-ranging and expressed in terms of the kinds of question that require answers. An
example of questions which are related to the first elements is:
The clients and their understanding, experiences and problems.
A sample of questions concerning the second element is:
The methodology users philosophical views, for example, science or systems-based.
A sample question concerning the third element asks about the way in which the methodology provides specific
assistance for:
Understanding the situation of concern and setting of boundaries.
One feature of this framework is that it recommends that the evaluation be conducted at three stages. First, before
intervention, second, during intervention, and finally, after intervention. These stages are important features of the
framework and introduce the important element of organisational learning. (Avison, 2006)
5.3 Avison and Taylor Approach
Avison and Taylor identify five different classes of situation and appropriate approaches as follows: (Avison, 2006)
1) Well-structured problem situations with a well-defined problem and clear requirements. A traditional SDLC approach
might be regarded as appropriate in this class of situation.
2) As above but with unclear requirements. A data, process modeling, or a prototyping approach is suggested as
appropriate here.
3) Unstructured problem situation with unclear objectives. A soft systems approach would be appropriate in this
situation.
4) High user-interaction systems. A people-focused approach, like ETHICS, would be appropriated here.
8
5) Very unclear situations, where a contingency approach, such as multiview, is suggested.
5.4 Presenting a Framework for Hard and Soft Systems Methodologies Comparison
As explained before, comparing methodologies is a very difficult task. There are different views as many as
methodology writers. It is not necessary that analysts views adopt on users views and in some cases are discordant. In
this framework, we have six basic elements and some of them broken into some sub elements. The headings are not
mutually exclusive and there are interrelationships between them. For example, aspects of philosophy are reflected in
some senses. In the following section, we explain each of these elements:
5.4.1 Philosophy
We know the philosophy as a principle or set of principles that are background of methodology. The question of
philosophy is an important aspect of methodology because choosing areas covering by methodology, orienting to people,
data or systems, the bias or otherwise toward an IT solutions and other aspects are made on the base of methodology
philosophy. Therefore it has influence on all other aspects. It may be explicit but in most methodologies, is implicit
because methodology authors rarely stress on their philosophy. Philosophy consists of paradigm, objectives, domains,
and applications.
5.4.2 Model
The second element of the framework concerns of analysis of the model that the methodology adhere to. The model is
based on the view of methodology of the world. It is an abstraction and representation of important factors of the
information systems or the organisation. Model is a means of communication. It is also a representation that provides
insight into the problem or area of concern.
Models have four different types:
1) Verbal;
2) Analytic or mathematical;
3) Iconic, pictorial or schematic;
4) Simulations.
Models which concern with information systems methodologies are often of the third type.
5.4.3 Tools
A key element of the framework is the identification of tools used in a methodology.
5.4.4 Scope
Scope is an indication of the stages of the life cycle of systems development that methodology covers. Some
methodologies do not follow a life cycle. They may adopt with an iterative, evolutionary or spiral model. Nevertheless,
we suppose that an examination of methodologys scope in relation to the life cycle is useful.
5.4.5 Background
With this element, we can identify each methodology in which environment (academic or commercial) created.
5.4.6 Participations
It contains of users and analysts. Participations determine that can it be undertaken by users or must involved by
professional analysts and what skills levels are required.
Table 1 compares introduced methodologies. In this table and in scope column, there signs are used:
;: Methodology covers this stage with its details.
:: Methodology shows this area but with less details and practice.
: It is introduced in methodology but no procedure, technique, or rule is provided.
6. Conclusion
The Proposed comparison framework is represented with attention to the key features of each methodology. These
features play an important role in successful performing the methodologies. This represented framework can help
organisations to select their suitable methodology based on methodologys features, to apply it in organisation, and to
ensure that methodology will be performed successfully.
Acknowledgement
This study was supported by the institution of "Bonyad Melli Nokhbegan". Their contribution is gratefully
acknowledged.
9
References
Avison, D. & Fitzgerald, G. (2006). Information Systems Development Methodologies, Techniques & Tools, 4th Edition,
McGraw Hill.
Checkland, P. B. (1981). Systems Thinking, Systems Practice. New York, Wiley.
Checkland, P.B. & Scholes, J. (1990). Soft Systems Methodology in Action. John Wiley & Sons, Chichester, UK.
Course information systems development, [Online] available: http://www.unl.ac.uk (May 12, 2008).
DehneRoel, F. & Wieringa, J. (2003). The Yourdon Systems Method and the Toolkit for Conceptual Modeling, Faculty of
Mathematics and Computer Science, March 20.
Formal RUP, [Online] available: http://www.rational.com (June 2, 2008).
Harry, M. (1997). Information Systems in Business. London, Pitman.
Inman, D. (2006). Course Business, Computing & Information management, London South Bank University, [Online]
available: http://www.bcim.lsbu.ac.uk/inmandw (July 10, 2008)
Kruchten, P. (2003). The Rational Unified Process: An Introduction, 3rd Edition, Dec 10, Addison Wesley Professional.
Niaki, A. (2001). Information Systems Creation and Development Methodologies Comparison, ISIran publication,
Tehran, Iran.
Poon, K.H. (2005). SSADM - An Introduction, Office of the Government Chief Information officer.
Swindells, N. & Moreno, A. (2005). Information Engineering - A New International Standard for the Long-term
Archiving of Property Data. Ensuring Long-term Preservation and Adding Value to Scientific and Technical data (PV
2005), The Royal Society, Edinburgh, 21-23, November, 2005.
Scott, W. (2005). A Managers Introduction to the Rational Unified Process (RUP), December 4, Ambysoft.
Watkins, J. (2003). Multi methodology: An Alternative Management Paradigm to Process Quality Improvement,
Appendix C.
Yourdon Inc., (1993). Yourdon System method: Model- Driven Systems Development, Yourdon Press.
10
Table 1. Methodology comparison and selection framework
P
a
r
t
i
c
i
p
a
t
i
o
n
s

S
y
s
t
e
m

u
s
e
r
s

a
n
d

e
x
p
e
r
t
s

S
y
s
t
e
m
s

u
s
e
r
s

(
e
x
p
e
r
t
s

a
s

c
o
u
n
s
e
l
o
r
)

P
r
o
f
e
s
s
i
o
n
a
l

I
S

d
e
v
e
l
o
p
e
r
s

P
r
o
f
e
s
s
i
o
n
a
l

I
S

d
e
v
e
l
o
p
e
r
s

P
r
o
f
e
s
s
i
o
n
a
l

I
S

d
e
v
e
l
o
p
e
r
s

P
r
o
f
e
s
s
i
o
n
a
l

I
S

d
e
v
e
l
o
p
e
r
s

B
a
c
k
g
r
o
u
n
d

A
c
a
d
e
m
i
c

A
c
a
d
e
m
i
c

C
o
m
m
e
r
c
i
a
l

C
o
m
m
e
r
c
i
a
l

C
o
m
m
e
r
c
i
a
l

A
c
a
d
e
m
i
c

S
c
o
p
e
Maintenance
- - -
- -
Evaluation
- - -
:
-
:
Implementation
- - -
:
-
;
Examination
- - -
;
-
;
Planning
- - -
:
-
:
Physical design
-
: ; ; ; ;
Logical design
-
; ; ; ; ;
Analysis ; ; ; ; ; ;
Feasibility : : ; : ; :
Strategy ;
-
; :
T
o
o
l
s

- -
A
d
v
i
c
e

t
o

u
s
e
u
s
e
b
e
c
a
u
s
e

o
f

c
o
m
p
l
e
x
i
t
y

a
n
d
t
i
m
e
A
d
v
i
c
e

t
o

u
s
e
u
s
e
b
e
c
a
u
s
e

o
f

c
o
m
p
l
e
x
i
t
y

a
n
d
t
i
m
e
M
a
j
o
r

M
o
d
e
l
R
i
c
h
p
i
c
t
u
r
e

S
o
c
i
a
l
-
t
e
c
h
n
i
c
a
l

m
o
d
e
l
D
F
D

E
R
D
D
F
D

O
b
j
e
c
t

M
o
d
e
l
P
h
i
l
o
s
o
p
h
y
A
p
p
l
i
c
a
t
i
o
n

S
i
t
u
a
t
i
o
n
s
i
n
c
l
u
d
e

h
u
m
a
n
a
c
t
i
v
i
t
i
e
s

A
n
y

s
y
s
t
e
m

w
i
t
h

a
n
y

s
i
z
e

B
i
g

S
y
s
t
e
m

B
i
g

S
y
s
t
e
m

B
i
g

S
y
s
t
e
m

B
i
g

S
y
s
t
e
m

D
o
m
a
i
n

i
n
f
o
r
m
a
t
i
o
n

a
n
d

o
r
g
a
n
i
s
a
t
i
o
n
a
l

s
y
s
t
e
m
s

p
l
a
n
n
i
n
g

N
o
t

w
e
l
l
-
d
e
f
i
n
e
d

o
r

u
n
s
t
r
u
c
t
u
r
e
d

D
e
t
e
r
m
i
n
e
d

p
r
o
b
l
e
m

s
o
l
v
i
n
g
m
e
t
h
o
d
o
l
o
g
y

D
e
t
e
r
m
i
n
e
d

p
r
o
b
l
e
m

s
o
l
v
i
n
g
m
e
t
h
o
d
o
l
o
g
y

S
t
r
a
t
e
g
i
c
i
n
f
o
r
m
a
t
i
o
n

a
n
d

o
r
g
a
n
i
s
a
t
i
o
n
a
l

s
y
s
t
e
m
s
p
l
a
n
n
i
n
g
D
e
t
e
r
m
i
n
e
d

p
r
o
b
l
e
m

s
o
l
v
i
n
g
m
e
t
h
o
d
o
l
o
g
y

D
e
t
e
r
m
i
n
e
d

p
r
o
b
l
e
m

s
o
l
v
i
n
g
m
e
t
h
o
d
o
l
o
g
y

O
b
j
e
c
t
i
v
e
s

F
o
r

c
o
m
p
l
e
x

s
i
t
u
a
t
i
o
n
s

a
n
a
l
y
s
i
s

I
m
p
r
o
v
e

w
o
r
k

l
i
f
e

q
u
a
l
i
t
y

a
n
d

i
n
c
r
e
a
s
e

u
s
e
r

j
o
b

s
a
t
i
s
f
a
c
t
i
o
n

C
o
m
p
u
t
e
r
i
z
e
d

i
n
f
o
r
m
a
t
i
o
n

s
y
s
t
e
m
s

d
e
v
e
l
o
p
m
e
n
t

C
o
m
p
u
t
e
r
i
z
e
d

i
n
f
o
r
m
a
t
i
o
n

s
y
s
t
e
m
s

d
e
v
e
l
o
p
m
e
n
t

C
o
m
p
u
t
e
r
i
z
e
d

i
n
f
o
r
m
a
t
i
o
n

s
y
s
t
e
m
s

d
e
v
e
l
o
p
m
e
n
t

C
o
m
p
u
t
e
r
i
z
e
d

i
n
f
o
r
m
a
t
i
o
n

s
y
s
t
e
m
s

d
e
v
e
l
o
p
m
e
n
t

P
a
r
a
d
i
g
m
S
y
s
t
e
m
i
c

S
y
s
t
e
m
i
c

S
c
i
e
n
t
i
f
i
c

S
c
i
e
n
t
i
f
i
c

S
c
i
e
n
t
i
f
i
c

S
c
i
e
n
t
i
f
i
c

A
p
p
r
o
a
c
h

S
o
f
t
S
o
f
t
H
a
r
d
H
a
r
d
H
a
r
d
H
a
r
d
S
S
M

E
T
H
I
C
S

Y
S
M

I
E
S
S
A
D
M

R
U
P
Computer and Informatiion Science
F
1
The problem
situation:
unstructured
2
The probl
situation
expresse
Figure 2.

igure 1. RUP G
lem
n:
ed
3
Root definition of
relevant systems
7
Action to impro
the problem
situation
Compa
w
Soft Systems M

General Struct
f
4a
Formal system
concepts
ove
m
5
arison of 4
with 2
Methodology S

ture (Scott, 200
ms
4
Conceptu
models
6
Feasible,
desirable
changes
Stages (Check

05)
4b
Other systems
thinking
ual
s
kland, 1981)
Feb
s
bruary, 2009
11
12
Model Checking the Inconsistency and Circularity

in Rule-Based Expert Systems
Desheng Xu (Corresponding author)
Department of Computer science and technology
University of Science and Technology Beijing
106 mailbox 30#Xueyuan Road, Haidian District, Beijing 100083, China
Tel: 86-10-6239-0097 E-mail: xudesheng98@163.com
Kejian Xia
30#Xueyuan Road, Haidian District, Beijing 100083, China
Dezheng Zhang
Huangsheng Zhang
The research is financed by the National High-Tech Research and development Plan of China (863) under Grant
(No.2007AA01Z170); the Beijing Natural Science Foundation of China under Grant (No.4062022)
Abstract
In the past several years, various techniques were proposed to analyze various types of structural errors, such as
inconsistency (conflict rules), and circularity (circular depending rules), of rule-based systems. Model checking is a
technique for the verification of temporal logic specifications in state transition systems. In this paper, we model the
rule-based systems as finite state transition systems and express consistency and acyclic as Linear Temporal Logic (LTL)
logic formula and then use the technique of model checking to detect inconsistency and circularity in Rule-Based
Systems with the model checker NuSMV.
Keywords: Model checking, Inconsistency, Circularity, Rule
1. Introduction
Expert systems have been widely used in many real world applications. The central part of an expert system is a rule
base that codifies the knowledge from domain experts in the form of inference rules. Often these inference rules are
built into a rule base incrementally over years and subject to frequent refinements. Due in part to the above construction
process of a rule base and in part to the different and even conflict views provided by domain experts, a rule base can
contain many structural errors. According to (D.L. Nazareth. 1989), the typical types of structural errors include
13
inconsistency (conflict rules), incompleteness (missing rules), redundancy (redundant rules), and circularity (circular
depending rules).But we just focus on the inconsistency and circularity in this paper.
Many different techniques have been proposed to detect the above structural errors in rule-based systems (a quite
comprehensive list of references can be found in (M. Ramaswamy, et al., 1997)). Earlier work mainly focused on
detecting structural errors by checking rules pair-wisely. Recent work aimed at detecting structural errors caused from
applying multiple rules in longer inference chains. The majority of the recent verification techniques involve using
some graphical notation such as Petri nets and graphs. Several of the above approaches cannot detect structural errors
accurately and report spurious errors when compound antecedent clauses overlap. The approach in (D. Zhang and D.
Nguyen,1994) could only detect structural errors matching a set of pre-defined syntactic patterns. The approaches in (G.
Valiente,1993) and(S. J.H. Yang, et al.1998)did not address inconsistency errors. In (M. Ramaswamy, S. Sarkar, and Y.S.
Chen, 1997), an adjacency matrix technique was used, which has a higher computational cost both in terms of space
(the sparseness of typical adjacency matrices) and time (the number of addition and multiplication involved in matrix
operations).
Model checking (William Chan,1998) is a technique for the verification of temporal logic specifications in state
transition systems. In this paper, we present a model to model the rule set of rule-based systems as state-transition
systems and express consistency and acyclic as LTL logic (A.Pnueli . 1981) formula and then use the technique of
model checking to detect inconsistency and circularity in Rule-Based Systems with the model checker NuSMV. This
technique is simple, efficient, and automated. We highlight the unique features of this new approach and demonstrate its
application through an example by the model checker NuSMV.
2. Model checking
Model checking (William Chan,1998) is an automatic approach to formal verification based on state exploration. This
verification is performed by software tools, which given a state transition system and a property, model checking
algorithms exhaustively explore the state space to determine whether the system satisfies the property. Fig. 1 is a
schematic of the process of model checking, a model of the specification and a property are fed to a model checker. The
result is either a claim that the property is true or else a counterexample (a sequence of states from some initial state)
falsifying the property. In practice, counterexamples often provide valuable debugging information, and can be used by
the user to modify the model, or the property checked. This iterative process is inherent in our work. In this context the
model is a state transition system and the specification is formalized with temporal logic which pinpoints desired
behavior over paths and states in the model.
In the rest of Section 2, we give an overview of the basics of LTL model checking and NuSMV, the model checker that
we used.
2.1 The LTL Model Checking Problem
In temporal-logic model checking, we are given a state transition system, which models a software or hardware system,
and a property specified as a formula in a certain temporal logic, and determine whether the system satisfies the formula.
A common logic for model checking is the Linear Temporal Logic (LTL), which extends propositional logic with
certain temporal operators. Typical formulas include the following(The symbol p is an atomic proposition, ):
F p : p holds eventually or sometime in the future.
G p : p holds globally or always in the future.
Formally, a state transition system <P, R, I> consists of a set of states P, a state transition relation R _ PP, and a set
of initial states I_P. A path is an infinite sequence of states such that each consecutive pair of states is in R. The set of
states P is often encoded by a set of state variables, such that each state corresponds to some valuation for the variables
and no distinct states correspond to the same valuation (that is, the mapping of P to the variable valuations is
one-to-one).
For simplicity we discuss just a subset of LTL, namely the subset with only the temporal operators F and G, which are
sufficient to understand our examples. We can recursively define this restricted class of LTL formulas as follows: We
say that a proposition is any Boolean combination of predicates on the state variables. A formula is either a proposition,
a Boolean combination of formulas, or of the form F f, G f, where f is a formula. And we use the symbolic ! , | , &
to denote the logic NOT, OR, AND , respectively. Each formula is evaluated at some state q. A proposition holds at q if
q satisfies the proposition.
The system satisfies a formula if the formula holds at all initial states. If not, a model checker typically attempts to find
a counterexample. For example, if the formula !(F P5=1 & F P5=0) is false, a counterexample is a finite path starting at
some initial state sometime satisfies the proposition P5=1 and sometime P5=0.
Readers familiar with temporal-logic model checking may notice that, although a LTL formula is usually interpreted
over a Kripke structure which is a model for the representation of a finite-state concurrent system. The model consists of
14
the set of states, transition relation and labeling (which gives semantics to the structure). In our definition, a state is not
explicitly labeled, but can be thought as being labeled implicitly by its corresponding state-variable valuations. This
more restricted formulation is sufficient for our presentation.
2.2 NuSMV
Verification of the inconsistency and circularity problem in Rule-Based Systems has been carried out by the symbolic
model checker NuSMV(R. Cavada, et al.2005) that originated from reengineering, reimplementation and extension of
CMU SMV, the original BDD-based model checker developed at CMU. NuSMV allows for the representation of
specifications expressed in CTL and Linear Temporal logic(LTL) using BDD-based and SAT-based model checking
techniques. The primary purpose of the NuSMV input language is to describe the transition relation of a finite Kripke
structure. The input file describes both the model and the specification (with possible fairness constraints). The states
are defined by a collection of state variables, which may be of Boolean or scalar type. The transition relation of the
Kripke structure is determined by a collection of parallel assignments, which are introduced by a keyword ASSIGN.
The semantics of assignment in NuSMV is similar to that of a single assignment data flow languages. A program can be
viewed as a system of simultaneous equations, whose solutions determine the next state. When a set is assigned to a
variable, the result is a nondeterministic choice among elements of the set. The case statements of NuSMV are
evaluated from the top down: if several expressions to the left of a : are true, then the command corresponding to the
first, top-most true expression will be executed.
3. Inconsistency and circularity problem in Rule-Based Systems and our method
3.1 Model the Rule Base
An inference rule (clause) has the following general form (William Chan, et al. 1998): P Q, or IF P THEN Q, where
P and Q are called premise and conclusion respectively. P (or Q) can be an atomic propositional logic formula (a
proposition or its negation) or a compound propositional logic formula containing multiple propositions and logical
connectives: ., v). In the following discussion, we use lower case letters to denote atomic formulas and capital letters
to denote compound formulas. We assume that all the inference rules in a rule base are syntactically valid propositional
logic formulas (syntax errors are easily detected by a parser). Furthermore, we assume that P and Q are in conjunctive
normal form P1. . Pm and Q1. .Qn respectively, in which each Pi (1 s i s m) (or Qj (1 s j sn)) is a disjunction
of propositional symbols and their negations, i.e. p1v vpk.. The above assumption does not limit the error detecting
capability of our approach since any propositional logic formula can be easily and mechanically transformed into a
semantically equivalent conjunctive or disjunctive normal form. The following example is modified from an example in
[11]. A rule base R is defined as follows:.
R={ r1: p1 p2 . p5
r2: p2 p5
r3: p3 p1 . p2
r4: p1 v p4 p5
r5: p5 p4 }
Formally, we model the rule-based system we describe as above as a finite-state transition system <P, R, I>consists
of a set of states P, a state transition relation R _ PP, and a set of initial states I_P, where P is a set of state variables
{p1,p2,pn}(n is the number of atomic propositions) which correspond to the atomic propositions appear in the rule
base, and all of these variables of enumeration type {1, 0, neither}.When variables pi=1, it means pi is true, and when
pi=0,it means pi is true, if pi=neither, it means neither pi nor pi is true. In our definition, a state transition relation is not
explicit presentation. The initial states I is also not explicit presentation. We use the variable InitTrue of enumeration
type {r1, r2,rm, neither}(m is the number of rules in the rule base) to denote the initial state implicitly, when InitTrue
= ri, it indicates that the premise of rule ri is true. Initially, the initial value of state variables pi is neither, and the initial
value of InitTrue is a nondeterministic choice among elements of the set {r1,r2rm }. If InitTrue = ri, it indicates that
rule ri is enable (Doing this is similar to the procedure we input some facts enable rule ri), and then we start the
reasoning to change the values of state variables according to the rules in the rule base. If the value of state variable pi is
non-neither and there no other rules assign 1 or 0 to it, then we reset pi to neither. And we also set the subsequent value
of InitTrue to be neither. This is the state transition relation of this model. For the rule base R in Fig.1, suppose the
initial value of InitTrue is r3, then the state transition is illustrated in Fig.4.This is the way how we model the rule-based
system as a finite-state transition system.
3.2 Inconsistency
Inconsistency results in conflict facts and must be resolved for correct functioning of an expert system. There are two
cases of contradiction.
15
Case 1: When we start reasoning from a rule, sometime pi is true and some time pi is true, then the contradiction
occurs. So, if there exists both pi and pi in the rule base, we should check whether the LTL formula !(F pi =1 & F pi=0)
be satisfied or not in the model. For the instance we presented in Fig.1 , initially, we enable rule r1,then p5 and p2 is
valued to 1. and then rule r2 is fired by p2, as the result, p5 is evaluated to 0.This results a contradiction.
Case 2 : If there are some rules make pi true and some make pi true, then the premises make pi true and pi true
respectively should not be true at the same time. So we use LTL formula !F((p4=1| p1=1) & (p2=1)) to check the
example we given whether satisfies this property.
3.3 Circularity
Circularity occurs when several inference rules have circular dependency. Circularity can cause infinite reasoning and
must be broken. An example of circularly dependent rules is as follows:
r1: p q
r2: q p
If a rule is a part of a circle, then the state variables such as pi form the rule will be assigned to 1 or 0 infinite often. So,
if there no circularity in the rule set, then all of state variables pi should always equal to neither in the future. We use the
LTL formula F G(p1=neither & p2=neither & & pn=neither) to express this property. If this formula cant be
satisfied, we conclude there must be some circularity occurs.
4. The experiment about the example presented in 3.1
We use the NuSMV to check the example presented in 3.1,the source code we used as below:
MODULE main
VAR
p1:{0,1, neither};

p5:{0,1, neither };
InitTrue:{r1,r2,r3,r4,r5, neither };
ASSIGN
init(p1) := neither;

init(p5) := neither;
init(InitTrue):={r1,r2,r3,r4,r5};
next(InitTrue):= neither;
next(p1):= case
InitTrue=r1 | InitTrue=r4 | p3=1 : 1 ;
p1 != neither : neither;
1 : p1;
esac;
next(p2):= case
InitTrue=r2 | p1=1 | p3=1: 1 ;
1: p2;
esac;
next(p3):= case
InitTrue=r3 : 1 ;
16
1: p3;
esac;
next(p4):= case
InitTrue=r4 | p5=1 : 1 ;
1: p4;
esac;
next(p5):= case
InitTrue=r5 | p4=1 | p1=1 : 1 ;
p2=1 : 0;
1: p5;
esac;
LTLSPEC !(F p5=1 & F p5=0);
LTLSPEC !F((p4=1| p1=1) & (p2=1));
LTLSPEC F G(p1=neither & p2=neither & p3=neither & p4=neither & p5=neither);
The NuSMV shows that all the three LTL formulas are false, and produce the counterexamples as Fig.2 indicates rule r2
is inconsistent with rules r1 and rules {r3,r4}, rules r4 and r5 are in self-loop.
5. Conclusion and future work
In this paper we have proposed a finite-state transition system model which can be used for describing the behaviors of
the rule-based system. We have focused on the inconsistency and circularity problem of a given rule base, and identified
two sorts of inconsistency problems and circularity problem. We also present two kinds of LTL formulas and a kind of
LTL formula for checking the consistency and circularity of the rule base respectively, and carry a experiment to explain
our method. Our technique has the following advantages:
(1) It is general and is capable to detect all potential inconsistency and circularity errors without imposing any
restrictions on the form of rules;
(2) It can produce the results automated, The result is easy to understand and we can locate the errors easily with the
help of the counterexamples;
(3) It is easily extensible to predicate forms such that each pi is a predicate instead of simple proposition. Although
describing of transition relation between states becomes more complicated and requires more careful.
But since the conversion from rules to the model we input into the model checker is manual work, we will develop a
software tool to implement the method we proposed automatic in the future. And from theoretical point of view, to
check the Inconsistency and circularity in rule base, we have to check the validity of each possible state space. The
possible state spaces in our method is 3
n
(n is the number of state variables), which is exponential. So, unfortunately, the
size of the state space is exponential in the size of the rule base, resulting in the state explosion problem. But since many
methods were proposed to deal with this problem, we still can do some research when the rule base is too huge to apply
the method proposed in this paper directly.
References
A.Pnueli . (1981). A temporal logic of programs. Theoretical Computer Science, 13:45-60, 1981.
D. Zhang and D. Nguyen, (1994). PREPARE: A Tool for Knowledge Base Verification, IEEE Trans. On Knowledge
and Data Engineering, vol. 6, no., 6, Dec. 1994, 983-989.
D.L. Nazareth and M.H. Kennedy, (1991). Verification of Rule-Based Knowledge Using Directed Grphs, Knowledge
Acquisition, vol. 3, 1991, 339-360.
D.L. Nazareth, (1993).Investigating the Applicability of Petri Nets for Rule-Based System Verification, IEEE Trans.
on Knowledge and Data Engineering, vol. 4, no., 3, 1993, 402-415.
D.L. Nazareth. (1989), Issues in the Verification of Knowledge in Rule-Based Systems, Intl J. of Man-Machine
Studies, vol. 30, 1989, 255-271.
Computer
G. Valient
Intl J. of E
M. Ramas
IEEE Tran
R. Agarwa
Intl J. of M
R. Cavada
S. J.H. Ya
Internation
William C
(1998). M
498-520, J
X. He, W.
and Softwa
and Informati
te, (1993). Ver
Expert Systems
wamy, S. Sark
ns. on Knowled
al and M. Tann
Man-Machine
a, A. Cimatti, E
ang, A.S. Lee
nal Computer a
han, Richard J
Model checking
July 1998.
Chu, and H. Y
are Technology
ion Science
rification of K
s, vol. 6, no. 3
kar, and Y.S. C
dge and Data E
niru, (1992).
Systems, vol.3
E. Olivetti, M.
e, W.C. Chu, a
and Software A
J. Anderson, P
g large softwa
Yang. (2003).
y, vol. 45, no.1
Figure 2
Specifi
Mo

Knowledge Ba
, 1993, 341-35
Chen, (1997).
Engineering, v
A Petri Net Ba
36. 1992, 447-4
Pistore, and M
and H. Yang,(
Application Co
Paul Beame, St
are specificatio
A New Appr
10, 2003, 663-
Figure 1. M
2. One of the co
fication
odel

ased Redundan
55.
Using Directe
vol. 9, no., 2, 1
ased Approach
468.
M. Roveri. (200
(1998).Rule
onference (CO
teve Burns, Fra
ons. In IEEE
roach to Verify
670.
Model checking
ounterexample
Model
checker

ncy and Subsu
ed Hypergraph
1997, 221-237.
h for Verifying
05).NuSMV 2.
Base Verificat
OMPSAC98),
ancesmary Mo
Transactions
fy Rule-Based
g at a glance.
es produced by
Tr
Fals
Count

umption Using
hs to Verify Ru
.
g the Integraity
.4 User Manua
tion Using Pe
Vienna,Austri
odugno, David
on Software E
Systems using
y NuSMV.
rue or
se with a
erexample
Feb
g Graph. Trans
ule-Based Expe
y of Productio
al, 2005.
etri Nets, Pro
ia, 1998.
Notkin, and J
Engineering 2
g Petri Nets,
bruary, 2009
17
sformations,
ert Systems,
n Systems,
oc. Of 22nd
on D. Reese.
24(7), pages
Information
18
End-User Computing Satisfaction (EUCS)

in Computerised Accounting System (CAS):
Which the Critical Factors? A Case in Malaysia
Azleen Ilias
Mohd Zulkeflee Abd Razak
Rahida Abdul Rahman
Mohd Rushdan Yasoa'
School of International Business and Finance Labuan
Universiti Malaysia Sabah
Labuan International Campus
87000, F.T Labuan, Malaysia
Tel: 60-8746-0517 E-mail: neelza80@yahoo.co.uk
Abstract
This paper is to examine critical factors; content, accuracy, format, ease of use, timeliness, satisfaction with system
speed and system reliability in End-User Computing Satisfaction (EUCS) that influence most end-users satisfaction.
The research was conducted using a set of questionnaire consist of seven factors; content, accuracy, format, ease of use,
timeliness, satisfaction with system speed and system reliability to measure end-users satisfaction. In addition, this
study covered 90 end-users of Computerised Accounting System (CAS) in finance department from 62 Responsibility
Centres. This study is analyzed with reliability analysis, descriptive analysis and multiple regressions. Overall, this
study indicates that most of end-users are almost satisfied with Computerised Accounting System (CAS). The results
show that ease of use, content, and accuracy has a significant effect on end-users satisfaction. Therefore, the empirical
results of this study can provide support for the Doll and Torkzadeh model (1988), which related to the factors
contributing end-users satisfaction toward accounting system.
Keywords: End-User Computing Satisfaction, Computerised Accounting System, Multiple Regressions
1. Introduction
1.1 Introduction
Since the beginning of 1980s, many researches had conducted survey in information system field highlighting the
tremendous development in end-user computing world. In fact, the growth of end-user computing is one of the significant
phenomena of the 1980s in the information management world (David, 1983). A study by the International Data
Corporation predicted that four out of five administrative and professional workers would be using personal computing to
support their work and personal activities by 1990.
In general practices, there are different between the practices in the public or government sectors and the business/private
companies. Hence, the applications of the Computerised Accounting System (CAS) in these organisations also differ
from one to another. For instance, the private organisation utilizes the general ledger system and the accounting principles
based on accrual basis. However, the public organisations employ vote accounting system, which is based on cash basis.
The financial management of the organisation is also related to budgeting using a code and warrant system (Statutory
Bodies Act 1980 (Account and Annual Report) (Act 240)). Thus, it seems to be interesting to investigate the level of
satisfaction among end-user computing in government sector, since it is different compared to the company or business
organisation.
Most of the end-user claimed that they are not satisfied with the Computerised Accounting System (CAS) which it was
using in the government sector office. The previous study showed that the main factor contributing to dissatisfaction
19
because of the limitation of stand-alone system among the various accountant general departments. Furthermore, it seems
clear that a study on End-User Computing Satisfaction (EUCS) in Malaysia is best described as limited especially in
government sectors. Therefore, the scarcity of empirical research in the area of end user satisfaction among Computerised
Accounting System (CAS); especially in the Malaysian context is significant to this research.
Perhaps, having a good financial or computerised accounting information system will increase the performance of an
organisation. Eventually, this enables the organisation to increase revenue in order to follow the recommendation of
self-financing by the ministry. Therefore, the objective of the study is merely to examine the most critical factor that
influences End-User Computing Satisfaction (EUCS) towards Computerised Accounting System (CAS).
The remainder of this paper is organized as follows. A review of related literature on end-users computing satisfaction
and hypotheses is discussed. Next, the methodology employed in this study, research instruments used and data analysis
method involved are described. Finally, the empirical results and discussion of the study are drawn.
1.2 Research Objective
1) To examine critical factors of content in End-User Computing Satisfaction (EUCS) that influence most end-users
satisfaction.
2) To examine critical factors of accuracy in End-User Computing Satisfaction (EUCS) that influence most end-users
satisfaction.
3) To examine critical factors of format in End-User Computing Satisfaction (EUCS) that influence most end-users
satisfaction.
4) To examine critical factors of ease of use in End-User Computing Satisfaction (EUCS) that influence most end-users
satisfaction.
5) To examine critical factors of timeliness in End-User Computing Satisfaction (EUCS) that influence most end-users
satisfaction.
6) To examine critical factors of system speed in End-User Computing Satisfaction (EUCS) that influence most
end-users satisfaction.
7) To examine critical factors of system reliability in End-User Computing Satisfaction (EUCS) that influence most
end-users satisfaction.
2. Literature review
2.1 Literature review
Unlike the other researchers, Pather et al. (2003) argued that the advent of e-Commerce has shifted the location of the
traditional user of Information Systems out of the physical domain of the organisation or business. E-commerce
businesses now have to deal with a new type of user viz. the e-Customer. Thus, they disputed that established
instruments that measure user satisfaction of IS in traditional (brick and mortar) businesses are not completely
appropriate. The authors, building on a comprehensive literature study, derived an appropriate model for exploring the
measurement of e-customer satisfaction in the South African context.
Markovic & Wood (2004) addressed the issue of user satisfaction with a computer lab in a university. Both formal and
informal data gathering techniques were used to provide comprehensive data for this research. Data was gathered from
both users and managers in order to provide a complete picture of the current situation. This data led to a research study
of user satisfaction among students and support staff. The research revealed that satisfaction with hours and software
and hardware performance had the greatest impact on user satisfaction followed closely by quality of support staff.
Bengts (2004) studies usability as a constituent of end- user computing satisfaction. Different measurement instruments
and rating scales for user satisfaction have been created; however, the relationship between satisfaction and usability
remains unclear. A web-based system with three different user interface alternatives was implemented and the system
was used by information technology students to practice SQL-queries in a university course. 43 students reported their
preference and the underlying reasons by answering both structured and open-ended questions in a web-based
questionnaire. The results also indicated that availability of desired features, simple interaction and user-control are as
constituents of satisfaction more important than simple screen design and error- free usage.
Huang et al. (2004) argued that while end-user computing satisfaction has been studied extensively, new aspects such as
purchasing convenience, product prices in the system and product delivery have to be included. In their study, they
developed an instrument for reliably and accurately measuring business-to-employee success. Testretest reliability and
construct validity were examined. Finally, they concluded that convenience, delivery, interface, accuracy, price and
security influence employee assessments of satisfaction. Managers can use the instrument developed in their study to
assess the success of their business-to-employee systems.
20
This study is continuing from Azleen et al. (2007) that measured the level of satisfaction among the end users of CAS in
Labuan F.T government sectors and also determine the relationship of seven factors (content, accuracy, format, ease of
use, timeliness, satisfaction with system speed, system reliability) that influence satisfactory level among the end users
toward the CAS. Furthermore, Azleen et al. (2007) also measured the level of EUCS among CAS end users in private
companies. They evaluated the significant relationship between EUCS factors and the overall EUCS in Malaysian context
and examined the differences of perception on overall EUCS among the demographic variables. This study found that
EUCS factors more reliable as compared to previous studies and the correlation between satisfaction and EUCS factors
are fairly strong. However, the study failed to detect significant differences of perception on overall EUCS on gender,
level of education and position.
The scope of the discussion is related to EUCS; the previous factors that contribute to EUCS, Doll and Torkzadeh Model
(1988); i.e., content, accuracy, format, ease of use, and timeliness and the modification made by Chin and Lee (2000), i.e.
satisfaction with system speed, and system reliability (self developed). The model will become the fundamental
guidelines to examine factors contributing to EUCS among Accountant General Department.
2.2 Research Hypotheses
The hypothesis of overall End-User Computing Satisfaction (EUCS) is constructed based on seven factors: (1) content (2)
accuracy (3) format (4) ease of use (5) timeliness (6) satisfaction with system speed and (7) system reliability. These
seven factors measure is formulated to test the relationship with the overall End-User Computing Satisfaction (EUCS)
that more focused questions, aimed specifically on CAS. Thus, the hypotheses for this study are summarized as follows:
H1: There is significant effect between content and end-users satisfaction towards Computerised Accounting System
(CAS)
H2: There is significant effect between accuracy and end-users satisfaction towards Computerised Accounting System
(CAS).
H3: There is significant effect between format and end-users satisfaction towards Computerised Accounting System
(CAS).
H4: There is significant effect between ease of use and end-users satisfaction towards Computerised Accounting
System (CAS).
H5: There is significant effect between timeliness and end-users satisfaction towards Computerised Accounting System
(CAS).
H6: There is significant effect between system speed and end-users satisfaction towards Computerised Accounting
System (CAS).
H7: There is significant effect between system reliability and end-users satisfaction towards Computerised Accounting
System (CAS).
3. Research Methodology
3.1 Sampling and Instrumentation
This study relies on survey design as it deemed more appropriate compared with other designs of research in order to
achieve the objective of the study. The population of this study covered the end users of Computerised Accounting
System (CAS) in each Responsibility Centres in East Malaysia. Responsibility Centres known as government agencies
that responsible to manage and control financial management in order to collect and report revenue and cost information
by areas of responsibility. However, only 62 Responsibility Centres were chosen due to time and cost limitation. We
have distributed two questionnaires for each department and the total population are 124 respondents. Sekaran, U. (2003)
has stated that the sample should be taken for this population are 97 respondents. Nevertheless, only 90 respondents
have completed and submitted these questionnaires.
Basically, the instrument of this study is based on the instruments, which was developed by Chin and Lee (2000). It
presents a new set instrument while focusing on the same five construct domains. They are: content, accuracy, format,
ease of use, and timeliness (Doll & Torkzadeh, 1988); and satisfaction with system speed (Chin & Lee, 2000).
According to their findings, the relationship between the overall measures of satisfaction than the baseline model is
expected to be relatively strong.
3.2 Data Analysis Method
To analyse the data, this study conducts descriptive analysis and multiple regression analysis using the SPSS 13.0
(statistical package for social science) software for windows. The study also tested reliability of the instrument so that it
enables to produce a robust and valid result.
21
4. Data Analysis and Results
4.1 Reliability Analysis
Table 1 summarizes the reliability analysis for each items and satisfaction. Each items show above 0.7. It seems that
this study provides more reliable instruments because most of the score is higher than Doll and Torkzadeh (1988). The
content score in Doll and Torkzadeh (1988)s study is 0.89 as compared to 0.918 in this study; accuracy=0.91 (0.875);
format=0.78 (0.927); ease of use=0.85 (0.927); timeliness = 0.82 (0.751). The coefficient alpha for system speed is
higher (0.800) than Chin and Lee study (0.72) and system reliability is a new factor provides 0.759 and presents a
reliable dimension for EUCS measurement. The reliability of the instruments of the study can be compared with Amdan
et al (2006). From the table, only system reliability and ease of use show better than Amdan et al (2006).
4.2 Descriptive Statistics
The results of the descriptive analysis are shown in Table 2. From the result, the highest mean is satisfaction with the
format (3.69) and the lowest mean is timeliness (3.34). It seems like most of the staff satisfied with format factor and
less satisfied with timeliness factor. The highest standard deviation is ease of use (0.69) and the lowest is system
reliability (0.56). The ease of use is deviate too far from the mean and system reliability is too close to the mean.
4.3 Multiple Regressions
Multiple regression analysis was performed in analysing the data. Properties of casual paths, including standardised
coefficients beta, t values and significant results, for each equation in the hypothesized model, are presented in Table 5.
Table 3 presents the result of model summary that system reliability, timeliness, format, accuracy, content, ease of use
and system speed predict the end-users satisfaction towards computerized accounting system (CAS) usage. From this
model, R represents the simple correlation .922 and R Square is .850, which shows that 85% changes in satisfaction can
be explained by system reliability, timeliness, format, accuracy, content, ease of use and system speed. However, 15%
changes can also be explained by other variables that influence the level of satisfaction. The result of the Adjusted R
Square tested shows that system reliability, timeliness, format, accuracy, content, ease of use and system speed
contribute significantly (F = 66.564; P < .001) as shown in Table 4 and predict 83.8% of variation in end-users
satisfaction.
4.4 Discussion
The results show that Hypothesis 1, Hypothesis 2 and Hypothesis 4 has significant effect with end-users satisfaction.
The significant relationship was found between ease of use and end-users satisfaction towards computerized
accounting system (CAS), (=.280, t=3.501, p=.001). This factor will be the most critical factor that contribute in
end-users satisfaction and it seems that users need to have accounting system that easy to be understood and handled
according to the structured module. This result has support to accept Hypothesis 1. Next, null hypothesis for Hypothesis
2 is rejected because second most critical factor is between content of accounting system and satisfaction based on the
findings (=.216, t=2.702, p=.0080) is revealed. This is because most of the end-users satisfied with the content of
accounting system. The last factor that contribute to the satisfaction is accuracy of accounting system (=.183, t=2.088,
p=.040) and Hypothesis 4 is substantiated. Most likely this circumstance is contributed by the high quality of
accounting information.
Furthermore, Hypothesis 3, Hypothesis 5, Hypothesis 6 and Hypothesis 7 are also constructed to examine the
significant effect with end-users satisfaction. Hypothesis 3 avers that format of the accounting system has significant
effect on end-users satisfaction. Table 3 illustrates that the effect of format of the system on end-users satisfaction was
not substantiated (=.087, t=1.120, p=.266). Next, Hypothesis 5 asserts that timeliness has significant effect on
end-users satisfaction. However, table 3 shows that the timeliness has insignificant effect with end-users satisfaction
and null hypothesis is accepted (=.105, t=1.740, p=.086). Further, null hypothesis for Hypothesis 6 is accepted
according to the result in table 3 (=.125, t=1.538, p=.128). The factor of system speed also do not effect significantly
to end-users satisfaction. It seems not important to influence the usability and emotional responses among users.
Finally, the hypothesis regarding reliability of accounting system has significant effect on end-users satisfaction
(Hypothesis 7) was not confirmed (=.087, t=1.120, p=.266). Undoubtedly, some of end-users are not aware with
accounting system security or privacy. This is because most of accounting system in government sector is supervised by
information technology unit or department. Thus, most of end users only intend to use and utilize this system regardless
the reliability of its security system.
5. Conclusion
This research attempted to measure the end-users satisfaction of CAS using multiple regression by determine the
significant effect of seven factors that influence end users satisfaction. Overall, this study indicates that most of end
users are almost satisfied with the CAS. It can be explained by descriptive statistics result which indicates high
satisfaction for each of EUCS factors. In addition, this study has enabled to identify the most critical factor that
22
influences end-user computing satisfaction towards CAS. The results show that ease of use, content, and accuracy has a
significant effect on end-users satisfaction. In other words, the usability of CAS is enormously important in producing
accurate output or content of CAS.
6. Implications and future research
The study demonstrated the EUCS factors in the government sectors with the CAS. The study also suggests that content,
accuracy, format, ease of use, timeliness, satisfaction with system speed and system reliability must be emphasized to
the standardized CAS. However, software developers must emphasize the user friendly and usability of the system
which might generate better output especially in terms of accuracy of the content.
The finding of this study is preliminary and shall not be generalized to the other government sectors. This is due to the
different in the adoption of the information technology and CAS especially in information system usage. Besides, the
number of the sample is relatively small. The small sample size is limited only 90 end-users of CAS. This small sample
size may be jeopardizing the population of the end user among the government sectors. The aspect of time also affected
the data collection method; i.e., by using questionnaire. A respondent who is not particularly interested in answering the
questionnaire is more likely interspersed to answer the question. This is because some of them are very busy with their
tasks and duties. As a result, they did not answer the questionnaire genuinely.
The future study can propose the other data collection method such as in depth interview with the end user computing.
In this study, questionnaire has been used to collect the data through owing to the fact that financial and time are of the
essence. The sample size should also be increased. A larger sample size would be required to ensure that the
generalisation ability of research. The future research should also consider the type of software, whether the department
purchases or develops the system. It is essential to measure the different perception on the overall EUCS among these
two types of the system since the end users are expected to have a different level of satisfaction.
References
Amdan, M., Rushdan, M.Y., Azleen, I., & Fahmi, M.G. (2006). The study of end-users computing satisfaction (EUCS)
on computerised accounting system (CAS) among peninsular Malaysia public universities: A survey in bursars office.
Proceedings of International Borneo Business Conference 2006. 2, 594-608.
Azleen., I., Mohd Rushdan,Y., Mohd Zulkeflee,A.R., Rahida., A.R. . (2007). The Study Of End-user Computing
Satisfction (EUCS) On Computerised Accounting System (CAS) Among Labuan F.T Government Sectors: A Case
Study In The Responsibility Centres. Labuan E-journal Of Muamalat And Society. vol.1: 1-14.
Azleen, I., Mohd Rushdan, Y.,. & Mohd Zulkeflee, R., & Rahida, A.R. (2007). The Study Of End-User Computing
Satisfaction (EUCS) On Computerised Accounting System (CAS) Among Labuan F.T Private Companies.
International Management Accounting Conference IV. 15-17 August 2007, Kuala Lumpur, pp. 1-17, ISBN:
978-983-3198-11-5.
Bengts, M. (2004). Usability as a constituent of end-user computing satisfaction, Department of Computer Science and
Information Systems, University of Jyvskyl. Unpublished Master Thesis.
Chin, W.W., & Lee, M. K. O. (2000). A proposed model and measurement instrument for the formation of is
satisfaction: The case of end-user computing satisfaction. Proceedings of The Twenty First International Conference On
Information Systems. 175-186.
David, H.B. (1983). A field study of end-user computing: Findings and issues. MIS Quarterly, 7(4), 34-47.
Doll, W. J., & Torkzadeh, G. (1988). The measurement of end-user computing satisfaction. MIS Quarterly. 12(6),
259-274.
Huang, J.H., Yang, C., Jin, B.H, & Chiu, H. (2004). Measuring satisfaction with business-to employee systems,
Computers in Human Behavior 20, pp.1735.
Markovic, B. & Wood, J. (2004). User satisfaction with CBA computer labs, CCSC: South Central Conference, pp
232-239.
Pather, S., Erwin, G., & Remenyi, D. (2003). Measuring e-commerce effectiveness: A conceptual model, Proceedings
of SAICSIT 2003, Pp. 143-152.
Sekaran, U. (2000). Research method for business A skill building approach, (3
rd
) Edition, John Wiley & Sons.
Statutory Bodies Act 1980 (Account and Annual Report) (Act 240)
23
Notes
Table 1. Reliability Analysis
Variable
Alpha
(Coefficient)
Alpha
(Coefficient)
Alpha
(Coefficient)
Alpha
(Coefficient)
Current study
Doll and Torkzadeh
(1988)
Chin and Lee
(2000)
Amdan et al.
(2006)
Satisfaction (DV) 0.771 - - 0.8550
Content 0.918 0.89 - 0.9243
Accuracy 0.875 0.91 - 0.8834
Format 0.927 0.78 - 0.9256
Ease of Use 0.927 0.85 - 0.8912
Timeliness 0.751 0.82 - 0.8711
System Speed 0.800 - 0.72 0.9126
System Reliability 0.759 - - 0.7204
Table 2. Descriptive Analysis
N Minimum Maximum Mean Std. Deviation
Content 90 2.33 5.00 3.76 .58
Accuracy 90 2.43 5.00 3.55 .59
Format 90 2.43 5.00 3.69 .68
Ease of Use 90 2.14 5.00 3.60 .69
Timeliness 90 2.50 5.00 3.34 .59
System Speed 90 2.33 4.83 3.48 .56
System Reliability 90 2.29 4.71 3.49 .57
Satisfaction (Dependent Variable) 90 2.14 5.00 3.61 .63
Valid N (listwise) 90
Table 3. Model Summary
Model R R Square Adjusted R Square
1 .922 .850 .838
a. Predictors: (Constant), System reliability, Timeliness, Format, Accuracy, Content, Ease of Use and System Speed
b. Dependent Variable: Satisfaction
Table 4. ANOVA
Model Sum of
Squares
df Mean
Square
F Sig.
1 Regression 30.274 7 4.325 66.564 .000
Residual 5.328 82 .065
Total 35.602 89
a. Predictors: (Constant), System reliability, Timeliness, Format, Accuracy, Content, Ease of Use and System Speed
b. Dependent Variable: Satisfaction
24
Table 5. Coefficients between EUCS factors and end-users satisfaction towards Computerised Accounting System
(CAS)
Standardized
Coefficients
t Sig.
Model Beta
1 (Constant) -1.991 .050
EASE .280** 3.501 .001
CONTENT .216** 2.702 .008
ACCURACY .183** 2.088 .040
TIME .105 1.740 .086
SPEED .125 1.538 .128
RELIABLE .095 1.392 .168
FORMAT .087 1.120 .266
a. Dependent Variable: Satisfaction
** Significant at the 0.05 level (2-tailed).
25
Layer Model and Algorithm of Organization Authorization

Based on Position Network
Feng Ding (Corresponding author)
Institute of Information and Decision Technology
Dalian University of Technology
Dalian 116023, China
E-mail: dingff@gmail.com
Yanzhang Wang
E-mail: yzwang@dlut.edu.cn
Xin Ye
E-mail: yejstar@163.com
This work was partially supported by the National Graduate Student Program of Building World-Class Universities
(Grant No.[2007]3020)
Abstract
After the RBAC model formalized in 1992 by David Ferraiolo and Rick Kuhn, the models of organization authorization
have been developing fast. Among these models, the one based on position network will be used widely and has a
prospective future. Algorithms for single-layer model and multi-layer model based on position network were cited in
this article. Meanwhile an algorithm example was given at the end of this article.
Keywords: Position, Network, Organization, Authorization, Layer, Model, Algorithm
1. Preface
In modern complex government systems, there are too many roles to be setup and provided to the users. If the duties or
privileges of users are changed, or the objects in system are changed, it will take lots of time to adjust the relationship of
User-Role and Role-Object. The integration of different government application systems has been affected by this
bottleneck. For improving the efficiency and security and reducing the workload of system maintenances, Position
was added between the User and Role. This makes it more close to the reality of Chinese government.
After turning this into reality, we found that a straight and effective algorithm was needed to evaluate the rationality of
the authorization process, make sure the strictness, and find out the problems. So the single-layer model algorithm and
multi-layer model algorithm for organization were put forward.
2. Algorithm for Organization Authorization System
We can easily connect the users, positions, roles and objects in different systems with network model. Show as figure 1.
How to look at the connection clearly and check out whether it is right? We need to setup a kind of model system to
evaluate the feasibility of the connection.
26
According to the relationship of the user ontology, the users in the e-government systems have the character of
social network. Meanwhile the position ontology also has the same character. For making the description and
expression of the character clearly, we describe the algorithm as follows.
2.1 Single-layer Model Algorithm for organization authorization system
Because of the complexity of the government, the mapping of user ontology and position ontology is not one-to-one. It
is multi-mapping. Show as figure 2.
Define a set of user ontology:
, recorded as vector ;
Define a set of position ontology:
, record as vector ;
We record the relationship of and as , and describe it with the form :
Shorten the as , then we get: .s
Especially, under this structure, if the only aim is to show the connection of and , we can use 0 and 1 to
realize it.
If has no right to access , , record as ;
If has the right to access , , record as .
If so, we can get a matrix like , which is formed only with 0 and1.
For keeping consistence with traditional model, the roles in the organization authorization system still have the
definitions and attributes as in model. And also the roles can be formed as layer structure. It is multi-mapping
between the positions and roles. Show as figure 3.
Define a set of role ontology:
We record the mapping between and as . We describe the mapping with the form of
:
} ,..., , {
2 1 m
user user user
T
m
user user user U ] ,..., , [
2 1
=
} ,..., , {
2 1 n
pos pos pos
T
n
pos pos pos P ] ,..., , [
2 1
=
n m
pos user
m
user
n
pos
mn
u
T
P U
(
(
(
(
=
(
(
(
(
(
(
(
(
=
mn m m
n
n
n m m m
n
n
n
m
T
u u u
u u u
u u u
pos user pos user pos user
pos pos pos
user
user
user
P U
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
] ,..., , [
......
2 1
2 22 21
1 12 11
2 1
2 2 2 1 2
1 2 1 1 1
2 1
2
1
T
P U
P
U | |
n m mn P
u U
=
m
user
n
pos
m
user
n
pos 0 =
n m
pos user 0 =
mn
u
m
user
n
pos 1 =
n m
pos user 1 =
mn
u
(
(
(
(
1 ...... 0 0
...... ...... ...... ......
0 ...... 1 1
1 ...... 0 1
RBAC
RBAC
} ,..., , {
2 1 i
rol rol rol
T
i
rol rol rol R ] ,..., , [
2 1
=
i n
rol pos
n
pos
i
rol
ni
p
T
R P
(
(
(
(
=
(
(
(
(
(
(
(
(
=
ni n n
i
i
i n n n
i
i
i
n
T
p p p
p p p
p p p
rol pos rol pos rol pos
rol rol rol
pos
pos
pos
R P
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
] ,..., , [
......
2 1
2 22 21
1 12 11
2 1
2 2 2 1 2
1 2 1 1 1
2 1
2
1
27
Shorten for , then: .
We still use the descriptions for roles and operations in model, and setup multi-mapping for them. By allowing
this, the workload for distributing the system function is reduced a lot. Show as figure 4.
We define a set of operation:
We record the mapping of and as , and record it as the form of :
Shorten for , then: .
By the matrix of the mapping of User-Position, Position-Role and Role-Operation, we can get the
matrix for single layer organization authorization. With , the relationship of different parts in the system can be
found out easily.
We record as the elements in , and is the mapping of users and operations. So we get Single-layer Model
Algorithm for Organization Authorization System:
If , it shows that the user has the right to get the operation ; if , it shows that the user
has no right to get the operation . From this matrix , all the operations that each user can get are shown clearly.
This can help the system administrator check the rationality of each user, and enhance the reliability of the whole
system.
From the above process, shows the relationship between users
and positions, recorded as: ;
T
R P
R
P | |
i n ni R
p P
=
RBAC
} ,..., , {
2 1 j
oper oper oper
T
j
oper oper oper O ] ,..., , [
2 1
=
j i
oper rol
i
rol
j
oper
ij
r
T
O R
(
(
(
(
(
=
(
(
(
(
(
(
(
(
(
=
ij i i
j
j
j i i i
j
j
j
i
T
r r r
r r r
r r r
oper rol oper rol oper rol
oper oper oper
rol
rol
rol
O R
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
] ,..., , [
......
2 1
2 22 21
1 12 11
2 1
2 2 2 1 2
1 2 1 1 1
2 1
2
1
T
O R
O
R
| |
j i
ij O
r R
=
j m
T T
(
(
(
(
(
(
(
(
(
(
(
(
(
=
=
=

ij i i
j
j
ni n n
i
i
mn m m
n
n
j i ij i n ni n m mn
O R P
r r r
r r r
r r r
p p p
p p p
p p p
u u u
u u u
u u u
r p u
R P U T
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
] [ ] [ ] [
2 1
2 22 21
1 12 11
2 1
2 22 21
1 12 11
2 1
2 22 21
1 12 11
mj
t T
mj
t
j m mj j i ij i n ni n m mn
t r p u T

= = ] [ ] [ ] [ ] [
0 =
mj
t m j 0 =
mj
t m
j T
(
(
(
(
(
(
(
(
ni n n
i
i
mn m m
n
n
p p p
p p p
p p p
u u u
u u u
u u u
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
2 1
2 22 21
1 12 11
2 1
2 22 21
1 12 11
i n ni n m mn R
p u U

= ] [ ] [
28
And shows the relationship between positions and roles, recorded
as: .
and are very helpful in the application of checking the middle layer mapping in organization authorization
system.
2.2 Multi-layer Model Algorithm for organization authorization system
In the practice of government, a structure like pyramid has been formed between different layers and sections. And this
caused that the same structure has been used in the former constructed systems. And also there are lots of pyramid
structures in positions.
Meanwhile for the easy configuration of organization authorization system, the pyramid structures were formed
between roles in many systems based on .
Because of the multi-layer structures, we need to expand the single-layer model for further detection to verify the
authorization process.
For the multi-layer, we add a column vector between each two layers. Each column vector consists of the mapping of
the elements of the two layers.
As to the multi-layer structure of positions, if the number of positions is , and they have been divided into layers,
we describe each position as , and show the abstract description as figure 5.
We describe the whole position set as , and describe the layer as
; if there are layers, we can get vectors.
We get the relationship of layer and layer with the multiplication of column vector and the transpose of
column vector , and record it as :
When has
some kind of relationship with , we record the value of as 1. If there is no
relationship, we record it as 0. So we can get a matrix which only consists of 0 and 1. Suppose there are
layers in position structure, we can get matrices in total. Record them as .
Multiply these matrices one by one, we can get the whole position structure matrix :
For the roles with abstract layer structure, figure 6 gives an example.
We use to describe the mapping matrix between the two nearby role layers, then we can get the whole role
(
(
(
(
(
(
(
(
(
ij i i
j
j
ni n n
i
i
r r r
r r r
r r r
p p p
p p p
p p p
......
...... ...... ...... ......
......
......
......
...... ...... ...... ......
......
......
2 1
2 22 21
1 12 11
2 1
2 22 21
1 12 11
j i ij i n ni O
r p P

= ] [ ] [
R
U
O
P
RBAC
n t
pos
t n
n n n
pos
pos
pos
pos
pos
pos
pos
pos
pos
(
(
(
(
......
......
......
......
......
...... ......
2
1
2
1
2
1
l
] , 1 [ ,
......
2
1
t l
pos
pos
pos
p
n
l
e
(
(
(
(
= t t
l 1 + l l
1 + l
) 1 (
+ l l
p
n n
n n n n
n
n
n
n
l l
pos pos pos pos pos pos
pos pos pos
pos
pos
pos
p
+
(
(
(
(
=
(
(
(
(
=
......
...... ...... ...... ......
......
......
] ,..., , [
......
2 1
2 2 2 1 2
1 2 1 1 1
2 1
2
1
1 ,
i
pos
k
pos ] , 1 [ , , n k i pos pos
k i
e
n n t
1 t n n
t t
p p p
) 1 ( 23 12
,......, ,
1 t n n P
) positions of layers : , , ( ,
1
1
) 1 (
total t N n l p P
t
l
l l
e =
[
=
+
) 1 (
+ k k
r
29
structure matrix :
The organization authorization system structure with multi-layer positions and roles shows in figure 7.
When getting the position structure matrix and role structure matrix, we can get the Multi-layer Model Algorithm for
Organization Authorization System:
3. Example for Single-layer Model Algorithm
For a given existing organization authorization system, we analyze its internal relationship between users, positions,
roles and operations.
Figure 8 shows the mapping between users and positions.
We get:
;
Figure 9 shows the mapping between positions and roles:
We get:
;
Figure 10 shows the mapping between roles and operations:
We get:
;
The whole system mapping is shown in figure 11.
From we get:
;
From we get:
R
) roles of layers : , , ( ,
1
1
) 1 (
total h N h k r R
h
k
k k
e =
[
=
+
j i ij
h
k
k k i n ni
t
l
l l n m mn O R P
r r p p u R R P P U T
=
+
=
+
= =
[ [
] [ ] [ ] [

1
1
) 1 (
1
1
) 1 (
) roles of layers : , positions of layers : , , , ( total h total t N h t l e
(
(
(
(
=
(
(
(
(
= =
1 0 0 0 0
1 1 1 0 0
0 1 1 1 0
0 0 1 1 1
] [
45 44 43 42 41
35 34 33 32 31
25 24 23 22 21
15 14 13 12 11
u u u u u
u u u u u
u u u u u
u u u u u
u U
n m mn P
(
(
(
(
(
(
=
(
(
(
(
(
(
= =
1 0 0
1 1 0
1 1 1
0 1 1
0 0 1
] [
53 52 51
43 42 41
33 32 31
23 22 21
13 12 11
p p p
p p p
p p p
p p p
p p p
p P
i n ni R
(
(
(
=
(
(
(
= =
1 1 1 0 0
0 0 1 1 0
0 0 0 1 1
] [
35 34 33 32 31
25 24 23 22 21
15 14 13 12 11
r r r r r
r r r r r
r r r r r
r R
j i ij O
i n ni n m mn R
p u U

= ] [ ] [
(
(
(
(
=
(
(
(
(
(
(
(
(
(
(
=
1 0 0
3 2 1
2 3 2
1 2 3
1 0 0
1 1 0
1 1 1
0 1 1
0 0 1
1 0 0 0 0
1 1 1 0 0
0 1 1 1 0
0 0 1 1 1
R
U
j i ij i n ni O
r p P

= ] [ ] [
30
;
From we get:
;
Now we can get the conclusion from the above calculation results:
From the mapping , the relationship of users and roles, we can see that there are 3 paths for user 1 to access role
1, 3 paths for user 2 to access role 2, 3 paths for user 3 to access role 3, and only 1 path for user 4 to access
role 4. Among these users, user 1, 2, 3 get full the rights to access all the roles, but user 4 does not.
From the mapping , we can see that position 1 can only access two operations, but position 3 can get the right
to access all the operations in the system.
From the mapping , there are 5 paths for user 1, user 2 and user 3 to get operation 2 and operation 3. Obviously the
authorization process was duplicated, and threatened to the safety of the system. If some user has more than 1 path to
get an operation, when someday the system administrator changes the configuration of the system and bans one path of
the user, the user can still access the operation through other paths. This can cause some uncertainty of the system.
References
Adams, S Farrell. (1999). Internet X. 509 Public Key Infrastructure: Certificate Management Protocols[EB/OL] .RFC
2510. March 1999.
Anders Kristensen. (1999). Formsheets and the XML Forms Language .Computer NetWorks, 1999, 31 :1189 .
Arpinar I B et al. (1999). Formalization of workflows and correctness issues in the presence of concurrency .Distributed
and Parallel Databases, 1999, 7(2):199~248.
Charlie Kaufman, Radia Perlman, Mike Speciner. (2002). Network Security: Private Communication in a Public World
[M] .America: Prentice Hall, 2002.
Christensen S R, Petrucci L. (2000). Modular analysis of Petri nets .The Computer Journal, 2000, 43(3):224~242 .
Christopher Steel, Ramesh Nagappan, Ray Lai. (2006). Core Security Patterns Best Practices and Strategies for J2EE,
Web Services and Identity Management[M] .2006.
Dave Crane, Eric Pascarello, Darren James. (2005). Ajax in Action [M]. America: Manning, 2005.
Eva Maler, Prateek Mishra. (2003). Assertions and Protocol for the OASIS Security Assertion Markup
Language(SAML)v1.1[EB/OL] .http://www.oasisopen.org/committee/security/. 2003-09.
Ferraiolo David F,Sandhu Ravi,Gavrila Serban,et al. (2001). Pro2 posed NIST standard for role2based access
control[J] .ACM Transac-tions on Information and Systems Security, 2001, 4 (3) :224-274 .
Holloway L E, Krogh B H, Giua A. (1999). A survey of Petri net methods for controlled discrete event
systems .Discrete Event Dynamic Systems: Theory and Applications, 1999, 7(2):151~190 .
Jesse James Garrett. (2005). Ajax: A New Approach to Web Applications
[EB/OL] .http://www.adaptivepath.com/publications/essays/archives/000385.php. 2005-2.
Liberty Alliance Project. Liberty Architecture View v1.1[EB/OL] .http://www.Projectliberty.org. 2003-01.
Liu K C, Ong T N. (1999). A modeling approach for handling business rules and exception .The Computer Journal,
1999, 42(3):206~223 .
Luo Z W, Sheth A, Kochut K, Miller J. (2000). Exception handling in workflow system .Applied Intelligence, 2000,
13(2):125~147 .
Mazzeo A, Mazzeocca N, Russo S, Vittorin V. (1997). A systematic approach to the Petri net based specification of
(
(
(
(
(
(
=
(
(
(
(
(
(
(
(
(
=
1 1 1 0 0
1 1 2 1 0
1 1 2 2 1
0 0 1 2 1
0 0 0 1 1
1 1 1 0 0
0 0 1 1 0
0 0 0 1 1
1 0 0
1 1 0
1 1 1
0 1 1
0 0 1
O
P
j i ij i n ni n m mn
r p u T

= ] [ ] [ ] [
(
(
(
(
=
(
(
(
(
(
(
(
(
(
(
(
(
(
=
1 1 1 0 0
3 3 5 3 1
2 2 5 5 2
1 1 3 5 3
1 1 1 0 0
0 0 1 1 0
0 0 0 1 1
1 0 0
1 1 0
1 1 1
0 1 1
0 0 1
1 0 0 0 0
1 1 1 0 0
0 1 1 1 0
0 0 1 1 1
T
R
U
O
P
T
Computer
concurrent
OASIS
http://docs
Perry D E
Engineerin
R. Fieldin
http://www
Ravi S San
1996, 29(2
Ryan Asle
SANDHU
standard .
2000, :47-
Volchkov
2001,:39-4
W M P v
Knowledg
and Informati
t systems .Rea
eXtensible
s.oasis-open.or
E, Romanovsky
ng, 2000, 26(9
ng, J. Gettys
w.w3.org/Proto
ndhu, Edward
2):38-47 .
son, (2005). N
U R S, FERRA
Proceedings o
63 .
A. (2001). R
45 .
van der Aalst
ge Engineering
ion Science
al-Time System
Access
rg/xacml/2.0/a
y A, Tripathi
) :817~819 .
s, J. Mogul,
ocols/rfc2616/r
d J Coyne, Hal
Nathaniel T Sch
AIOLO D, KUH
of the Fifth A
Revisiting Sing
t. (2001). A
g, 2001, 38(3):3
Fi

ms, 1997, 13(3)
Control
ccess-control-x
A. (2000). Cu
et al. (1999
rfc2616.html:
l L Feinstein, e
hutta. Foundat
HN R. (2000)
ACM Worksh
gle Sign-On A
reference mod
335~263 .
igure 1. User-P

): 219~236 .
Markup
xacml-2.0-cor
urrent trands in
9). Hypertext
October, 1999
et al. (1996). R
ions of Ajax [M
. The NIST m
hop on Role B
A Pragmatic A
del for team-
Position-Role-

Language(X
re-spec-OS.pdf
n exception ha
Transfer Pro
9.
Role-Based A
M] .America: A
model for role-b
Based Access
Approach in a
-enabled work
-Operation Mo

XACML)
f. 20052-01.
andling .IEEE
otocol-HTTP/1
ccess Control
Apress, 2005.
based access c
s Control [C].
New Contex
kflow manage
odel
Feb
version 2
.
E Transaction o
1.1 [EB/OL]
Models .IEEE
control: Towar
. Berlin, Germ
xt .IT Professi
ement systems
bruary, 2009
31
2.0[EB/OL].
on Software
.RFC2616.
E Computer,
rds a unified
many:ACM,
onal, IEEE,
s .Data and
Vol
32
l. 2, No. 1
F

Figure 2. U
Figure 4. Role-

User-Position M
Operation Ma

Mapping
apping
Figure 6

Figu
Figur
6. Roles Layer
Co
ure 3. Position-
re 5. Positions
rs
omputer and In
-Role Mapping
Layers
nformation Sci
g
ience
Computer and Informati
Figure 7. C

ion Science
Complex Orga
Figure 8. Use

anization Autho
er-Position Ma
Figure 10.

orization Syste
apping
Role-Operatio

em Model Bas
Figure 9. Po
on Mapping

sed on Position
osition-Role M
Feb
ns and Roles
Mapping
bruary, 2009
33
Vol
34
l. 2, No. 1
Fig

gure 11. Relati

ionship in Org

ganization Auth
Co
horization Proc
omputer and In
cess
nformation Sciience
35
Chi-Square Test for Anomaly Detection in XML Documents

Using Negative Association Rules
K. Premalatha (Corresponding Author)
Kongu Engineering College
Perundurai, Erode, TN, India
E-mail: kpl_barath@yahoo.co.in
A.M. Natarajan
Bannari Amman Institute of Technology
Erode, TN, India
Abstract
Anomaly detection is the double purpose of discovering interesting exceptions and identifying incorrect data in huge
amounts of data. Since anomalies are rare events, which violate the frequent relationships among data. Normally
anomaly detection builds models of normal behavior and automatically detects significant deviations from it. The
proposed system detects the anomalies in nested XML documents by independency between data. The negative
association rules and the chi-square test for independency are applied on the data and a model of abnormal behavior is
built as a signature profile. This signature profile can be used to identify the anomalies in the system. The proposed
system limits the unnecessary rules for detecting anomalies.
Keywords: Anomaly detection, Chi-square test, Negative association rule, XML
1. Introduction
XML is a simplified subset of the Standard Generalized Markup Language (SGML). It provides a file format for
representing data, a schema for describing data structure, and a mechanism for extending and annotating Hyper-Text
Markup Language (HTML) with semantic information. The XML data model carries both data and schema information,
being naturally suitable to represent semi-structured data. It is a standard for representing and exchanging information
on the Internet.
XML is a markup language for structured documentation. Structured documents are documents that contain both
content and some indication of what role that content plays. Almost all documents have some structure. A markup
language is a mechanism to identify structures in a document. The XML specification defines a standard way of adding
markup to documents. Information marked up as XML data is becoming increasingly persistent that allow data to be
imported, accessed and exported in the XML format. XML database may prove more efficient and easier to store the
data in XML format. As XML document storage formats become popular, the task of detecting anomalies within
XML document collections becomes more important.
Deviation from the normal or common order or form or rule or deviation from the normal standard, especially as a
result of congenital defects is called anomaly or outlier. Otherwise (Jiawei Han, Micheline Kamber. 2004.) very often,
there exist data objects that do not comply with the general behavior or model of the data. Such data objects, which are
grossly different from or inconsistent with the remaining set of data, are called outliers or anomaly. Due to anomalies,
data may be inconsistent. Since anomalies are rare event which violate the frequent relationships among data. Anomaly
Detection may refer to an unsupervised data mining technique that produces a data mining model for identifying cases
(records) that deviate from the norm in a dataset. The general step for anomaly detection schemes is
Build a profile of the abnormal behavior. Profile can be patterns or summary statistics for the overall population
(Dataset)
Use the abnormal profile to detect anomalies. Anomalies are observations whose characteristics agree significantly with
the abnormal profile
36
2. XML Documents
Extensible Markup Language (XML) is a simple, very flexible text format derived from SGML. Originally designed to
meet the challenges of large-scale electronic publishing, XML is also playing an increasingly important role in the
exchange of a wide variety of data on the Web and elsewhere. In the real world, computer systems and databases
contain data in incompatible formats. XML data is stored in plain text format. This provides a software- and
hardware-independent way of storing data. This makes it much easier to create data that different applications can share.
With XML, data can easily be exchanged between incompatible systems. One of the most time-consuming challenges
for developers is to exchange data between incompatible systems over the Internet. Exchanging data as XML greatly
reduces this complexity, since the data can be read by different incompatible applications.
XML documents must contain a root element. This element is the parent of all other elements. The elements in an XML
document form a document tree. The tree starts at the root and branches to the lowest level of the tree. All elements can
have sub elements (child elements):
<root>
<child>
<subchild>.....</subchild>
</child>
</root>
The terms parent, child, and sibling are used to describe the relationships between elements. Parent elements have
children. Children on the same level are called siblings (brothers or sisters). All elements can have text content and
attributes. Fig 1 shows the sample XML document.
The root element in the example is <bookstore>. All <book> elements in the document are contained within
<bookstore>. The <book> element has 4 children: <title>,< author>, <year>, <price>.
3. Association Rule Mining
The definition by Agrawal et al (R. Agrawal, T. Imielinski & A. Swami. 1993.) the problem of association rule mining
is defined as: Let be a set of n binary attributes called items. Let be a set of
transactions called the database. Each transaction in D has a unique transaction ID and contains a subset of the items in
I. A rule is defined as an implication of the form where and . The sets of items X and Y
are called antecedent (left-hand-side) and consequent (right-hand-side) of the rule.
To select interesting rules from the set of all possible rules, constraints on various measures of significance and interest
can be used. The best-known constraints are minimum thresholds on support and confidence. The support supp(X) of an
itemset X is defined as the proportion of transactions in the data set which contain the itemset. The confidence of a rule
is defined
.
Confidence can be interpreted as an estimate of the probability P(Y | X), the probability of finding the RHS of the rule in
transactions under the condition that these transactions also contain the LHS
The interestingness of an association rule can be defined in terms of the measure associated with it, as well as in the
form an association can be found. The most common framework in the association rules generation is the
support-confidence one. Although these two parameters allow the pruning of many associations that are discovered in
data, there are cases when many uninteresting rules may be produced. The measure interest is used to discover
interesting rules.
4. Chi-Square Test
Generally speaking, the chi-square test is a statistical test (Glenn A. Walker.) used to examine differences with
categorical variables. There are a number of features of the social world we characterize through categorical variables -
religion, political preference, etc. To examine hypotheses using such variables, use the chi-square test.
The chi-square test is used in two similar but distinct circumstances:
For estimating how closely an observed distribution matches an expected distribution - we'll refer to this as the
goodness-of-fit test
For estimating whether two random variables are independent.
) .Support(Y Support(X) - Y) Support(X Y) , Interest(X =
37
4.1 Goodness-of -fit test
The chi-square test is a goodness of fit test: it answers the question of how well do experimental data fit expectations.
The chi-square test of independence can be used for any variable; the group (independent) and the test variable
(dependent) can be nominal, ordinal, or grouped interval.
The following algorithm illustrates calculating a goodness-of-fit test with chi-square:
1) Establish hypotheses.
2) Calculate chi-square statistic. Doing so requires knowing:
The number of observations
Expected values
Observed values
3) Assess significance level. Doing so requires knowing the number of degrees of freedom.
4) Finally, decide whether to accept or reject the null hypothesis.
4.2 Testing Independence
The other primary use of the chi-square test is to examine whether two variables are independent or not. What does it
mean to be independent, in this sense? It means that the two factors are not related. Typically in social science research,
we're interested in finding factors that are related - education and income, occupation and prestige, age and voting
behavior. In this case, the chi- square can be used to assess whether two variables are independent or not. More
generally, we say that variable Y is "not correlated with" or "independent of" the variable X if more of one is not
associated with more of another. If two categorical variables are correlated their values tend to move together, either in
the same direction or in the opposite.
5. Anomaly detection in XML documents
The system is based on rules that define signatures and it detects anomalies that fall in abnormal signature profile. Fig 1
shows the anomaly detection in XML documents.
5.1 Two-dimensional (2-D) representation of XML documents
An XML document is defined (Jong P. Yoon, Vijay Raghavan, Venu Chakilam. 2001.) as a sequence of elementary
paths with associated element contents. An elementary path is a sequence of nested elements where the most nested
element is simple content element. In a two-dimensional representation of XML documents a row represents an XML
document and column represents an elementary paths. Fig 2 and 3 show the sample mushroom XML document and
2-D representation of XML document
In the above example the elementary paths are
ep1 = <Mushroom><Cap><Shape>
ep2 = <Mushroom><Cap><Surface>
ep3 = <Mushroom><Cap><Color>
ep3 = <Mushroom><Brusises>
ep4 = <Mushroom><Odor>
ep5 = <Mushroom><Gill><Attachment> etc.,
5.2 Mining Negative Association rules
There are rules that imply negative relationships such rules are called Negative Association Rules. A negative
association rule (Xindong wu, Shichao zhang. 2004.) also describes relationships between item sets and implies the
occurrence of some item sets characterized by the absence of others. To eliminate unwanted rules and focus on potential
interesting ones, the system predict possible interesting negative association rules by incorporating domain knowledge
of the data sets. The negative association rule can be written in the form XY, XY, , where X and Y are
itemsets.
Measures for XY
38
Measures for X Y
Interesting Negative Rules for XY
Interesting Negative Rules for X Y
ms minimum support threshold and mc minimum confidence threshold
From the above measures interesting negative association rules are identified and these antecedents and consequents are
applied for chi-square test to identify the independency.
5.3 Chi-Square test
Chi-square test is a statistical test to verify the independence between two variables. Using Chi-square test,
independency between the two variables are identified by finding contingency tables and expected frequencies
The following is a contingency table, a tabular representation of a rule.
R1 and R2 represent the Boolean states of an antecedent for the conclusions C1 and C2. The X11, X12, X21, X22
represent the frequencies of each antecedent-consequent pair. The R1T, R2T, CT1, CT2 are the marginal sums of the
rows and columns, respectively.
5.3.1 Calculating Chi-square:
1) Calculate and fix the sizes of the marginal sums,
2) Calculate the total frequency, T, using the marginal sums.
3) Calculate the expected frequencies for each cell
Formula:
Where and are the row total for i
th
row and the column total for j
th
column.
4) Select the test to be used to calculate based on the highest expected frequency, m.
5) Calculate using the chosen test.
6) Calculate the degrees of freedom. df=(r-1)(c-1)
A critical factor in using the chi-square test is the degrees of freedom, which is essentially the number of independent
random variables involved.
7) Use a chi-square table with and df to determine if the conclusions are independent from the antecedent at the
selected level of significance
8) For the selected level of significance
if

Reject the null hypothesis of independence
else
2
2
>
2
=
T
C R
eij
Tj ij
2
) (
ij
R Tj
C
2

39
Accept the null hypothesis of independence
5.4 Signature generation
From the chi-square test the strongly independent attributes are identified and are stored in the form of XML file as a
signature profile. For example
<cap_shape:b>(<cap_surface:r>
(support=20% CPIR=70% interest=50%>
For the above negative association rule the abnormal profile in XML format is shown in fig 4
Using XQUERY the test data is checked with abnormal signature profile for detecting anomalies.
Fig 5 shows the proposed system model.
6. Discussion and Experiment Results
Giulia et al (Giulia Bruno, Paolo Garza, Elisa Quintarelli, Rosalba Rossato. 2007.) identifies the anomalies in simple
nested XML documents using quasi-functional dependencies and association rule The system query the original
datasets to extract the instances that violate the dependences and for each quasi-functional dependency relating the sets
X and Y query all the stored association rules that involve X c and Y, with a low confidence (i.e., with a confidence
lower than a fixed threshold).
Proposed system identifies anomalies using negative association rules and enhanced with chi-square test. The anomalies
are identified whose confidence value is grater than minimum confidence threshold in negative association rule which is
improved with chi-square test.
The proposed system uses XML Mushroom data set includes 8124 hypothetical samples corresponding to 23 species of
gilled mushrooms. Table 2,3 & 4 show the number of rules generated by the proposed system. The number of rules
generated by the system is low because the uninteresting rules are filtered by interesting measure and chi-square
independence test. The anomalies are identified between two attribute levels.
References
Giulia Bruno, Paolo Garza, Elisa Quintarelli, Rosalba Rossato, (2007), Anomaly Detection in XML databases by
means of Association Rules, 18
th
international workshop on Database and Expert Systems Applications , pp.387-391,
IEEE .
Glenn A. Walker, Common Statistical Methods for Clinical Research with SAS Examples, SAS publishing, Second
Edition.
Jiawei Han, Micheline Kamber. (2004). Data mining Concepts and Techniques, Simon Fraser university, Reprinted.
Jochen Hipp, Ulrich Gntzer, & Gholamreza Nakhaeizadeh. Algorithms for association rule mining - A general survey
and comparison. SIGKDD Explorations, 2(2):1-58, 2000.
Jong P. Yoon, Vijay Raghavan, Venu Chakilam. (2001). BitCube: A Three-Dimensional Bitmap Indexing for XML
Documents, Journal of Intelligent Information Systems, pp. 241-254(14).
R. Agrawal, T. Imielinski & A. Swami. Mining Association Rules. Between Sets of Items in Large Databases",
SIGMOD Conference 1993: 207-216.
Xindong wu, Shichao zhang. (2004). Efficient Mining of Both Positive and Negative Association Rules, ACM
Transactions on Information Systems, Vol. 22, No. 3.
Table 1. Contingency Tables
C1 C2 Marginal Sums
R1 X11 X12 R1T=X11+X12
R2 X21 X22 R2T=X21+X22
Marginal Sums CT1=X11+X21 CT2=X12+X22 T=X11+X12+ X21+X22
40
Table 2. Number of rules generated at minimum support =0.2
Minimum Confidence Positive Association rule
(Confidence < mc)
Interesting Negative Association rule
(Confidence > mc)
0.6 333 66
0.7 409 50
0.8 460 43
0.9 513 32
Table 3. Number of rules generated at minimum support =0.3
(Confidence < mc)
Interesting Negative Association
rule (Confidence > mc)
0.6 92 47
0.7 137 34
0.8 162 29
0.9 191 24
Table 4. Number of rules generated at minimum support Level=0.4
(Confidence < mc)
Interesting Negative Association
rule (Confidence > mc)
0.6 48 0
0.7 70 25
0.8 82 24
0.9 100 21
41
Figure 1. Sample XML document
Figure 2. Sample mushroom document
<bookstore>
<book>
<category="COOKING">
<title lang="en">Everyday Italian</title>
<author>Giada De Laurentiis</author>
<year>2005</year>
<price>30.00</price>
</book>
<book>
<category="CHILDREN">
<title lang="en">Harry Potter</title>
<author>J K. Rowling</author>
<year>2005</year>
</book>
<book>
<category="WEB">
<title lang="en">Learning XML</title>
<author>Erik T. Ray</author>
<year>2003</year>
</book>
</bookstore>
42
ep1 ep2 ep3 ep4 ep5 ep6 ep23
D1 c s b b p f
Figure 3. 2-D representation of Mushroom document
Figure 4. Format of abnormal signature profile
Figure 5. Anomaly detection in XML documents
<Nrule support = 20% CPIR = 70% interest=50%>
<antecedent>

<cap_shape>b</cap_shape>
</antecedent>
<consequent>

<cap_surface>r</cap_surface>
</consequent>
</Nrule>
43
Research of Sales Contract Management System Based on WEB

Hualun Lai
Business School
University of Shanghai for Science and Technology
Shanghai 200093, China
Tel:86-21-5527-1343 E-mail: laihl@usst.edu.cn
Liangwei Zhong
CAD center
Tel:86-21-5527-6487 E-mail: zlv@usst.edu.cn
Jing Ni
Business School
Tel:86-21-5527-1343 E-mail: nijing501@126.com
Supported by Shanghai Leading Academic Discipline Project (S30504, J50503) and Shanghai Education Committee
( 05EZ30)
Abstract
On the basis of study on information stream of management of contract, a frame of an information system of
management of contract with holistic information is designed. Sale and production forecast is put forward in terms of
data mining technique and customer management concept and relevant system is developed. System information
includes customer management system, elaborate content of contract, finance management in the execution of contract,
record of change of contract, control of dispatch process, shares information with packaging system, and also shares
information with warehouse management system to realize the control of dispatch process. Data of customers and sale
of products information is obtained through data mining of customers and contract information. Good forecast for
market can be made by the data which also places perfect basis for the decision of production planning.
Keywords: Sell contract, Management system, WEB, Object-Oriented, Whole lifecycle
1. Introduction
The sales contract is the important production information of manufacturing firm, it includes the attribute of the user
request, specification and delivery time and the information which enterprise formulation and in the execution
productive plan process involves. The Manufacturing firm and user's link are connected by it. In the Manufacturing firm,
the contract schedule must be formulated between Master Production Schedule and Production Schedule. Contract
schedule must guarantee that all duties in the Master Production Schedule are completed and that production schedule
arranges well production as schedule in each contract. Contract schedule is the refinement of the Master Production
Schedule. To make the enterprise interior operation be Consistent, its function is not only Communication between
inside and outside, but also reducing the contradiction and the error between the Sale branch and Production
department.
Manufacturing industry enterprise's contract management includes management completed in ERP system and
independent management system. The two kinds of systems have some Weaknesses such as incomplete information
record, not thorough integrations with other systems and weak analysis ability. It is disadvantageous to enterprise's
informationization construction and influences seriously on enterprise market reaction rate and market competition
ability, which is more obvious to the large-scale manufacturing companies that have complex products than others.
On the foundation of contract management information flow is studied, contract management information system's
44
frame having complete information is designed, and sales and the production forecast method according to the data
mining technology and client management's theory is proposed. The corresponding system has been developed. The
system core goal is that a transformation to production schedule and job instruction is made, and that market demand is
mapped in the manufacturing promptly and accurately to enhances the enterprise reaction capacity to the market and
enterprise's competitive ability.
2. System function and Structure
With analysis of operating methods of contract management, the management processes and management content is
concluded based on whole lifecycle sales contract management. The structural feature of process is as shown in Figure
1. The functional modules are as follows:
(1) Contract input
Contract information, including customer information, the framework of the contract, a variety of equipment
specifications, model, quantity, and special note, and other information involved by contract is input by staffs of sale
department.
(2) Customer Management
Users can look for, modify the information of existing customers and add new customer information into system.
(3) System Maintenance
Including import and export data, backup and recovery, as well as the management of users (user addition, delete, and
the rights granted), operated by the system administrator.
(4) Search
The sales can inquiries contract information signed by the sale department, as well as execute-status of contract. The top
leaders can inquire information and execute-status of contract of all sale departments
(5) Contract Modify
Sales can modify the mistake contents of contract, such as payment, product name and specification caused by
carelessness. But other some contents can not be modified such as contract ID and contract type.
(6) Financial Check
Accountants check the payment details of contract content in the System on basis of original contract. The contract will
not continue the execute process without Accountants agree, and should be modified by the sales of departments. Only
if the contract contents match the rules of company, Accountant will verify the contract.
(7) Generation of production plans report
The Staff of central sale department make an production application based on the actual situation of stock. Technology
department make the BOM on the basis of production of application. Manufacture department staffs make the
production plan and order play according to the contents of ERP system, and the Manufacture report will be given to
factories.
(8) Financial check for delivering product
Accountants will check the delivering devices on the basis of arrived payment of customer.
(9) Generation of the bill of lading
The bill of lading will be generated according to the financial check and actual stock situation.
The system uses technology structure based on Web, and use the MS SQL Server as database. The development
language of system is JAVA, and ECLIPSE is as development environment.
Using the Web Service technology will make the System get further more development and third-party application
integration. Figure 2 showed as structure of the system. Web Service uses the standard output interface WSDL (Web
Service Description Language) serving for the client of internet/ intranet. It will not emphasis on using what technology
to get web solutions.
Web Service changes the programming-Oriented conception of WEB application into a service-oriented conception.
The most valuable technology is to integrated different model components and different model components. This
contract management system using Web Service technology provided technical support for the system integration in the
future of the inevitable internal and inter-enterprise.
3. Management of the Change of Contract Process
Customers often require that the contract be changed during the execution of the contract, and the change of contract
mainly covers the following range:
45
Elimination of contract
Increase and decrease of number
Change of equipment type
Change of equipment parameter
Change of transportation means and so on
The different extent of the change would affect the capital of contract as well as the execution of production plan.
Effective management of the data from the change provides data source to guarantee the implementation of data mining.
Meanwhile, the system attains the change of production cycle according to the change, and the change provides
production planning with basis.
Design of Database
A table recording change of contract is designed in database, and following is its structure:
No. of contract; date of change; content of change; reason of change,..
Content of change directly relate to contract in terms of code. After opening the database, relevant content would be
modified and be recorded. Reason of change is character field, which elaborately record the reason that customers put
forward and provide data processing with data source.
Other Change Management Results from Change of Contract
Change in many other aspects resulting from change of contract mainly affects production plan. System
communicates with PDM system, applies for change of design and exports changed data of BOM according to changed
data. Generally, there is no change of production planning process. Therefore, it is required that change of
production planning as well as changed data of purchasing planning be obtained by change module in production
system according to production capacity.
Management module in change of contract process will record data which is created in the process of change of contract
in database. Data becomes affluent with the increasing use of system.
4. The application of Data Mining Technology
(1) Data Mining System
Data Mining System is an independent three-tier structure. The bottom layer is data, and read data sources from the
SQL Server. The middle is mining, and has multi-dimensional database and excavation engine. Top-layer can switch
data with user interface.
(2) The realization of Data Mining
(3) Association Rules Mining
Data Mining (Data Mining), refers to mine hidden, useful and not yet found the information and knowledge from a
large number of the raw data, is considered an effective method to resolve the current "data rich and information poor".
Association Rules Mining is one of important areas of Data Mining (Liu, xin, etc. 2007).
According to different situations association rules can be classified as follows:
a) According to dealing with the types of variables of rules, the association rules can be divided into Boolean and Value
type.
b) According to data abstract level of rules, the association rules can be divided into single and multi-association rules.
c) According to relating to the dimensions of the data of rules, the association rules can be divided into one-dimensional
and multi-dimensional.
(4) Association Rules Mining Algorithm
From the large-scale database mining association rules require two steps:
a) Finding all the frequent set.
b) Producing strong association rules by the frequent set.
The study of Association Rules Mining focuses on the first step (Li, Shujuan, etc. 2004). The discovery ways of
association rules, the most influential algorithm is the Apriori algorithm, which uses a gradual approach called iterative
search. The basic algorithm is as follows: firstly, traversing one time to goal database, recording appearance times of
each item and attribute. that is, calculating the degree of support for each project to collect project constitutes frequent
table-L1which all the support degree is not less than the minimum support degree of user, and then link all the elements
right of L1, form candidate-C2, once again traversing the services database, calculated support degree for each set of
46
options of C2, to collect project constitutes frequent table-L2 which all the support degree is not less than the minimum
support degree of user, and then link all the elements right of L2, form candidate-C3, traversing the services database to
get L3, repeatedly implementing the course until the options stopping.
(5) Multi-association Rules Mining device
Apriori algorithm is more complicated, using an improved algorithm of the SDA in the paper, and using this improved
algorithm to develop a multi-level association rule mining device, two interrelated can be found in relational association:
property correlation and cross-attributes link. The former is a link among different properties; the latter is an inner
association of one property or attributes collection, at the same time that the association integrates with another set of
attributes or properties (Yang, guang & Wang, rui. 2007.).
Through the analysis of the data mining system gets the trend of product sales, as well as products associated with the
user, the factor weight to the contract changes, the integration of contract change and association of products and users
etc..
5. Integrated with packaging system
Delivering product is an important part of contract management, the completion of shipments of equipment marks
contract completed. Generally, customers will clearly stipulated batches delivering product in a different time in the
contract, because of the size of the goods being different, a device may be divided into a number of packing cases, or
many devices may be put a pacing case due to the smaller size of the goods.
Delivering product list must be marked information of packing cases, including the number of cases, identification
number etc., the information is not included in the contract information and need integrated with packaging system to
get product packing information of a contract.
In the database system level system integrates with packaging system, in the packaging systems database has packing
case and contract data table, this table will relate with equipment constitute table in contract management, and generate
detail data including delivering product list of packing case.
6. Integrated with financial system and inventory system
Shipments must meet the requirements of the stock and financial conditions, and to financial the user must pay the
money of contract demand before delivering product, to inventory, products must check to inventory before delivering
product. The information of inventory and financial needs integrate with inventory system and financial system.
System integrates with financial system and inventory system in the database level, the database in the financial system
has detail information of the process of payment of user, and the database in the inventory management system has
detail information of products, this table will relate with delivering product control management table in contract
management, and strictly control delivering product behavior of products.
7. Conclusion
1) Contract management system has closer ties with customer management systems, packaging system, financial system,
and inventory management system etc., and easy to integrate with these systems, owing to the system architecture
applying B / S system based on WEB.
2) Contract management system using data mining techniques provided the basis for forecasting, and data sources for
data mining based on complete record to contract changes.
3) Information of system integrated with financial system and inventory information system reaches purpose of
controlling delivering product, and put an end to human error.
References
Huang, meihua and Zheng, shaohong. (2005). Development and Research of management system based on B/S for
Contract and Cost Controlling. COMPUTER AND DIGITAL PROJECT. P64. 2005. Vol. 33. 03.
Li, Shujuan etc. (2004). Development of Contract Management System based on JSP. COMPUTER ENGINEERING.
2004.12.
Liu, xin etc. (2007). The Application Research of Data Mining Algorithms for Intrusion Detection. COMPUTER AND
INFORMATION TECHNOLOGY. P92~94. 2007.06.
Liu, Yabin etc. (2005). Master Eclipse. Electronics Industry Press. 2005.
Shi, Weiwei and Feng, Mei. (2007). Conqure JavaScript. Posts & Telecom Press.
Yang, guang and Wang, rui. (2007). Research of Mining Algorithm based on Association Rules. COMPUTER AND
INFORMATION TECHNOLOGY. P46~50.2007. 04.
47
Figure 1. function Structure
Figure 2. System Structure
48
Complexity Analysis in Heterogeneous System

Kuldeep Sharma & Deepak Garg
Department of Computer Science and Engineering
Thapar University
Patiala-147004, India
Abstract
Complexity analysis in heterogeneous system is one of the more complicated topics in the subject mathematics for
computing. It involves an unusual concept and some tricky algebra. This paper shows how various algorithms take part
in heterogeneous system and how they affect the overall system complexity? All analysis and methods are situation
specific. Here we show a comparative study of various method and techniques of algorithm analysis giving their
specific advantages and disadvantages. We will specify general guidelines for calculating the complexity of
heterogeneous system which still gives sleepless nights to the researchers.
Keywords: Heterogeneous system, Complexity analysis, Time complexity, Space complexity
1. Introduction
In recent years heterogeneous system has begun to play a vital role in computing. The potential use of heterogeneous
system in safety-critical applications increases the importance of their reliability. The system includes
telecommunications, consumer products, robotics, and an automotive control system is called heterogeneous system.
Heterogeneous system contains many different kinds of hardware and software working together in cooperative fashion to
solve problems. There may be many different representations of data in the system. This might include different
representations for integers, byte streams, floating point numbers, and character sets. Most of the data can be marshaled
from one system to another without losing significance. Attempts to provide a universal canonical form of information is
lagging. There may be many different instructions sets. An application compiled for one instruction set cannot be easily
run on a computer with another instruction set unless an instruction set interpreter is provided. Heterogeneous system may
be based on software as well as hardware, e.g. we have a system
in which our data is on site A and on site B, Site A having the SQL Server and Site B having DB2. In this scenario we may
have same hardware but platforms are different. In another case we may have SQL server or DB2 on both site but
hardware may be different.
An algorithm is a well-defined sequence of steps to solve a problem of interest. The complexity of an algorithm means
a function representing the number of steps (times) required to solve a problem under the worst-case behavior. The
worst-case behavior of an algorithm, means the maximum number of steps executed (or time taken) to solve a problem.
Hence, for any algorithm, even before implementing it, its complexity function should be analyzed. Such analysis will
help us predict the maximum magnitude of time required to solve a problem using the algorithm. There is one more
type of complexity function, namely volume complexity function, which represents the maximum primary memory
requirement while executing the algorithm (Design And Analysis of Algorithms).
Time complexity function of an algorithm gives the worst-case estimate in terms of the number of steps to be executed
for the algorithm. The order of the algorithm is defined like (n
2
), O (n!), O (2
n
), etc. The big O means the maximum
order of the algorithm. O (n
2
) means that order of the algorithm is n
2
, which indicates the maximum number of steps
required to solve any problem by that algorithm is n
2
where n is the problem size. The time complexity function may
be either polynomial or exponential.
Volume complexity: The Volume complexity function of an algorithm represents the amount of prime memory space
required while executing the algorithm. Again this may be either polynomial or exponential. In the case of the branch
and bound technique, if Breath First Search (BFS) is used, the function representing the memory space required to store
the sub-problems will be in exponential from; if Depth First Search (DFS) is used, the function representing the
memory space required to store the sub-problems will be in polynomial form. The type of the algorithm as well as data
structure affects the volume complexity. This analysis will be helpful in deciding the types of computer to be used for
implementing the algorithm.
49
2. Methods for Analysis
Apriori means designing then making. The principle of apriori Analysis was introduced by Donald Knuth. In apriori
Analysis first we analyze the system or problem then we design or write the code for the problem.
Posterior Analysis refers to the technique of coding a given solution and then measuring its efficiency. Posterior
Analysis provides the actual time taken by the program. This is useful in practice. The drawback of posterior analysis is
that it depends upon the programming language, the processor and quite a lot of other external parameters.
Micro analysis refers to perform the instruction count for all operations. It counts each and every operation of the
program. Micro analysis takes more time because it is complex and tedious (Average lines of codes are in the range of 3
to 5 million lines).Those operations which are not dependent upon the size or number of the input will take constant
time and will not participate in the growth of the time or space function, So they need not be part of our analysis.
Macro Analysis Perform the instruction count only for dominant operations or selective instructions which are costliest.
The following are a few examples of dominant (or basic) operations e.g. Comparisons and Swapping are basic
operations in sorting algorithms. Best, worst and average cases of a given algorithm express what the resource usage is
at least, at most and on average, respectively.
Amortized analysis refers to finding the average running time per operation over a worst-case sequence of operations.
In this kind of analysis we do not include probability we always use actual data so it guarantees the time per operation
over worst-case performance, hence it is different from average case.
E.g. In the implementation of the dynamic array, we double the size of the array each time it fills up. Because of this,
array reallocation may be required, and in the worst case an insertion may require O (n). However, a sequence of n
insertions can always be done in O(n) time, so the amortized time per operation is O(n) / n = O(1).( Allan Bording and
Ran El-Yaniv. 1998.) Aggregate analysis determines the upper bound T (n) on the total cost of a sequence of n
operations, and then calculates the average cost to be T (n) / n.
Accounting method determines the individual cost of each operation, combining its immediate execution time and its
influence on the running time of future
operations. Usually, many short-running operations accumulate a "debt" of unfavorable state in small increments, while
rare long-running operations decrease it drastically (Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and
Clifford Stein.).
Potential method is like the accounting method, but overcharges operations early to compensate for undercharges later.
3. Complexity of various algorithms in the system
The example given in figure 1 there are number of algorithm like authentication, encryption, fault tolerant, routing
algorithms, ACID (atomicity, consistency, isolation, durability) properties for transaction and, communication protocol
etc. In figure 1.terminals at a branch connected to a set of clustered web servers for authentication. The web servers are
in turn connected to a set of application servers for implementing banking rules and policies. The application servers
access mirrored and/or replicated data storage. Redundancy is present in the network also. Telephone banking using an
Interactive Voice Response System is used as a backup if the branch terminals break down. The design of the
authentication hardware and software requires fault tolerance the users should not have to Re-login if one or more
servers fail some state should be stored in the form of cookies in non-volatile storage somewhere. The banking
calculations require very high accuracy (30+ digit accuracy). Various kinds of fault tolerance schemes are used for
storage. For example, two mirrored disks always keep identical data. A write to one disk is not considered complete till
the other is written also. The servers have to respond within seconds to each user level request (deposit, withdrawal, etc)
the real time response of the system has to be evaluated using queuing theory and similar techniques. For TTS, the
response output speech samples have to be guaranteed to be delivered at periodic time intervals, say every 125ms
(micro-seconds). We have to calculate the complexity of overall system.
In a communication network routing is one of the latency sources. Routing algorithm has task to determine the suitable
port for delivering message address to one network node. Correct delivery of message from source node to destination
node is considered as optimal routing. The aim is to have the fastest and optimal routing available in the network.
The principle of routing with routing tables is straightforward. Ever node keeps a table with entries for each node. Using
these entries, it could be determined via which outgoing port a message had to be sent. Interval routing is a space
efficient routing strategy used in computer networks. The basis of interval routing idea is to label all of the nodes with
integer from one set (for example {1, 2, 3...n}) and labeling of all arcs with interval (in range of number of nodes).
These intervals means that a message addressed to a node labeled with u, via a port labeled with interval which includes
a node u is forwarded.
50
The mechanism used for detecting and correcting faults is called fault tolerance. It can be further classified based on
place of deployment, fault coverage, fault or error latency, cost of recovery .In real time systems fault tolerance can be
achieved by online fault (K. Chandy, J. Browne, C. Dissly and W. Uhrig. 1975.) detection followed by a
hardware-based checkpoint and rollback recovery mechanism. Cost is measured in terms of hardware and or time
overheads (F.P. Preparata, G. Metze, and R.T. Chien. 1967.). The complexity for k faults O (n
2
RL), where n is the
number of periodic tasks in a task set, R is the ratio of largest task period to smallest task period, and L is the number of
frequency levels supported by the processor (K. Shin and Y. Lee).
4. Measure complexity in Heterogeneous System
In table 3 we have five heterogeneous systems and each system has five different modules, in the table their module
wise complexity is given in the table. Now we will calculate the overall complexity of the system (T
C
).
Scenario1. In H_system 1 we have a very simple system and no module is running in parallel, all the modules are
running in sequence, simply by adding the complexity of each module will give the overall complexity of the system
(i.e. T
c
= n
2
+ n log n + 2
n
+ log n + n). T
C
=2
n
.It is an easy example and these types of systems are rare in use.
Now we have more dynamic system with inter connections and multiple dependencies with distributed, parallel or other
techniques of advance computing.
Scenario2. In H_system2 some of the modules are running parallel, so approach to measure the complexity will be
different. If module 3 and module 5 are running in parallel fashion, so the highest complexity from both the module will
be taken in to the account (the complexity of module 3 and module 5 is n
3
and n respectively so will take n
3
highest one),
so the total complexity (T
c
) of the system H_system 2 will be T
c
= log n +n! + n
3
+ n log n, T
C
=n
3
. These types of
systems are most common one. Some time all the modules are running in parallel manner then the highest complexity
will be the Total complexity (T
c
) of the system. In actual it may not be that simple and many more parameters may
affect overall complexity.
Scenario3. Heterogeneous system may include one or more critical module like security where it cannot be
compromised in context to complexity, so in this type of cases one module can affect the overall systems complexity.
Like in banking system or online payment system security, atomicity, accuracy, consistency, isolation and durability are
modules those cannot be compromised we have to take care these modules, they can increase the total complexity of the
system.
Scenario4. In this scenario we may have some dependencies. One module may dependent on another one so the
complexity will increase. Like in H_system 3 we cant start the second module until first module is not finished if the
complexity of module one is high then automatically the overall systems complexity will get effected and output of
Module 3 is dependent module 2 if output varies then it may has effect on T
C
.
Scenario5. Now we have system in which two modules (module 1and module 2) are running in parallel and have one
complex module 3 the output of module 5 is dependent on module 4. If we do some computations on client side rather
than to do all computations on server side, we do so the complexity of module 3 is decreased n
4
to n. the overall system
complexity will automatically decreased.
There are various other factors that will play significant roles in the overall complexity. Communication is one of them.
Depending upon the distance of locations where servers may be placed due to security or feasibility of the application.
Backup is another issue where system needs some time to reconsolidate with the back end links or system. Depending
upon the types of backup (hot or cold backup) different timings may be taken up by the system. In case of database
application there may be delays due to commit and other updates on various types of storage media, which needs to be
taken into account. These, will be looked into in our future study of this area.
Conclusion
This paper gives relative comparison of various techniques to analyze the complexity like Apriori Analysis, Posterior
Analysis ,Micro Analysis ,Macro Analysis, Amortized Analysis ,Big O Notation ,Theta Notation, Potential
Method ,Accounting Method and also lists there advantages and disadvantages, But still people face problem
determining the complexity of complex systems. Many points has come up relating to various specific problems
formulation and various design technique elements giving an insight into the area of complexity analysis of algorithms.
In this type of system, there are many algorithms that are running simultaneously as we discussed earlier like fault
tolerance, authentication, encryption, routing tables, communication protocols and for error recovery. Every algorithm
contributes in the system, so every algorithm have time and space complexity, if the algorithms are running parallel
then the complexity will be highest out of them. E.g. we have complexities n, n+1, n2 then the complexity will be n
2.
If
we have N algorithms and they are running sequentially then the complexity will be sum of all these.eg. We have N
algorithms running simultaneously then the total complexity will be Tc = 1+2++N.
51
References
Allan Bording and Ran El-Yaniv(1998). Online Computation and Competitive Analysis Cambrige University Press.20,
141. ISBN 0-521-56392-5.
Complexity Analysis of Routing Algorithms in computer network, M. Bielikov (Ed.), IIT.SRC 2005, April 27, 2005,
pp. 69-76.
Design And Analysis of Algorithms By R.Panneerselvam Prentice-Hall India. ISBN:978-81203-3278-2.
F.P. Preparata, G. Metze, and R.T. Chien, On the connection assignment problem of diagnosahlc systems. IEEE
Truns. Eleclron. Compirt.,Vol. C-16. pp. 848-854, Dec. 1967.
K. Chandy, J. Browne, C. Dissly and W. Uhrig, Analytic models for rollback and recovery strategies in data base
systems, IEEE Transactions on Software Engineering, Vol. 1, pp. 100-110, Mar. 1975.
K. Shin and Y. Lee, Error detection process-model, design and its impact on computer Performance, IEEE
Transactions on Computers.
Task feasibility analysis and dynamic voltage scaling in fault-tolerant real-time embedded systems Ying Zhang;
Chakrabarty, K.Design, Automation and Test in Europe Conference and Exhibition, 2004. Proceedings Volume 2, Issue,
16-20 Feb. 2004 Page(s): 1170 - 1175 Vol.2
Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and Clifford Stein. Introduction to Algorithms, Second
Edition. MIT Press and McGraw-Hill, 2001. ISBN 0-262-03293-7. Chapter 17: Amortized Analysis, pp.405430.
Vol. C (33), pp. 529-540, Jun 1984.
Table 1. Comparison of all Methods
Apriori Analysis Require less effort because we dont actually implement it.
Less risky,Simple and Easy
Posterior Analysis Whole effort may go waste if analysis gives negative data.
Actually resources or infrastructure is required.
Not applicable in many instances.
Micro Analysis Checks all instruction.
It takes more time,Not useful for large code.
Macro Analysis Check only dominant operation
It takes less time comparative to Micro Analysis
As useful as micro analysis
BigO Notation Represent worst case growth of an algorithm in time and space when they are
dependent on n.
Big O represents an upper bound.
Easy to calculate and widely acceptable.
Theta Notation It represents tight bound.
Useful for many problems.
Omega Notation It represents a lower bound.
Not used very often.
Amortized Analysis Probability is not involved.
Guarantees the time per operation over worst-case performance
Accounting Method It determines the individual cost of each operation.
Balancing can be done between heavy and light operations.
Potential Method
Use in Online Algorithms [1].
Good to calculate the overall cost of a data structure.
52
Table 2. Complexity of latency of message passing in wide area networks.
Routing Tables Interval routing
Space complexity
O(n log )
O(k log n)
Time complexity O(1) O(log(k ))
Note: - is maximum degree of the graph, n is number of nodes in the graph, k is interval.
Table 3. Heterogeneous Systems and module wise complexity.
Module
1
Module
2
Module
3
Module
4
Module 5
H_system
1
n
2
n log n 2
n
log n N
H_system
2
log n n! n
3
n log n n
H_system
3
n
3
N log n O(1) n log n
H_system
4
n
2
(log n ) /
2
n
4
N O(1)
H_system
5
n
2
n log n
n
2
log
n
(log n)
2
n log n
Figure 1. An Hetrogeneous system
53
Pixel Based Temporal Analysis Using Chromatic Property

for Removing Rain from Videos
Peng Liu
College of Computer Science and Technology, Harbin Institute of Technology
PO box 352, Harbin Institute of Technology, China
Tel: 86-451-8641-3631 E-mail: pengliu@hit.edu.cn
Jing Xu (Corresponding author)
Tel: 86-451-8641-3631 E-mail: xujing.hit@gmail.com
Jiafeng Liu
Tel: 86-451-8641-3631 E-mail: jefferyliu@hit.edu.cn
Xianglong Tang
Tel: 86-451-8641-3631 E-mail: txl60@public.hr.hl.cn
The research is financed by National Natural Science Foundation of China.No.60702032 (Sponsoring information), and
by Natural Scientific Research Innovation Foundation in Harbin Institute of Technology. No.HIT.NSRIF.2008.63
(Sponsoring information)
Abstract
The raindrops degrade the performance of outdoor vision system, and it brings difficulties for objects detection and
analysis in image sequence. In this paper, we propose an algorithm detect moving objects using chromatic based
properties in rain-affected videos captured by outdoor vision systems. Thus the raindrop removal algorithm includes
two parts that is removal raindrops in background and removal raindrops in moving objects. Since the degradation made
by raindrops is complex and appears as various changes. The raindrops detection function considering the chromatic
properties of image sequence is induced, which does not need the velocity and time information of raindrops. Therefore,
it is suitable for all the blur effects caused by raindrops. The removal raindrops are able to distinguish accurately the
raindrops-affected pixels from the immovable or movable objects. Although the objects are moving in the rain, the
algorithm is also effectual. The experiment results show that the proposal algorithm is able to remove the raindrops and
improve the quality of image sequence remarkable.
Keywords: Imaging model, Complex scenes, Video processing, Outdoor vision and weather
1. Introduction
The rain-affected image sequences annoy human viewers and degrade image quality. The degraded images also
decrease performance of computer vision algorithm in areas such as object detection, tracking, segmentation, video
surveillance.
54
Now many algorithms are proposed to remove obvious rain streaks. But there are still two difficult points needing
further research.1. To remove light rain. Rain streaks are too small to be detected in single frame when the background
is complex, but viewer can feel rain streaks in videos. 2. To get accurate intensity value. Due to the imaging precision
and video compression algorithms, its difficult to get the true intensity value of a rain-affected pixel. The existing
algorithms just have good effectiveness for detecting and removing obvious and clear rain streaks. So to resolve the two
difficult points described above, in this paper, we propose a general pixel based algorithm using chromatic property for
removing rain from an ordinary video.
1.1 Previous work
The methods of removing rain streaks can be classified into two types: pixel-based method and frequency-based method.
The first pixels-based method used a temporal median filter in (Starik, 2003). This works during moderate rain in which
the scenario is not seriously corrupted. But moving objects will cause blur effects.
For removing rain streaks during video acquisition, Garg and Nayar (2007, pp.3-27) proposed modifying the camera
parameters by increasing exposure time or reducing the depth of field. But for some types of cameras their parameters
are not adjustable. For removing rain streaks in video, Garg and Nayar (2004) supposed that there are few raindrops will
cover three consecutive frames. So if a pixel is covered by one raindrop, the intensity change of this pixel between this
frame and its previous frames is equal to the intensity change between this frame and its latter frame. Then they used a
linearity constraint to reject improperly detected pixels. Finally they calculated the binary rain field to segment rain area.
This method has two limits: 1. In heavy rain, raindrops could frequently affect the same position in three consecutive
frames. 2. Due to noise or video compress algorithm, the linearity constraint is not always valid.
Zhang at al (2006, pp.461-464) proposed another pixel-based algorithm using the chromatic constraint. They found If a
pixel is covered by raindrops, the varieties of the intensities of , , and , are approximately the same. This
property is used to segment moving objects and rain area. But in some videos, the intensity changes of , , and
in area where comprise object motion are far less than in rain-affected area. It is difficult to find an appropriate
threshold that is suitable for both stationary and dynamic objects.
Barnum at al (2007) used frequency information to detect and remove rain streaks. They used a blurred Gaussian to
approximate the blurred effect caused by a moving raindrop. It works when the streaks are clear. But for light rain or
heavy rain, the model of a blurred Gaussian is not always effective.
1.2 Our work
This paper focuses on rain removal in video. If a video is captured by a moving camera, the frames can be stabilized in
advance. So to make the problem simpler, we mainly focus on scenario comprising a stationary background and some
moving objects captured by a stationary camera.
First, we make a further study of the raindrops model, and give a general detecting method using chromatic property
which is able to avoid noise effect. Then we estimate the distribution of the detected variable to reject improperly
detected pixels. In the removal step, we use the imaging model to restore the background information. In this paper, we
dont use the information of raindrops velocities or shapes. Therefore our method is suitable for various types of
raindrops. The result shows that our method is effective and has a better performance.
2. Chromatic properties of rain streaks
Garg and Nayar (2007, pp.3-27) got a conclusion that when a raindrop is passing through a pixel, the intensity of its
image is brighter than the background. This imaging process is illustrated by the formula:
(1)
Where is the intensity of this pixel affected by the raindrop. is the time during which the raindrop projects onto
the pixel. is the irradiance caused by the raindrop during the time , is the average irradiance of the
background. is the exposure time of the camera.
If the background is stationary, or the motion of it is slow, we can use the average irradiance value to calculate the
background intensity of the pixel over the time duration .
(2)
Also we use the time-averaged irradiance to compute the intensity of raindrops.
R G B
R G B
0
( , ) ( , ) ( , )
T
r r b
I x y E x y dt E x y dt
t
t
= +
) )
r
I t
r
E t
b
E
T
b
E
b
I T
( , ) ( , )
b b
I x y T E x y =
0
1
( , ) ( , )
r r
E x y E x y dt
t
t
=
)
55
(3)
Use to denote the change in intensity at a pixel due to a raindrop, we obtain
(4)
Let , , rewrite Eq.(3) and Eq.(4)
(5)
(6)
Here means the equivalent ideal intensity caused by raindrops during the exposure time .
means the ratio of rain-affected time to the exposure time.
From Eq.(6) we know that the intensity change of pixels along a rain streak is proportional to its background values.
But usually the intensity values of pixels along a rain streak are similar.
(7)
proximate to the intensity of a little white sphere. Therefore the equivalent ideal intensity and the ratio
are uniform respectively in , , channels. That is, although a falling raindrop appears as complicate shape and
intensity change along the falling direction, the values of and at each pixel are equal respectively in
three channels. But the Eq.(6) shows the relation of rain-affected pixel and its background. When rain is heavy,
raindrops usually cover the same position in consecutive frames. Its hard to get the accurate value of background. In
the next part, we will derive the formula of the relation between rain-affected pixels in consecutive frames.
2.1 Relation of Background pixel and rain pixel
Consider pixels at the same position in two consecutive frames, one is background pixel, and the other is rain-affected
pixel. The brighter one is rain-affected pixel. denotes the vector of intensity change between frames in ,
, channels. is the vector of background intensity in three channels. We rewrite the Eq.(6)
(8)
Where .
From the discussion above, we get . We still use to denote this value.
2.2 Relation of Two Rain Affected Pixels
When rain is heavy, or the frames extracted from videos are not consecutive frames. Sometimes the pixels at position
in both frames are covered by raindrops. We use denotes the
intensity vector of the brighter one, use denotes the other one.
is its background intensity vector.
and is the parameters according to the brighter pixel.
and is the parameters of the other one. We get
(9)
(10)
From Eq.(10), we get , substitute it into Eq.(9), we get
(11)
( , ) ( , ) ( , )
r r b
T
I x y E x y I x y
T
t
t
= +
( , ) ( , ) ( , )
r b
I x y I x y I x y A =
( , ) ( , ) ( , ) ( , ) ( , )
r b r b
I x y I x y I x y E x y I x y
T
t
t A = =
( , ) ( , )
E r
I x y TE x y =
T
t
o =
( , ) ( , ) (1 ) ( , )
r E b
I x y I x y I x y o o = +
( , ) ( , ) ( , )
E b
I x y I x y I x y o o A =
( , )
E
I x y T / T o t =
1 2
2 1
( , ) ( , )
( , ) ( , )
b b
I x y I x y
I x y I x y
o
A A
=
( , )
E
I x y o
R G B
( , )
E
I x y o
( , ) I x y A
&
R
G B ( , )
b
I x y
&
( , ) ( , ) ( , )
E b
I x y I x y I x y o o A =
& & &
( , ) [ ( , ), ( , ), ( , )]
T
E ER EG EB
I x y I x y I x y I x y =
&
( , ) ( , ) ( , )
ER EG EB
I x y I x y I x y = = ( , )
E
I x y
( , ) x y ( , ) [ ( , ), ( , ), ( , )]
T
r rR rG rB
&
( , ) [ ( , ), ( , ), ( , )]
T
r rR rG rB
I x y I x y I x y I x y ' ' ' ' =
&
( , ) [ ( , ), ( , ), ( , )]
T
b bR bG bB
&
( , ) [ ( , ), ( , ), ( , )]
T
E ER EG EB
&
o
( , ) [ ( , ), ( , ), ( , )]
T
E ER EG EB
I x y I x y I x y I x y ' ' ' ' =
&
o'
( , ) ( , ) (1 ) ( , )
r E b
I x y I x y I x y o o =
& & &
( , ) ( , ) (1 ) ( , )
r E b
I x y I x y I x y o o ' ' ' ' =
& & &
( , ) ( , )
( , )
1
r E
b
I x y I x y
I x y
o
o
' ' '
=
'
& &
&
1
( , ) ( , ) ( , ) ( , ) ( , )
1 1
r r E E r
I x y I x y I x y I x y I x y
o o o
o o
o o
'
' ' ' ' =
' '
& & & & &
56
Comparing Eq.(11) with Eq.(8), it shows that Eq.(8) is the special form of Eq.(11) when . In this condition the
pixel with less intensity value is background pixel. And the rain-affected time is zero, so .
Since the terms of , and in Eq.(11) are equal respectively in three channels, let
, , we can rewrite Eq.(11)
(12)
Eq.(12) is the pixel-based imaging formula using chromatic properties.
3. Existence of solution and noise effect
Eq.(12) is the imaging model of rain-affected pixels which is used as a detecting function. But when is
affected by noise, it is difficult to get an accurate value so that is inaccurate. Let denote the
noise in three channels, from Eq.(12), we get
(13)
Using two channel to calculate the value of , for example using , channels:
(14)
Calculation of is of two types.
(1) does not approach to zero
When does not approach to zero, suppose , and . The term of noises is negligible.
Using Eq.(12), we will get three values respectively in , channels. So according to the imaging model of
pixels covered by raindrops, if the three values are similar, this pixel is a rain-affected pixel. Otherwise it belongs
to a moving object.
(2) approaches to zero
When approaches to zero, the term of noises is not negligible. Although a pixel is rain-affected pixel, the
three values calculating by three channels may be not equal to each other. In this condition, ,
, where , due to the fast movement of raindrops, , therefore , we
get . From Eq.(13), when a pixel is affected by a raindrop,
. Since the noise is caused by illumination or imaging process, its far less than
the intensity change caused by a mobbing object. We get . So if we use and to denote different
channels, when approaches to zero, the pixel at is affected by raindrops, otherwise
its of a moving object.
4. Segmentation of moving object
From the discussion above we know if a pixel is covered by a raindrop, either , or
does not approach to zero. So the equation
has a solution. We use a general formula to calculate the
variable .
(15)
Where , is a threshold to judge if approaches to
zero. For three channels we will get three values by using Eq.(15). That is
r b
I I ' =
& &
0 o' =
1
( , ) ( , )
1
E E
I x y I x y
o
o o
o
' '
'
& &
1
o o
o
'
'
1
( , ) ( , )
1
E E
A I x y I x y
o
o o
o
' ' =
'
& & &
1
K
o o
o
'
=
'
( , ) ( , )
r
I x y A K I x y ' A =
& & &
( , )
r
I x y '
&
K
, , ,
[ ]
T
R G B
=
&
( , ) ( , )
b
I x y A K I x y ' A = +
& & & &
K R G
( , ) ( , )
R G R G
bR bG bR bG
I x y I x y
K
I I I I
A A
= +
' ' ' '
K
bR bG
I I ' '
bR bG
I I ' '
bR bG
I I C ' ' > 0 C >
K R G
K
bR bG
I I ' '
bR bG
I I ' '
K ( ) 0
bR bG
I I ' '
( ) /(1 ) K o o o ' ' = / T o t = 0 T t <
1 K <
( ) 0
bR bG
K I I ' '
( ( , ) ( , )) ( )
R G R G
I x y I x y A A
0
R G
k ' k
'
( , ) ( , )
k k
I x y I x y A A ( , ) x y
( , ) ( , ) 0
R G
I x y I x y A A
( , ) ( , )
bR bG
I x y I x y ' '
( ( , ) ( , )) /( ( , ) ( , ))
R G bR bG
K I x y I x y I x y I x y ' ' = A A
k
( , ) ( , )
( ( , ) ( , )) (1 )
( , ) ( , )
R G
R G
bR bG
I x y I x y
K J I x y I x y J
I x y I x y
A A
= A A
' '
1, when ( , ) ( , )
0, when ( , ) ( , )
bR bG
bR bG
J I x y I x y C
J I x y I x y C
' ' = s
' ' = >
C ( , ) ( , )
bR bG
I x y I x y ' '
k
57
(16)
The final value of is .To determine whether a pixel is covered by a raindrop by values of is
very tedious. Here we use another method to segment moving object from rain field.
From Eq.(11) and Eq.(12), we know . Consider , because ,
and we get , that is . So
(17)
Where , and . So when the pixel at is covered by a raindrop, we have .
However when the pixel belongs to a moving object, the term of does not converge to zero. The
detecting function is
(18)
If dose not converge to zero, the pixel is of a moving object. We obtain the edges of moving object by using
this detecting function.
5. Detection and removal of rain
As mentioned in section 1.2, we suppose those frames abstracted from video are aligned in advance. Using the detecting
function, a frame is divided into two parts: static background and foreground. So the removal algorithm is twofold:
(a) To remove raindrops in background.
(b) To remove raindrops in foreground.
Many methods are suitable for removing raindrops in static background. We use Kalman Filter to suppress intensity
increase caused by raindrops. To remove raindrops in foreground, we align the foregrounds respectively, and then use
the method of removing raindrops in background like the first step.
6. Results and Analyses
Our experiments use a threshold of 3 gray levels to detect the intensity change of pixels. Use a threshold of 5 gray to
judge if the subtraction between channels such as approaches to zero. In calculating of
detecting function, the order of is 1, the threshold of the detecting function is 5.
Fig.2 (a) is an image of static scene from the video captured by Zhang and Li (2006, pp.461-464). (b) is removal result
using Garg and Nayars method (2007, pp.3-27). The result shows that the method of Garg and Nayar is not effective
when rain is heavy because their removal algorithm only calculates two consecutive frames. Image using method of
Zhang and Li (2006, pp.461-464) has better quality. But their method uses K-means clustering to calculate the
background color which is effective in static scene. But this method will damage the image quality in some areas of
dynamic scene by improperly regarded some moving object pixels as rain-affected pixels.
Fig. 3 shows results in dynamic scenes. Fig. 3 (b) and (c) are result of Zhangs method. The improperly detected pixels
obviously damage the visual quality of derained image. (d) is our result. (e) and (f) are local areas of (d). Compared
with (c), our results accurately detected rain-affected area. (g) is another frame in this video. (j) is derained frame. (h)
and (i) are local areas of (g). (k) and (l) are local areas of (j). Our results show a better performance in this conditon.
7. CONCLUSIONS
By further studying the model of raindrops, we obtain a detecting function using chromatic property which is suitable
for a general rain condition. It can segment rain-affected pixels from moving objects between two frames. Then
removing approach is classified into two steps: to remove raindrops in background and to remove raindrops covered
moving objects. The removal method makes the pixels at the same position of backgrounds in each frame similar and
shows a better visual quality. Another advantage is that our method does not use any information about the shape, the
velocity of raindrops, neither uses the value of cameras exposure time. Therefore it is effective in various conditions,
such as heavy rain, light rain, rain in focus, rain out of focus and so on. When the objects are moving in the rain, this is
1 1 1
2 2 2
3 3 3
( , ) ( , )
( ( , ) ( , )) (1 )
( , ) ( , )
( , ) ( , )
( ( , ) ( , )) (1 )
( , ) ( , )
( , ) ( , )
( ( , ) ( , )) (1 )
( , ) ( ,
R G
R G
bR bG
R B
R B
bR bB
B G
B G
bB bG
I x y I x y
K J I x y I x y J
I x y I x y
I x y I x y
K J I x y I x y J
I x y I x y
I x y I x y
K J I x y I x y J
I x y I x y
A A
= A A
' '
A A
= A A
' '
A A
= A A
' ' )
k
1 2 3
max( , , ) k k k k
( ) /(1 ) K o o o ' ' = K o ( 1) /(1 ) K o o o o ' ' =
0 1 o < 0 K o < K o s
1
( , ) ( , ) ( , )
n n n
E
K I x y I x y I x y o o
+
A s A <
/ T o t = T t ( , ) x y
1
lim ( , ) 0
n
E
n
I x y o
+

( , )
n
K I x y A
( , ) ( , )
n
F x y K I x y = A
( , ) F x y
( , ) ( , )
bR bG
I x y I x y ' '
K
58
also effective. Moreover the method of Garg (2007, pp.3-27) and Zhang (2006, pp.461-464) need many consecutive
frames to calculate the information of rain-affected area. However our algorithm is able to segment rain field from
moving object in two arbitrary frames. The results show that our method has better visual quality.
References
Barnum, P, Kanade, T., and Narasimhan S., (2007). Spatio-Temporal Frequency Analysis for Removing Rain and
Snow from Videos, PACV workshop at ICCV.
Braillon, C., Pradalier, C., Crowley, J.L., Laugier, C., (2006). Real-time moving obstacle detection using optical flow
models.Proc. 2006 IEEE Intelligent Vehicles Symposium, IV 2006, pp. 466-471.
Cucchiara, R., Grana C., Piccardi M., and Prati A., (2003). Detecting moving objects, ghosts, and shadows in video
streams, IEEE Trans on Pattern Anal. and Machine Intel., vol. 25, no. 10, pp. 1337-1442,2003.
C. Wren, A. Azarhayejani, Darrell, T., and Pentland, A.P., (1997). Pfinder: real-time tracking of the human body,
IEEE Trans. on Pattern Anal. and Machine Intell., vol. 19, no. 7, pp. 780-785, 1997.
C. Ulbrich, W.. (1983). Natural variations in the analytical form of the raindrop size distribution, J. of Applied
Meteorology, 22(10):176475, 1983.
Garg, K., and Nayar, S. K., (2004). Detection and removal of rain from videos,CVPR, 2004.
Garg, K., and Nayar, S. K., (2006). Photorealistic rendering of rain streaks, SIGGRAPH, 2006.
Garg, K., and Nayar, S. K., (2007). Vision and Rain,IJCV, pp.3-27,2007.
Han, B., Comaniciu, D., and Davis, L.S.,(2004). Sequential kernel density approximation through mode propagation:
applications to background modeling, Proc. Asian Conf. on Computer Vision, Jan. 2004.
Liu, Polley R., Meng, Max Q.-H., Liu, Peter X., Tong, Fanny F. L., Wang, Xiaona, (2006). Optical flow and active
contour for moving object segmentation and detection in monocular robot, Proc. IEEE International Conference on
Robotics and Automation, ICRA 2006, pp. 4075-4080.
Narasimhan, S.G. and Nayar, S.K. (2002). Vision and the Atmosphere, IJCV, 2002, 48(3):233254.
Starik, S., and Werman, M., (2003). Simulation of rain in videos. In Intl. Workshop on Texture Analysis and Synthesis,
2003.
Seki, M., Wada, T., Fujiwara, H., and Sumi, K., (2003) Background subtraction based on cooccurrence of image
variations, Proc. CVPR 2003, Vol. 2, pp. 65-72,2003.
Wayne, P. Power, and Schoonees, J. A., Understanding background mixture models for foreground segmentation,
Proc. of JVCNZ 2002, pp. 267-271, Nov. 2002.
Zhang, X., Li, H., Qi, Y., Kheng, W., and Ng., T. K., (2006). Rain removal in video by combining temporal and
chromatic properties. ICME, pp.461-464, 2006.
Computer
(a) is the o
using Zha
algorithm
frames on
and Informati
Figure 1.
Fig
original frame.
angs method.
uses 10 frame
the position in
ion Science
Rain affected
(a) is a rain-a
(b) sh
gure 2. One fra
(b) is the resu
The removal
es. (e) is the re
n (a), the remo

frame and the
affected frame
hows the inten
ame in static sc
ult using Garg
algorithm us
esult of our me
val algorithm

intensities cha
e with stationar
nsities change
cene from the v
s method. The
ses 200 frame
ethod using 50
uses 10 frame

ange at one fix
ry background
of the position
video captured
e removal algo
es. (d) is the r
0 frames. (f) is
s

xed position in
d in heavy rain.
n in (a)
d by Zhang and
orithm uses thr
result using o
s the intensitie
Feb
n this video.
.
d Li.
ree frames. (c)
our method. T
es change of th
bruary, 2009
59
is the result
The removal
he de-rained
Vol
60
The
resu
(d).
are
l. 2, No. 1
e video is a chi
ult of Zhangs
(g) is another
local areas of

ip from the mo
method. (c) is
r frame of this
(j)

Figure 3. So
ovie Magnoli
s a local area o
video. (h) and

ome experimen
ia. It has been
of (b). (d) is o
d (i) are local a

nt results in dyn
n used by Garg
our result calcu
areas of (g). (j)
Co
namic scenes.
g and Zhang. (
ulating 10 fram
) is our result c
omputer and In
a) is the origin
mes. (e) and (f
calculating 10
nformation Sci
nal frame. (b) i
f) are local are
frames. (l) an
ience
is the
eas of
nd (k)
61
Fault Recovery Mechanisms in Utility Accrual

Real Time Scheduling Algorithm
Idawaty Ahmad (Corresponding author)
Faculty of Computer Science and Information Technology
University Putra Malaysia
43400 UPM Serdang Selangor, Malaysia
Tel: 60-3-8946-6541 E-mail: idawaty@fsktm.upm.edu.my
Muhammad Fauzan Othman
Motorola Multimedia Sdn Bhd
3507 Prima Avenue, Jalan Teknokrat 5, Cyberjaya 63000 Malaysia
Tel: 60-3-8314-6777 E-mail: fauzan.othman@motorola.com
Abstract
In this paper, we proposed two recovery solutions over the existing error-free utility accrual scheduling algorithm
known as General Utility Accrual Scheduling algorithm (or GUS) (Peng Li, 2004). A robust fault recovery algorithm
called Backward Recovery GUS (or BRGUS) works by adapting the time redundancy model i.e., by re-executing the
affected task after its transient error period is over. The BRGUS is compared with a less complicated recovery algorithm
named as Abortion Recovery GUS (or ARGUS) that simply aborts all faulty tasks. Our main objectives are (1) to
maximize the total accrued utility and (2) to ensure correctness of the executed tasks on best effort basis and achieve the
fault free tasks as much as possible. Our simulation results reveal that BRGUS outperforms the ARGUS algorithm with
higher accrued utility and less abortion ratio, making it more suitable and efficient in adaptive real time system.
Keywords: Adaptive Real-Time System, Utility Accrual Scheduling, Fault Recovery, Time Redundancy, Discrete
Event Simulation
1. Introduction
A real time system is a system where the time at which events occur is important. In adaptive real time system, the
deadline misses and delays during overloads are tolerable and do not have great consequences. The definition of deadline
constraints in existing deadline based scheduling algorithms such as Earliest Deadline First (or EDF) is limited in
expressiveness by its singular metrics. A clear distinction has been made between the urgency and the important of a task
by Jensen and it is known as time/utility functions (or TUFs) (Jensen, 1985; Locke, 1986). As illustrated in Figure 1, the
urgency is measured as a deadline on X-axis and importance is captured by utility in Y-axis. A tasks time constraint
expresses the utility for completing a task as a function of when the task is completed. As shown in Figure.1, completion of
a task within its initial time and termination time will accrued some utility or zero utility otherwise. We specify the
deadline constraint of a task a binary value, downward step shaped TUF as shown in Figure 1.
The scheduling optimality criteria are based on maximizing accrued utility from those tasks. These criteria are named as
Utility Accrual (or UA) (Wu, 2004). A closer look at the UA algorithms in (Ravindran, 2005; Edward, 2007) indicates
that only the Aborted-Assured Utility Accrual (or AUA) and Handler-Assured Utility Accrual (or HUA) algorithms
consider fault in its scheduling decision. These algorithms consider the abortion and released handler for all task failures.
To the best of our knowledge, none of the prior works in UA scheduling domains consider time redundancy in their
fault recovery design model. In this paper, we apply the time redundancy model for fault recovery in UA scheduling
domain.
2. Literature Review
2.1 Fault
The term error and fault are often used as synonym. However, the definition in (Sasikumar, 1997) is appropriate in the
context of problems analyzed in this study. A fault is a defect in component or design of a system. Fault can be
categorized as permanent and transient. Permanent faults include hardware breakdowns, connection disruption as well
62
as design errors. A transient fault category is when an error disappear shortly after it appearance. These types of faults
are caused by software design error or environmental variations. We focused on transient faults since previous studies in
(Sasikumar, 1997; George, 2003) have indicated that majority of faults observed during the lifetime of a system are
caused by transient faults.
2.2 Fault Recovery
Almost every fault recovery mechanism relies on some form of redundancy. Space redundancy is employed by extra
hardware and software component that is introduced in the system only for fault recovery purposes (Mejia, 1994). A
less expensive approach is via the use of time redundancy, which is based on repeating the computation and typically
does not require a large amount of extra resources (Sasikumar, 1997). This paper focused on time redundancy paradigm,
since it is suitable for non-distributed and uniprocessor environment.
Two main approaches for fault recovery are backward and forward recovery technique. In forward recovery technique,
the system does not roll back to its previous safe state and it allows for another available resource to perform recovery.
This technique works well in distributed system environment whereas extra resources are widely available. In contrast,
the backward technique attempts to take the system back to its previous safe state and then proceeds to re-execute the
affected task (Sasikumar, 1997).
3. The Proposed Algorithms
3.1 Task Model
During the lifetime of a task, it may request one or more resources. As shown in Figure 2, a task specifies duration to
hold the requested resource in holdtime. We apply the Jensens TUFs (Jensen, 1985; Locke, 1986) to define the time
constraints of a task. Each task has an initial time and a termination time. If the termination time of a task is reached and
the task has not completed its execution, it will then be aborted. Aborting a task will change the task state from Normal
to Abort mode. Completion of task before the deadline in Normal mode accrues some uniform utility and accrues zero
utility otherwise. Following (Peng Li, 2004), our proposed algorithm measures the metric called Potential Utility
Density (or PUD) that was originally developed in (Jensen, 1985). The PUD of a task measures the amount of utility
that can be gained per unit time by executing the task. Thus, executing task in Abort mode will accrues zero PUD and
accrued zero utility to the system.
3.2 Fault Definition
In our fault model we assumed that the system have error containment capabilities to prevent the propagation of a
specific error to other tasks. The transient faults in a request can be effectively overcome by re-execution of the request
in the affected task. A request is suspended temporarily to model the transient error defects. Figure 3 shows the
procedures to detect and simulate the transient error in the system. The fault occurrences and its duration follow the
exponential distribution as detailed in Table 1. We further assumed that no error occurred during the fault recovery
process.
3.3 The Fault Recovery Algorithms
Figure 4 shows a description of Backward Recovery GUS (or BRGUS) and Abortion Recovery GUS (or ARGUS) for
error recovery. The recovery process follows three stages and is described as follows:
1) The transient erroneous period of the erroneous task, Trec is over.
After the transient error is over, the task needs to release the resource before executing the recovery algorithm in stage
2.
2) Executes the fault recovery algorithm, either ARGUS or BRGUS
In BRGUS, after the erroneous period of a request is over, the time taken to re-execute the request (i.e., holdtime) is
measured. The system is rolled back to its previous safe state (i.e., before error occurred) and then proceeds to
re-execute the affected task (Sasikumar, 1997). A request is eligible for re-execution if the holdtime of the request does
not exceed the remaining execution time of a task. Otherwise, the task will be aborted since re-executing the task will
finally result in abortion later during termination time. The AbortTime is the time taken for the task to abort the resource.
Our proposed recovery algorithm works in a best-effort basis, in the sense that a task is simply aborted if it does not
have enough time to be re-executed or continue re-execution otherwise. In ARGUS, all faulty tasks are simply aborted
and no error recovery performed (Edward, 2007).
3) Check the availability of the requested resource.
If the related resource is available, the requesting task continues its re-execution or abortion procedure directly. On the
other hand, if the resource is busy and currently being used by other task, the requesting task has to wait for the resource
in the unordered task list (or utlist).
63
4. Experimental Model
4.1 Simulation Model
We developed a discrete event simulator to verify the performance of our proposed algorithm. We used experiment
settings that are similar to those in (Peng Li, 2004) for comparison purpose. Figure 5 shows the entities involved in the
simulation model. It consists of a stream of 1000 tasks that are exponentially generated, an unordered task list (or utlist)
and a set of active resources. The requests from tasks are queued in the unordered task list before it can use the
resources. The scheduling algorithm decides which request to be executed by calculating the PUD of the request. The
tasks are assumed to be independent of each other. The events defined in the simulation model are the arrival of a task,
the completion of a task, a resource request, a resource release, the arrival of termination time of a task, the arrival of
fault in a request and the arrival of a fault recovery in a request. Table 1 summarized the details task settings used in our
model.
4.2 Performance Metric
The Accrued Utility Ratio (or AUR) metric defined in (Jensen, 1985) has been used in many algorithms stated in (Locke,
1986; Wu, 2004; Peng Li, 2004; Ravindran, 2005; Edward, 2007) and can be considered as standard metric in UA
scheduling domain. AUR is defined as the ratio of accrued aggregate utility to the maximum possibly attained utility.
The Abortion Ratio (or AR) is the ratio of aborted task to the total of task in the system. The Average Response Time (or
ART) measures the time taken for a task to complete its execution.
5. Results and Analysis
Figure 6 shows the AUR results under an increasing load. The error-free in GUS is the upper limit of our proposed
recovery algorithms. We observed that BRGUS algorithm accrued higher utility compared to ARGUS for the entire load
range. Clearly, the attempt to re-execute the faulty tasks significantly improved the accrued utility in BRGUS. Since
aborted tasks produced zero utility, this caused ARGUS to accrue lower utility. Figure 7 plots the success ratio of
BRGUS and ARGUS under an increasing error rate. We observed that the number of successfully executed tasks
decreases as the error rate and load is higher. Figure 8 verifies our speculation, proving that the abortion ratio in
BRGUS is lower than in ARGUS, which ultimately leads to higher utility accrued.
Figure 9 shows the ART effects of the proposed algorithms. Although BRGUS accumulates higher ART compared to
ARGUS, it is always lower than its execution time (i.e., 0.5seconds). This is identified as BRGUS tradeoff, whereby the
extra time taken is the time overhead incurred for re-execution of the faulty tasks during recovery process. The more
tasks get re-executed, the higher utility that BRGUS is able to accrue back to the system
6. Conclusion
This paper presents quantitative results of a best effort UA real time scheduling algorithm called BRGUS that applied
time redundancy paradigm for fault recovery. Simulation studies shows that the BRGUS achieved higher accrued utility
with smaller abortion ratio, conforming to our design objectives. Future work includes implementing in real time
operating system (RTOS) to observe the actual behavior of BRGUS algorithm.
References
Edward A.Curley (2007). Recovering From Distributable Thread Failures with Assured Timeliness in Real Time
Distributed System. Master Thesis, Virginia Polytechnic Institute and State University
George Marconi de Araujo Lima (2003). Fault Tolerance in Fixed Priority Hard Real Time Distributed Systems. PhD
dissertation, University of York
Jensen,E.D., Locke,C.D. and H.Tokuda (1985). A time driven scheduling model for real time system. Proceeding of
Real Time System Symposium, 112-212
Locke,C.D. (1986). Best-effort decision making for real time scheduling. PhD dissertation, Carnegie Mellon University
Mejia Alvarez.P and D.Mosse (1994). A Responsiveness Approach for Scheduling Fault Recovery in Real Time
Systems. Proceeding of Fifth IEEE Real-Time Technology and Applications Symposium,4-13
Peng Li (2004). Utility Accrual Real Time Scheduling Model and Algorithm. PhD dissertation, Virginia Polytechnic
Institute and State University
Ravindran,B., Jensen,E.D., Peng Li (2005). On Recent Advanced in Time/Utility Function Real-Time Scheduling and
Resource Management. Proceeding of Eight IEEE International Symposiums on Object-Oriented Real-Time Distributed
Computing (ISORC), 55-60.
Sasikumar Punnekat (1997). Schedulability Analysis for Fault Tolerant Real Time Systems. PhD dissertation,
University of York
Wu,H., Ravindran,B. and Peng Li (2004). CPU Scheduling for Statistically Assured Real-Time Performance and
64
Improved Energy Efficiency. Proceeding of the 2
nd
IEEE/ACM/IFIP International Conference of Hardware/Software
Codesign and System Synthesis, 55-60
Table 1. Simulation Parameters
Parameters Value Descriptions
load 0.2 to 1.5 Range of load
C_AVG 0.5 seconds Task Average Execution Time
iat Exponential(C_AVG/load) Task Inter-Arrival Time
holdtime Normal (0.25,0.25) Time duration for holding a resource
max_au Normal (10,10) Tasks Maximum Utility
mean_tasks Exponential (0.1) Inter- arrival of error in the system
MEAN_DURATION Exponential (0.1) Duration of error (transient)
MAX_RESOURCES 5 Number of available resources
Figure 1. The Step TUF [1], [4]
Figure 2. Task Model
Figure 3. Fault Assignment
65
Figure 4. Fault Recovery Algorithms
66
Figure 5. Simulation Model
Figure 6. Accrued Utility Ratio (AUR) vs. Average Load
67
Figure 7. Success Ratio (or SR) vs. Task Error Rate
Figure 8. Abortion Ratio (or AR) vs. Average Load
68
Figure 9. Average Response Time (or ART) vs. Task Error Rate
69
The Fusion of Remote Sensing Images Based on Lifting

Wavelet Transformation
Wei Zhang, Jinzhong Yang, Xiaohong Wang & Qinghua Yang
China Aero Geophysical Survey & Remote Sensing Center for Land and Resources
Beijing 100083, China
Qin Huang
Zhejiang University Press, Hangzhou
Zhejiang 310028, China
Zhenkun Wang
Department of Resources and Earth Science
China University of Mining & Technology
Beijing 100083, China
Abstract
The fusion of remote sensing images has become one of the new hotspots in recent years. It can not only improve
spatial resolution effectively, but can keep the integrity of the multi-spectral image. In this paper, we take the Hangzhou
area as an example and put forward a new image fusion based on lifting wavelet transformation, and carry out the
qualitative and quantitative comparison to the spatial characteristics and spectral characteristics with traditional fusion
methods, such as IHS transform, PCA transform and Brovey transform. The results show that by lifting scheme wavelet
transformation the characteristics and information of the fusion image have been preserved greatly, and the spatial
quality of the original image has been better improved. Furthermore, this method is much simpler and easier.
Keywords: Remote sensing, Image fusion, Lifting wavelet transformation, IHS transform, PCA transform, Brovey
transform
1. Introduction
The remarkable characteristic of modern remote sensing is to use multiple sensors, multiple resolutions, multi-spectral
and multi-temporal technology together to obtain the remote sensing image with different spatial resolution, temporal
resolution and spectral resolution(Zhu S L, Zhang Z M. 2000.). Spatial resolution and spectral resolution are the two
highly related factors for optical sensors. Usually, the sensor with higher spectral resolusion has lower spatial resolution,
though it can capture more radiation information from different object styles in various electromagnetic spectrums, and
it can not obtain optimum resolution(Mara Gonzlez-Audcana. 2004.).
Image fusion is an important image enhancement technology, which can fuse high spatial resolution and multi-spectrum
image. So far, we have developed various remote sensing image fusion technologies, such as Brovey transform(Vrabel,
Jim. 1996.), Principle Component Analysis (PCA)( Haydn R. 1982.), IHS transform(Chavez S, Sides C, Anderson A.
1991.), high-pass filtering(Chang S G, Yu B. 2000.), multi-resolution analysis (Aiazzi B, Alparone L, Baronti S, et.al.
2002.) and wavelet transformation(Chipman J, Orr M, Lewis N. 1995.)( Li H, Manjunath B S, and Mitra S K.
1995.)( Kingsbury G. 2000.). Brovey transform, IHS transform, PCA transform are most widely used in many remote
sensing commercial softwares. But they have two defects. Firstly, it requires three low-resolution images in Brovey
transform and IHS transform, or even more for PCA transform. Secondly, compared to low-resolution image, there will
be a general spectral distortion in fused images. During the fusion of remote sensing image, the spectral and detailed
information should be both taken into consideration. Wavelet transformation is a famous algorithm in recent years,
which can retain effective original spectral information while enhancing spatial resolution. First, original image is
70
decomposed into approximate and detailed image by wavelet transformation, and then fuse the image on each
characteristic fields by high frequency substitution. Though wavelet transformation has multi-resolution and more
advantages in the image fusion with different image resolution, it has larger computational complexity.
In this paper lifting scheme (Calderbank R, Daubechies I, Sweldens W. 1998.) and a new wavelet transformation based
on lifting scheme is introduced. To prove its effectiveness, the experiment with ETM+ multi-spectral image and
panchromatic imagery based on the Hangzhou area is conducted, and the spectral quality and spatial quality are also
analyzed and compared with the traditional fusion such as IHS transform, PCA transform and Brovey transform.
2. Common image fusion algorithms
2.1 IHS transform
IHS transform refers to a transformation between RGB space and IHS space, it can transform the color space from red
(R), green (G), blue (B) to hue (H), saturation (S) and intensity (I). The former describe the color from physics while the
latter from human subjective vision. Firstly, the spatial characteristic (I) and spectral characteristics (H, S) is separated.
Secondly, the high-resolution image is stretched according to I component, and then I was substituted by the
high-resolution image. At last, IHS inverse transform is carried out to improve resolution and enhance spectral
characteristics. IHS forward and reverse transformation formulas are as follows (Zhang W, Ding X W, Chen H L.
2007.).
2.2 PCA transform
Principle Component Analysis(PCA)is a multi-dimensional orthogonal linear transformation based on computing the
statistical character of the imagery, which is called K-L transform in mathematics(Zhang W, Mao Q M, Zhang X C,
et.al. 2005.). It aims to compress and integrate the multi-band imaginary information into one image, using main
components information to substitute the information of multi-band image, so that the information of different bands
could be furthest displayed in new and irrelevant images. PCA transform first requires PCA forward transformation to
obtain n (n3) components and stretches high resolution panchromatic band and other data to obtain the same mean and
variance with the first principal component, and substitutes the first principal component with stretched high resolution
image and fuse the image by PCA reverse transformation.
2.3 Brovey transform
Brovey transform is a simpler but widely-used RGB color fusion. The algorithm decomposes the phase space of the
multi-spectral image into color and intensity, which essentially substitute the I component of multi-spectral image with
high resolution image. It simplifies the image transformation coefficience to furthest reserve the multi-band image
information, and all the intensity information is transformed into high resolution panchromatic image. The formulas of
Brovey transform are as follows:
(1)

In this formula, R stands for the red band of multi-band image, G stands for the green band of multi-band image, B
stands for the blue band of multi-band image, and PAN stands for the intensity of high spatial resolution image.
In addition, Brovey transform need better pretreatments and noise filtering in order to reduce redundant data and
non-spectral information.
3. Image fusion based on lifting wavelet transformation
3.1 Lifting wavelet scheme
Lifting wavelet transformation is a new wavelet construction proposed by Sweldens based on lifting scheme. Lifting
scheme consists of split, predict and update. This reversible wavelet transformation is applied widely in the image
process and becomes the core part of JPEG2000. Besides, lifting method avoids FFT operation of a convolution and
requires only shift, addition and subtraction operations, which makes the hardware implementation easier.
The frame of lifting format is as follows:
split: suppose there is a source data set s
0
, and split the data set s
0
into two subsets s
1
and d
1
. The simplest way is to
split s
0
into even sequence d
1
and odd sequence s
1
, namely
split (s
0
) = (s
1
, d
1
) (2)
predict: according to correlation between image data, a prediction operator p which is irrelevant to data structure is
adopted. First act p filter on even signal to obtain the prediction value p(s
1
) of odd signal. In practical application, the
prediction value p(s
1
) is quite close to d
1
, which can be substituted by the difference between d
1
and p(s
1
):
+ + =
+ + =
+ + =
) (
) (
) ( Re
B G R B Pan Blue
B G R G Pan Green
B G R R Pan d
71
d
1
= d
1
-p(s
1
) (3)
This difference has less information than d
1
.
update: some characteristics o f the coefficient subset s
1
(such as mean) and s
0
are different, so the update operator u is
introduced to update s
1
and preserve the integrity of signal s
0
:
s
1
= s
1
+u(d
1
) (4)
So far the subset s
1
and d
1
substitute the original dataset s
0
. When the recursion of the previous process repeats, s
1
will
be divided into subsets s
2
and d
2
. d
2
is substituted by difference between d
2
and p(s
0
), and s
2
is substituted by the sum of
s
2
and u(d
2
). When the recursion goes on the original dataset s
0
is substituted by a more compact expression
.
The decomposition and reconstruction of lifting scheme wavelet show as Fig.1(Sweldens W. 1998.)
.
This paper uses (4,2) lifting scheme to decompose and reconstruct image data:
(5)
(6)
Formula (5) is the forward wavelet transformation based on lifting scheme, Formula (6) is the reverse wavelet
transformation based on lifting scheme.
3.2 Wavelet transform based on lifting scheme
Lifting wavelet transformation is a new wavelet construction based on more efficient and simpler-to-inverter lifting
scheme, which needs less operation storage space and has better expansibility. Lifting wavelet transformation fuses the
multi-spectral and panchromatic bands of ETM+ images. The concrete process shows in Fig2.
In this paper, we first match PAN image (high resolution image) and MS image (multi-spectral image), and resample
MS image to get the same pixel as PAN images, and then stretch PAN image according to mean and variance in the
single band image.
Lifting wavelet transformation of PAN image can be conducted to obtain four components by formula (5), which are
HH, LH, HLand LL. And among them, HH, LHand HL represent spatial detailed information of high
resolution image of P resolution, LLrepresents approximate image of P/2 resolution. Lifting wavelet
transformation of MS single-band image can be similarly conducted to obtain Mi-HH, Mi-LH, Mi-HLand
Mi-LL.
Substitute LLwith Mi-LLand reverse transformation by formula (6) to obtain Mi. Till then, the fusion of PAN
images and MS single-band images has been completed.
This process is repeated until the images of other bands are fused and the spatial detail information of high resolution
images are conducted to various bands of multi-spectral images, so that the original images have kept both spectral
information and spatial detail information.
4. Experimental Results
In order to prove the effect of lifting wavelet transform, the Landsat-7 ETM+ multi-spectral image with the resolution
of 30m and panchromatic image with the resolution of 15m are fused. First, the 7, 4 and 3 bands of ETM+ image are
resampled to obtain the same pixel as panchromatic image. The size of the 7, 4 and 3 bands and panchromatic band of
ETM+ image is 512512.
From Fig.3 we can see that all the spatial information of fused image is better than the original image by visual
interpretation. And among them, Brovey transform leads to larger spectral distortion. But it is difficult to identify which
method can keep better spatial detail information or less spectral distortion. Therefore, the quantitative evaluation of the
fused images is separately carried out from the spatial quality and the spectrum quality.
4.1 Evaluation of spatial quality
In this paper, the entropy and average gradient as the two indicators to evaluate the enhancement effect of the fused
images. Among them, entropy represents the abundance of image information, the higher the entropy is, the more
{ }
1 1
, , , ,
n n
s d s d
+
+ =
+ + + +
+ =
4
) ( ) 1 (
) 2 ( ) (
16
)] 4 2 ( ) 2 2 ( [ )] 2 2 ( ) 2 ( [ 9
) 1 2 ( ) (
1 1
0 1
0 0 0 0
0 1
n d n d
n s n s
n s n s n s n s
n s n d
+ + + +
+ = +
+
=
16
)] 4 2 ( ) 2 2 ( [ )] 2 2 ( ) 2 ( [ 9
) ( ) 1 2 (
4
) ( ) 1 (
) ( ) 2 (
0 0 0 0
1 0
1 1
1 0
n s n s n s n s
n d n s
n d n d
n s n s
72
information the image has. Average gradient shows the fine contrast, texture characteristic and clarity. Higher average
gradient has more image levels and the image is clearer. The formula of the entropy and average gradient is as follows:
(7)
H is entropy, P
i
is the grey value which is equals to the probability of i and L are total grey level.
(8)
Among them, is average gradient, and are the gradient on row-column direction, M and
N represent the row number and column number of image f.
According to formula (7) and formula (8), the entropy and average gradient of original ETM+ and fused images are
shown as tab2. We can see that compared to Brovey transform, IHS transform and PCA transform, the entropy and
average gradient of the image fused by lifting wavelet transformation are larger, which means this method can hence the
spatial detail information much better. Analyzed from algorithm principle, this method preserves the high frequency
information by low frequency substitution in wavelet domain, which corresponds to the detail information of the
multi-spectral image, so the entropy and average gradient of the fused image are larger.
4.2 Evaluation of spectral quality
In order to evaluate the spectral information, the correlation efficient and deviation index are adopted to measure the
fusion quality. The correlation efficient shows the similar degree of the two images and the alternation of spectral
information, while the deviation index shows the offset degree of the two images. The formula is as follows:
(9)
Among them, R(f
1
, f
2
) is the correlation efficient of f
1
and f
2
, and f
1
(i, j) and f
2
(i, j) are the grey values of f
1
and f
2
on the
point (i, j). u
1
and u
2
are the average grey values of the two images.
(10)
Among them, D is the deviation index of compared to f, M and N are the row number and column number of f, and
and f (i, j) are the grey values on the point (i, j) of the fused image and original image.
We can see that compared to Brovey transform, IHS transform and PCA transform, the image fused by lifting scheme
wavelet transformation has the highest correlation coefficient and the lowest deviation index, which means that this
method keeps more spectral information of the low resolution images. Analyzed from algorithm principle, Brovey
transform fuses high resolution panchromatic images and multi-spectral images directly by multiplication, while IHS
transform substitutes I component with panchromatic image and results in spectral distortion. PCA transform substitutes
PC1 component with panchromatic image, which results in the loss of spectral information. The method proposed in
this paper makes full use of the thoughts of wavelet transformation and fuses the images on the sub-band to preserve
spectral information.
5. Conclusions
In this paper, we introduce the lifting wavelet transformation fusion of low resolution multi-spectral image and high
resolution image. This method decomposes the panchromatic image with high resolution and the multi-spectral image
with low resolution by lifting wavelet transformation. It fuses the low-frequency component of the high resolution
image into the low-frequency component of the multi-spectral images, which further reserves spectral information of
original image and enhances spatial resolution of fused image. The results show that compared to Brovey transform,
IHS transform and PCA transform, the lifting wavelet transformation fusion is a better fusion method.
Acknowledgement
This study is supported by the major project of the Science and Technology bureau of Hangzhou city under grant
HZZFCG-2005-A2.
References
Aiazzi B, Alparone L, Baronti S, et.al. (2002). Context-driven fusion of high spatial and spectral resolution data based
1
0
ln
L
i i
i
H P P
=
=
_
( )( )
( ) ( )
2 2
1 1
1 1
, ,
1
1 1 2
M N
i j
i j
f i j f i j
G
M N

= =
V +V
V =

__
G V
( ) ,
i
f i j V ( ) ,
j
f i j V
( )
( )
( )
( )
( )
( )
( )
( )
( )
1 2
1 2
1 2
,
1 2
2 2
1 2
, ,
, ,
,
, ,
f f
i j
f f
i j i j
f i j f i j
R f f
f i j f i j

(

=
( (

( (

_
_ _
( ) ( )
( ) ,
, ,
1
,
i j
f i j f i j
D
MN f i j
=
_
( ) , f i j
73
on oversampled multi-resolution analysis. IEEE Transactions on Geosciences and Remote Sensing, 2002,
40(10):2300-2312.
Calderbank R, Daubechies I, Sweldens W. (1998). Wavelet transform that map integers to integers. Applied and
computational harmonic analysis, 1998, 5(1):332-369.
Chang S G, Yu B, Vetterli M. (2000). Adaptive wavelet thresholding for image denoising and compression. IEEE
Transaction on Geoscience and remote sensing, 2000, 9(9):1532-1546.
Chavez S, Sides C, Anderson A. (1991). Comparison of three different methods to merge multi-resolution and
multi-spectral data: Landsat TM and SPOT Panchromatic. Photogrammetric Engineering and Remote Sensing, 1991,
57(3):295-303.
Chipman J, Orr M, Lewis N. (1995). Wavelets and image fusion. IEEE Transactions on Image Processing. 1995,
3(1):248-251.
Haydn R. (1982). Application of the IHS color transform to the processing of multi-sensor data and image enhancement.
Proceedings of the International Symposium on Remote Sensing of Arid and Semi-Arid Lands[C]. Cairo, Egypt,
1982:599-616.
Kingsbury G. (2000). The dual-tree complex wavelet transform with improved orthogonality and symmetry properties.
IEEE International Conference on Image Processing, 2000, 1(1):375-378.
Li H, Manjunath B S, and Mitra S K. (1995). Multisensor image fusion using the wavelet transform. Graphical Models
and Image Processing, 1995, 57(2):235-245.
Mara Gonzlez-Audcana. (2004). Fusion of multi-spectral and panchromatic images using improved IHS and PCA
mergers based on wavelet decomposition. IEEE Transactions on Geosciences and Remote Sensing, 2004, 42 (6):
1291-1299.
Sweldens W. (1998). The lifting scheme: A construction of second generation wavelets. Journal of Mathematical
Analysis, 1998, 29 (2):511-546.
Vrabel, Jim. (1996). Multispectral imagery band sharpening study. Photogrammetric Engineering & Remote Sensing,
1996, 62 (9):1075-1083.
Zhang W, Ding X W, Chen H L. (2007). SAR-image enhancement and fusion based on improved IHS method. Journal
of Marine Sciences, 2007, 25(1):73-79. (In Chinese)
Zhang W, Mao Q M, Zhang X C, et.al. (2005). On features of fusion technique of remote sensing images with different
high resolutions. Donghai Marine Science, 2005, 23 (1):23-31. (In Chinese)
Zhu S L, Zhang Z M. (2000). The acquisition and analysis of remote sensing image. Beijing: Science Press, 2000. (In
Chinese)
Table 1. IHS Forward and Reverse Transformation Formulas
Condition Forward Transformation Formula Reverse Transformation
R>B G or 0 H<1
I = (R + G + B ) / 3
H = (G B) / [ 3 (IB ) ]
S =1B / I
R = I (1 + 2 S 3 S H )
G = I (1 S + 3 S H )
B = I (1 S )
G >R B or 1 H<2
I = ( R + G + B ) / 3
H = ( B R) / [3 ( I R) ] + 1
S = 1 R / I
R = I ( 1 S )
G = I (1 + 5 S 3 S H )
B = I (1 4 S + 3 S H )
B>G R or 2 H<3
I = ( R + G + B ) / 3
H = ( R G ) / [ 3 ( I G ) ] + 2
S = 1 G / I
R = I (1 7 S + 3 S H )
G = I (1 S )
B = I (1 + 8 S 3 S H )
s s
s s
s s
74
Table 2. Statistic parameters of spatial performance between original ETM+ images and fusion images
Methods
Statistics
original
image
Brovey
transform
IHS transform
PCA
transform
lifting wavelet
transformation
Entropy
Band7 3.8945 4.4069 5.2968 5.2255 5.4947
Band4 3.9982 4.6351 5.1402 5.1850 5.5129
Band3 3.6828 4.4202 5.2752 5.3013 5.4828
Average
gradient
Band7 16.1102 23.4662 20.6459 24.0035 24.1366
Band4 13.2060 17.6053 19.3091 20.9073 21.1347
Band3 13.6423 19.8649 17.5965 15.4116 23.4283
Table 3. Statistic parameters of spectral performance between original ETM+ images and fused images
Methods
Statistics
original
image
Brovey
transform
IHS transform
PCA
transform
lifting wavelet
transformation
correlation
coefficient
Band7 1.0000 0.9060 0.9321 0.9197 0.9741
Band4 1.0000 0.9410 0.9445 0.9336 0.9802
Band3 1.0000 0.9117 0.9411 0.9797 0.9812
deviation
index
Band7 0.0000 0.5535 0.5040 0.4971 0.3731
Band4 0.0000 0.4651 0.3854 0.3804 0.3166
Band3 0.0000 0.4963 0.3707 0.2895 0.2381
Figure 1. Decomposition and reconstruction of lifting scheme wavelet
Figure 2. Remote sensing image fusion based on lifting scheme wavelet transformation
+
+
-
-
1+j
V
M
G
V
j
V
j
even/
odd
sp lit
P
U U
P
even/
odd
merge
odd
even
odd
even
Computer

and Informati
(a) Origina
(c) Fusion im
(e) Fusion
ion Science
al ETM+743 b
mage using Bro
image using P
Figure 3

bands image
ovey transform
PCA transform
3. Experimenta

m (d
m (f) Fu
al results by di

(b) Original E
d) Fusion imag
usion image us
ifferent fusion

ETM+PAN im
e using IHS tra
sing the propo
n methods
Feb
age
ansform
sed method
bruary, 2009
75
76
Individual Differences, Perceived Ease of Use, and Perceived Usefulness

in the E-Library Usage
Yusliza Mohd.Yusoff (Corresponding author)
School of Management, Universiti Sains Malaysia
11800 Minden, Penang, Malaysia
Tel: 6-03-5543-5847 E-mail: yusliza1977@yahoo.com
Zikri Muhammad
School of Humanities, Universiti Sains Malaysia
11800 Minden, Penang, Malaysia
E-mail: zikri69@hotmail.com
Mohd Salehuddin Mohd Zahari
Faculty of Hotel and Tourism Management
Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia
E-mail: salehuddinm@yahoo.com
Ermy Syaifuddin Pasah
Faculty of Administrative Science & Policy Studies
E-mail: ermysyaifuddin@yahoo.com
Emmaliana Robert
Faculty of Administrative Science & Policy Studies
E-mail: princess_emmaliana83@yahoo.com
Abstract
This study examines the usage of e-library among students in a public university in Malaysia using the Technology
Acceptance Model. Data from a questionnaire survey of 201 students were analyzed using linear regression. The results
show that individual differences (computer self-efficacy and knowledge of search domain) had a significant positive
relationship with perceived ease of use. Perceived ease of use showed significant relationship with perceived usefulness
but non significant with the actual usage of the e-library. For perceived usefulness, it showed a significant relationship
with the actual usage of the e-library. Lessons and discussions for individual differences, perceived ease of use,
perceived usefulness, and actual usage are presented.
Keywords: Computer self-efficacy, Knowledge of search domain, Perceived ease of use, Perceived usefulness, Actual
usage, E-library, Malaysia
1. Introduction
The term e-library has been associated with many connotations such as digital library, hybrid library, or virtual library.
There are different definitions in the literature as to what constitutes an e-library. Akla (2002) as cited in Ramayah,
Aafaqi, and Ignatius (2004), for example, defined e-library as the digital library that requires technology to link the
resources of many libraries and information services. Deb, Kar, and Kumar (2003) defined e-library as a system that is
accessible from anywhere via the Internet, to deliver knowledge directly to their users, without being confined to the
contents neither of a physical library nor by being caught in a web of unorganized, unmanaged information. The
emergence of e-library provides more opportunities for users to access a variety of information resources.
77
There a lots of reasons why e-library is important especially for the students in doing their course work assignments, so
that they can do their work more efficiently and faster. Chen (2000) pointed out that e-library use the internet to create
and store massive amounts of digital media information. To satisfy the demands of library users, e-library provides
highly efficient and consistent methods for the search and retrieval information. Furthermore, e-library is the integration
of materials, collection, information, services and operation (Chen, 1999). Ke (2000) shows that e-library are composed
of three important factors; electronic collection, electronic operation, and electronic service. Besides that, the goal of
e-library is to perform online all the functions of the traditional library, plus many more available in todays digital
world (Deb et al., 2003).
Millions of dollars have been invested into the development of e-library. However, research on the e-library is still in its
infancy (Xie, 2006) particularly the applicability of the Theory of Acceptance Model (TAM) on user acceptance of
e-library. Therefore, there is a need for research to identify the factors that determines users adoption of e-library. Using
the TAM as theoretical framework, this study investigates the effect of the set of individual differences to the usage of
e-library.
2. Review of literature and hypotheses development
2.1 Technology Acceptance Model (TAM)
Davis (1986) developed the Technology Acceptance Model (TAM) in studying the determinants of IT usage. The goal
of TAM was to provide an explanation of the determination of computer acceptance that is generally capable of
explaining user behaviour across a broad range of end-user computing technologies and user populations, while at the
same time being both parsimonious and theoretically justified (Davis, 1989). TAM can be seen as an adaptation of the
generic Fishbein and Ajzens Theory of Reasoned Action (TRA) and was developed to explain individual system use in
the workplace (Davis, 1989). TAM posits that perceived ease of use (PEOU) and perceived usefulness (PU) are
important factors that determine the users attitude toward his or her intention to use and actual usage of information
systems (IS). According to TAM, usage behavior is a direct function of behavioral intention which in turn a function of
attitude toward usage reflect feelings of favorableness or unfavorableness toward using the technology and PU which
reflect the belief that using the technology will enhance performance. Attitude is determined jointly by PU and PEOU
(Davis, 1989). Furthermore, a key purpose of TAM is to provide a basis for discovering the impact of external variables
on internal beliefs, attitudes, intentions, and usage. The TAM developed by Davis (1986) is shown in Figure 1.
2.1.1 Individual Differences
According to Hong, Thong, Wong, and Tam (2002), individual differences are believed to be most relevant to both IS
success (Harrison & Rainer, 1992; Zmud, 1979) and human computer interaction research (Dillon & Watson, 1996).
Individual differences play a major role in determining user performance on information retrieval systems (Chen,
Czerwinski, & Macredie, 2000). Previous studies have examined various individual differences such as computer
self-efficacy (Chau, 2001; Hong et al., 2002; Ramayah & Aafaqi, 2004; Thong, Hong, & Tam, 2004; Ramayah, 2006;
Amin, 2007; Hasan, 2006), knowledge of search domain (Hong et al., 2002; Thong et al., 2004; Ramayah, 2006),
computer experience (Thong et al., 2004; Ramayah, 2006), and demographic variables (Lu, Yu, Liu, & Yao, 2003) on
technology acceptance. Following research by Hong et al. (2002), this study will examine computer self-efficacy and
knowledge of search domain as individual differences in the context of e-library usage. In the following section, these
individual differences will be explored that may influence user acceptance of e-library.
2.1.2 Computer Self-Efficacy
In the context of using IS, computer self-efficacy represents an individuals perceptions of his or her ability to use
computers in the accomplishment of task, rather than reflecting simple component skills (Compeau & Higgins, 1995, p.
191). Prior IS research has found that computer self-efficacy has a positive impact on perceived ease of use (Agarwal,
Sambamurthy, & Stair, 2000; Venkatesh, 2000; Hong et al., 2002; Thong et al., 2004; Hasan, 2006; Amin, 2007). The
proposed relationship between computer self-efficacy and PEOU is based on the theoretical argument by Davis (1989)
and Mathieson (1991). Library and information science researchers have also recognized the possible impact of
computer literacy on increasing usage of information retrieval systems (Jacobson & Fusani, 1992; Davies, 1997). Thus,
H
1
: Computer self-efficacy is positively related to perceived ease of use of the e-library.
2.1.3 Knowledge of Search Domain
Knowledge of search domain is another individual difference that may positively affect PEOU of e-library. According to
Thong et al. (2004), in a study of information seeking behaviour in a hypertext environment, domain experts were found
to conduct faster and more focused searches than did the novices (Marchionini, Lin, & Dwiggins, 1990). Prior studies
(Hong et al., 2002; Thong et al, 2004; Ramayah, 2006) have shown that knowledge of search domain does have a
significant and positive relationship with PEOU. The associated hypothesis for testing in this study is as follows:
H
2
: Knowledge of search domain is positively related to perceived ease of use of the e-library.
78
2.1.4 Perceived Ease of Use
PEOU has been extensively studies in the IS literature. PEOU is the degree to which a person believes that using a
particular system would be free of effort (Davis, 1989; Dholakia & Dholakia, 2004). The easier it is for a user to
interact with a system, the more likely he or she will find it useful (Thong et al., 2004). There is substantial empirical
support for this view (Chau, 2001; Hong et al., 2002; Thong et al., 2004; Ramayah & Aafaqi, 2004; Ramayah et al.,
2004; Lallmahamood, 2007; Shim & Viswanathan, 2007; Amin, 2007). In the digital libraries setting, Hong et al. (2002)
submitted that digital libraries need to be both easy to learn and easy to use. Based on the evidence that link perceived
ease of use with perceived usefulness in the context of e-library, it is hypothesized that:
H
3
: Perceived ease of use is positively related to perceived usefulness of the e-library.
When the application of the e-library is perceived to be easier to use, it is more likely to be accepted by the users.
Although most researches have found perceived usefulness to be directly related to usage, there are some findings that
show no significant effect on usage such as Ndubisi, Jantan, and Richardson (2001). There are also many researches
that have found ease of use to be influential in system usage (Adams, Nelson & Todd, 1992; Davis, 1989; Ramayah,
Siron, Dahlan, & Mohammad, 2002; Ramayah & Jantan, 2004; Ramayah & Aafaqi, 2004). Hence, the hypothesis
posited in this study is as follows
H
4
: Perceived ease of use is positively related to actual usage of e-library.
2.1.5 Perceived Usefulness
PU is defined as the degree to which a person believes that using a particular system would enhance his or her job
performance (Davis, 1989). Within the organizational context, a system that is high in perceived usefulness is one that
the user believes will have a positive use-performance relationship. In fact, IS adoption research suggests that a system
that does not help people perform their jobs is not likely to be received favourably (Nysveen, Pedersen, &
Thornbjomsen, 2005, p. 537). The ultimate reason that users exploit digital libraries is that they find the systems useful
to their information needs or search tasks (Hong et al., 2002). Past researches (Ramayah et al., 2004; Ramayah &
Aafaqi, 2004; Davis, 1989; Mathieson, 1991; Adams et al., 1992; Segars & Grover, 1993, Igbaria, Iivari, & Maragahh,
1995; Igbaria, Zinatelli, Cragg, & Cavaye, 1997; Ndubisi et al., 2001; Ramayah, Ignatius, & Aafaqi, 2002, Ramayah,
Sarkawi, & Lam, 2003) has shown that PU influences computer usage directly. Hence, it is hypothesize that:
H
5
: Perceived usefulness is positively related to actual usage of e-library.
3. Research model
Based on the above discussion, an augmented TAM was proposed (Figure 2). The model was adapted from the previous
study by Hong et al. (2002), Thong et al. (2004), and Ramayah et al. (2004).
4. Methodology
4.1 Research design and sampling
The population for this study was students from one of the Malaysian public university. A convenience sample was
drawn from this population (n=201). Convenience sampling refers to the collection of information from members of the
population who are conveniently available to provide it (Sekaran, 2003). A survey-questionnaire approach was used to
gather data in this study. A questionnaire was personally administered to the student in the university. This method is
chosen because the designed questionnaire could be collected within a short period of time and any doubts that the
respondents might have on any question could be clarified on the spot (Sekaran, 2003).
The average age of the students was 23 years, and the majority (96%) were single. More than half of the students were
female (58.7%), while 82.6 per cent were Malay, 11.9 per cent were Sarawak Bumiputera, and 5 per cent Sabah
Bumiputera. In terms of status of candidacy, most students reported they are full-time students (77.6%), the remainder
part-time students. In respect of faculty, almost half of the students came from Faculty of Administrative Science and
Policy Studies (43.3%), Faculty of Information Technology and Quantitative Sciences (21.4%), Faculty of Architecture,
Planning and Surveying (2.5%), Faculty of Mechanical Engineering (1.5%), Faculty of Applied Sciences (3.5%),
Faculty of Hotel and Tourism Management (6.5%), Faculty of Business Management (15.4%), Faculty of Civil
Engineering (0.5%), Faculty of Art and Design (1%), and Faculty of Accountancy (4.5%).
4.2 Measurements
Computer self-efficacy was measured with the ten-item measure developed by Compeau and Higgins (1995).
Knowledge of search domain was tapped by Daviess (1997) two-item scale. PEOU and PU were measured by four
items each, taken from Hong et al. (2002). These items in the questionnaire were rated on five-point Likert scales raging
from 1 = strongly disagree o 5 = strongly agree, except computer self-efficacy, which was rated from 1 = not at all
confident to 5 = totally confident. Actual usage was assessed with the frequency of usage and daily usage measure
developed and validated by Igbaria et al. (1995) and Thompson, Higgins, & Howell (1991). Frequency of e-library
79
usage is measured on a five-point scale ranging from 1 = never/almost never to 5 = several times a day, whilst daily
e-library usage is measured by the amount of time spent daily on the databases. A five-point scale ranging from 1 =
never/almost never to 5 = more than three hours per day is used to measure daily e-library usage.
5. Findings
5.1 Means, standard deviations, reliability, and correlations between all study variables
A summary of the means, standard deviations, reliability, and correlations is provided in Table 1.
It can be seen that the mean value for each of the studys variables ranges from 2.856 to 3.479, with a standard deviation
of .461 to 1.123. It revealed that all variables charted higher than the midpoints of their respective scales except for actual
usage. To assess reliability among the studys variables, Cronbach's alpha values were generated. Alpha values of .765
(computer self-efficacy), .776 (knowledge of search domain), .846 (PEOU), .855 (PU), and .820 (actual usage) suggested
more than adequate reliability within the respective factors (Nunnally & Bernstein, 1994).
According to correlations between variables, PEOU and PU have a strong and positive relation (r=.558, p<.01).
Computer self-efficacy (r=.243, p<.01) and knowledge of search domain (r=.434, p<.01) affects PEOU in a positive
manner, as well. However, this effect is not as strong as PU. PEOU (r=.183, p<.01) and PU (r=.224, p<.01) also
positively affect actual usage. The zero order correlation between computer self-efficacy and PU and computer
self-efficacy and actual usage were non-significant.
5.2 Direct effects of computer self-efficacy and knowledge of search domain and perceived ease of use variables
The results of the tests of effects of computer self-efficacy and knowledge of search domain and PEOU are presented in
Table 2. It can be seen that R
2
value of .258 suggests that 25.8 per cent of the variance in the PEOU is explained by
computer self-efficacy and knowledge of search domain. In addition R
2
change (0.258) is significant. This implies that
the additional 25.8 per cent of the variation in the PEOU is explained by knowledge of search domain and computer
self-efficacy. The Durbin-Watson index (1.963) shows that auto-correlation was not the problem. Knowledge of search
domain ( = .459, p < .01) and computer self-efficacy ( = .159, p < .05) had significant and positive effects on PEOU.
Thus, H
1
and H
2
were supported.
5.3 Direct effects of perceived ease of use and perceived usefulness variables
As depicted in Table 2, the coefficient determination (R) was found to be .325 indicating that 32.5 per cent of PU is
explained by PEOU. The R change (.325) is significant. This implies that the additional 32.5 per cent of the variation
in PU is explained by PEOU. The value of Durbin Watson index is 1.883. PEOU ( = .570, p < .01) had significant and
positive effects on PU. Based on this result, it provided support for H
3
.
5.4 Direct effects perceived ease of use and perceived usefulness and actual usage variables
An examination of Table 2 revealed that only PU ( = .177, p < .05) had significant and positive effect on actual usage.
Contrary to expectation, there was no significant relationship between PEOU and actual usage. Based on these results,
H
5
and H
4
are rejected. R is 5.5 percent. The R change is 5.5 per cent and was found to be significant (p<.01). This
findings show that an additional 5.5 percent of the variations in the actual usage is explained by PEOU and PU. The
Durbin-Watson index is at 2.111, which is within the acceptable range 1.5 to 2.5 (Coakes & Steed, 2003). It shows no
auto-correlation problems detected in the model.
6. Discussion and conclusions
This study tested the relation between individual differences (computer self-efficacy and knowledge of search domain)
and PEOU on the one hand and PEOU and PU towards the actual usage of e-library on the other. In addition, PEOU is
examined in relation to PU. In general, four hypotheses were confirmed, but one was rejected.
6.1 Computer self-efficacy
As predicted, there was a positive relationship between computer self-efficacy and PEOU of the e-library. Thus, it
received support for research hypothesis. This finding is consistent with previous research that has found positive
relationship between computer self-efficacy and PEOU of use in IS research (e.g., Hong et al., 2002; Thong et al., 2004;
Ramayah & Aafaqi, 2004; Hasan, 2006; Amin, 2007). This finding implies that the high level of computer self-efficacy
indicated that the students will be have high levels of confidence in using e-library in general especially in doing their
assignment and more likely to find the e-library is easy to use. Hence, high level of computer self-efficacy indicates that
the students will be often using e-library in their study. Besides, students can also improve their performance in their
study once they are confident in using e-library.
6.2 Knowledge of search domain
As shown in the regression result of this study, knowledge of search domain was found to be contributing to PEOU of
e-library. This result support prior research of Hong et al. (2002), Thong et al. (2004), and Ramayah (2006) on user
80
acceptance of digital libraries or e-library. Domain experts could conduct faster and more focused searches than novices
as they are able to separate relevant information from irrelevant responses and increase effective searches (Ramayah,
2006). This finding supports the recommendation of library science researchers to provide customized interfaces to
different individuals (Archer, Head, & Yuan, 1996; Meadow, Wang, & Yuan, 1995). When the students are familiar and
knowledgeable with e-library it will encourage the students using e-library frequently in their study.
6.3 Perceived ease of use
The result from this study also suggested the significant and positive relationship between PEOU and PU. This result is
in line with previous studies on this link (Chau, 2001; Hong et al., 2002; Lallmahamood, 2007; Shim & Viswanathan,
2007; Ramayah & Aafaqi, 2004; Ramayah et al. 2004; Thong et al., 2004; Amin, 2007). This indicates that if students
find that the e-library is easy to use, they will be more willing to use it for information retrieval in order to improve the
quality of their assignments.
The finding of an insignificant relationship between PEOU and actual usage did concur with Ramayah and Aafaqis
(2004) and Igbaria and Iivaris (1995) insignificant finding. More work needs to be done in this area.
6.4 Perceived usefulness
PU was also found to be positively related to actual usage of the e-library. It implies that if students feel that a system is
useful, their usage level will be higher. This finding support prior research that has found a direct positive relationship
between PU and actual usage (Ramayah et al., 2004; Ramayah & Aafaqi, 2004; Davis, 1989; Mathieson, 1991; Adams
et al., 1992; Segars & Grover, 1993, Igbaria et al., 1995, 1997; Ndubisi et al., 2001; Ramayah, Ignatius, & Aafaqi, 2004;
Ramayah, Jantan, & Aafaqi, 2003).
References
Adams, D. A., Nelson, R. R, & Todd, P. A. (1992). Perceived usefulness, ease of use and usage of information
technology: A replication. MIS Quarterly, 6(2), 227-247.
Agarwal, R., Sambamurthy, V., & Stair, R. M. (2000). Research report: The evolving relationship between general and
specific computer self-efficacyAn empirical assessment. Information Systems Research, 11(4), 418-430.
Amin, H. (2007). Internet banking adoption among young intellectuals. Journal of Internet Banking and Commerce,
12(3), 1-13.
Archer, N. P., Head, M. M., & Yuan, Y. (1996). Patterns in information search for decision making: The effects of
information abstraction. International Journal of Human-Computer Studies, 45(5), 599-616.
Chau, P. Y. K. (2001). Influence of computer attitude and self-efficacy on IT usage behaviour. Journal of End User
Computing, 13(1), 26-33.
Chen, Y. N. (1999). Another type of overview of electronic libraries: Journal of Information, Communication in Library
Services, 5(3), 141-150.
Chen, C. C. (2000). Theory and practice of integrated searches at libraries. Wenhua, Taipei, 4-5.
Chen, C., Czerwinski, M., & & Macredie, R. (2000). Individual differences in virtual environmentsIntroduction and
overview. Journal of the American Society for Information Sciences, 51(6), 499-507.
Coakes, S. J., & Steed, L. G. (2003). SPSS analysis without anguish version 11 for Windows. Milton, QLD. John Wiley
and Sons, Australia.
Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS
Quarterly, 19(2), 189-211.
Davis, F. D. (1986). A technology acceptance model empirical testing new end-user information systems: Theory and
results. Doctoral Dissertation, Sloan School of Management, Massachucetts Institute of Technology, Cambridge, MA.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology,
Graduate School of Business Administration. MIS Quarterly, 13(3), 319-340.
Davies, C. (1997). Organizational influences on the university electronic library. Information Processing and
Management, 33(3), 377-392.
Deb, S., Kar, D. C., & Kumar, S. (2003). Setting up electronic library: The case of TERI. Proceeding of the ASIS&T
2003 Annual Meeting Humanizing Information Technology: From Ideas to Bias and Back, 19-22 October, California.
Dholakia, R. R., & Dholakia, N. (2004). Mobility and markets: Emerging outlines for m-commerce. Journal of Business
Research, 57(12), 1391-1396.
Dillon, A., & Watson, C. (1996). User analysis in HCIThe historical lessons from individual difference research.
International Journal of Human-Computer Studies, 45(6), 619-637.
81
Harrison, A. W., & Rainer, R. K. (1992). The influence of individual differences on skill in end-user computing. Journal
of Management Information Systems, 9(1), 93-111.
Hasan, B. (2006). Delineating the effects of general and system-specific computer self-efficacy beliefs on IS acceptance.
Information & Management, 43, 565-571.
Hong, W., Thong, J. Y. L., Wong, W. M., & Tam, K. Y. (2002). Determination of user acceptance of digital libraries: An
empirical examination of individual difference and system characteristic. Journal of Management Information System.
18(3), 97-124.
Igbaria, M., Iivari, J., & Maragahh, H. (1995). Why do individuals use computer technology: A finnish case study.
Journal of Information and Management, 29, 227-238.
Igbaria, M., Zinatelli, N., Cragg, D., & Cavaye, A. (1997). Personal computing acceptance factors in small firms: A
structural equation model. MIS Quarterly, 29(3), 278-305.
Igbaria, M., & Iivari, J. (1995). The effect of self-efficacy on computer usage. Omega. 23(6), 587-605.
Jacobson, T., & Fusani, D. (1992). Computer, system, and subject knowledge in novice searching of a full-text, multifile
database. Library and Information Science Research, 14(1), 97-106.
Ke, H. R. (2000). Discussion of experience in establishing on electronic library: The example of National Chiao Tang
University, Shu Fah Quarterly, 47, 10-33.
Koay, P. L. (2002). Receptiveness of e-banking by Malaysian consumers. Unpublished MBA Dissertation, School of
Management, Universiti Sains Malaysia, Penang, Malaysia.
Lallmahamood, M. (2007). An examination of individuals perceived security and privacy of the internet in Malaysia
and the influence of this on their intention to use e-commerce: Using an extension of the Technology Acceptance Model.
Journal of Internet Banking and Commerce, 12(3), 1-26.
Lu, J., Yu, C. S., Liu, C., & Yao, J. E. (2003). Technology acceptance model for wireless internet. Internet Research,
13(3), 206-222.
Marchionini, G., Dwiggins, S., Katz, A., & Lin, X. (1993). Information seeking in full-text end-user-oriented search
systems: The roles of domain and search expertise. Library and Information Sciences Research, 15(1). 35-69.
Mathieson, K. (1991). Predicting user intention: Comparing the Technology Acceptance Model with the Theory of
Planned Behavior, Information Systems Research. 12(3), 173-191.
Meadow, C. T., Wang, J., & Yuan, W. A. (1995). A study of user performance and attitude with information retrieval
interfaces. Journal of the American Society for Information Science, 46(7), 490-505.
Ndubisi, N. D., Jantan, M., & Richardson, S. (2001). Is the Technology Acceptance Model valid for entrepreneurs?
Model Testing and Examining Usage Determinants. Asian Academy of Management Journal, 6(2), 31-54.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory. McGraw-Hill, New York, NY.
Nysveen, H., Pedersen, P. E., & Thornbjomsen, H. (2005). Intentions to use mobile services: Antecedents and
cross-service comparisons. Journal of Academy of Marketing Science, 33(3). 330-346.
Ramayah, T. (2006). Doing e-research with e-library: Determinants of perceived ease of use of e-library. International
Journal of Technology, Knowledge and Society, 1(4), 71-82.
Ramayah, T. & Aafaqi, B. (2004). Role of self-efficacy in e-library usage among student of a public university in
Malaysia. Malaysia Journal of Library and Information Science, .9(1), 39-57.
Ramayah, T., Aafaqi, B., & Ignatius, J. (2004a). E-library usage among students of a public university in Malaysia. The
Proceedings of the International Conference on Management Education, 28
th
-29
th
June, Kuala Lumpur, Malaysia.
Ramayah, T., Ignatius, J. , & Aafaqi, B. (2004b). PC Usage among students: The case of a private institution of higher
learning. The proceedings of the International Conference on Management Education, 28-29
th
June, Kuala Lumpur,
Malaysia.
Ramayah, T., & Jantan, M. (2004). Internet usage among Malaysian students: The role of demographic and motivational
variables. PRANJANA: The Journal of Management Awareness, 7(2), 59-70.
Ramayah, T., Jantan, M. & Aafaqi, B. (2003). Internet usage among students of Institutions of Higher Learning: The
role of motivational variables. The proceedings of the 1
st
International Conference on Asian Academy of Applied
Business Conference, 10-12
th
July, Sabah, Malaysia.
Ramayah, T., Siron, R., Dahlan, N., & Mohammad, O. (2002). Technology usage among owner/managers of the SMEs:
The role of demographic and motivational variables. The proceedings of The 6
th
Annual Asian-Pacific Forum for Small
82
Business on Small and Medium Linkages, Networking and Clustering, 16-18
th
October, Kuala Lumpur, Malaysia.
Ramayah, T., Sarkawi, F., & Lam, S. C. (2003). Attitude and intention to use web-based supply chain management
(SCM) among SMEs. The proceedings of the Asia Pacific Seminar on E-Customer Relationship Management, 8-9
th
July,
Shah Alam, Selangor, Malaysia.
Sekaran, U. (2003). Research methods for business: A skill building approach. Singapore: Wiley& Son.
Segars, A. H., & Grover, V. (1993). Re-examining perceived ease of use and usefulness: A confirmatory factor analysis.
MIS Quarterly, 17(1), 517-725.
Shim, S. J., & & Viswanathan, V. (2007). User assessment of personal digital assistants used in pharmaceutical detailing:
System features, usefulness and ease of use. The Journal of Computer Information Systems, 48(1), 14-21.
Thompson, R. Higgins, C., & Howell, J. (1991). Personal computing: Toward a conceptual model of utilization. MIS
Quarterly, 15(1), 125-143.
Thong, J. Y. L., Hong, W., & Tam, K. Y. (2004). What leads to user acceptance of digital libraries? Communications of
the ACM, 47(11), 79-83.
Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into
the technology acceptance model. Information Systems Research, 11(4), 342-365.
Xie, H. (2006). Evaluation of digital libraries: Criteria and problems from users perspectives. Library & Information
Science Research, 28(3), 433-452.
Zmud, R. W. (1979). Individual differences and MIS success: A review of the empirical literature. Management Science,
25(10), 966-979.
Table 1. Descriptive statistics, reliability, and correlations
Cronbachs
Alpha
Mean S.D. 1 2 3 4
1. Computer
Self-Efficacy
.765 3.461 .461
2. Knowledge of
Search Domain
.776 3.231 .654 .146*
3. PEOU .846 3.394 .606 .243** .434**
4. PU .855 3.479 .614 .122 .283** .558**
5 Actual Usage .810 2.856 1.123 -.094 .236** .183** .224**
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Table 2. Results of regression
Variables
Perceived Ease of Use
(Std Beta)
Perceived Usefulness
(Std Beta)
Actual Usage
(Std Beta)
Computer Self-Efficacy .159**
Knowledge of Search
Domain
.459***
PEOU .570*** .085
PU .177**
R .258 .325 .055
R Change .258 .325 .055
Sig. F Change .000 .000 .004
Durbin Watson Index 1.963 1.883 1.997
Note: **p<.05, ***p<.01
83
Figure 1. Technology Acceptance Model (TAM)
Figure 2. Research framework for e-library usage
Computer Self-Efficacy
Knowledge of Search Domain
Actual Usage
Individual Difference
H1
H2
H3
H4
H5
Computer Self-Efficacy
Knowledge of Search Domain
Actual Usage
Individual Difference
H1
H2
H3
H4
H5
84
Research and Application of the 3D Virtual Community

Based on WEBVR and RIA
Meng Zhang, Zhihan Lv, Xiaolei Zhang, Ge Chen & Ke Zhang
College of Information Science and Engineer
Ocean University of China
Qingdao 266100, China
E-mail: shentian1984@163.com
The research is financed by the Innovation Fund of Medium or Small Science and Technology Enterprise of Science &
Technology Ministry (No. 07C26223101272). (Sponsoring information)
Abstract
Starting from the development of VR and WEB technologies, in this article, we analyze the characters and the
development foregrounds of the 3D virtual community based on the WEBVR (Web Virtual Reality) technology, and
introduce the characters of the RIA (Rich Internet Application) technology system. And we put forward the technical
frame of the 3D virtual community based on WEBVR and RIA, and introduce the information organization, storage,
interaction and other key technologies of the 3D virtual community based on WEBVR and RIA. Finally, taking the
Real-time Interactive 3D Simulation Web Community Platform as the example, we prove the extensive application
foreground of WEBVR and RIA in future web interaction.
Keywords: WEBVR, RIA, 3D virtual community, Value of new media
Virtual Reality (VR) is a sort of computer system which can create and experience the virtual world. The virtual world
is created by the computer, and it is the reappearance of the real world, and it can also be the world in the mind, and
users can naturally interact with the virtual world by virtue of many sensor channels such as vision, hearing and feeling.
It can create a 3D virtual world which can real-time reflect the changes of entity objects and interactions for users by the
mode of simulation, and offer users a 3D interface where users can interact with the virtual world and make users
directly participate in and explore in the function and changes of simulation objects in the environment and produce
immersion feelings by helmet-mounted display (HMD), data gloves and other assistant sensors.
With the development of VR and relative domains, only the analog and simulation to real scenes have not completely
fulfill users high-layer demands, and real-time and dynamic scene digital information acquisition based on web and
effective management, transmission and analysis computation are inspired as a sort of new demand. Under this
background, the combination of WEB3D and VR is the necessary tendency for the technical development.
WEBVR is the web virtual reality technology based on Internet which is realized depending on the software technology.
In 1994, Mark Pesce and Tony Parisi created the browser which was called as Labyrinth, and it was the initial shape of
3D browser on WWW. The game of Second Life developed by US Linden Lab perfectly combined VR and WEB3D,
i.e. in a 3D space, establishing virtual online world can copying various scenes, details and experiences of human living
to the Internet. After Google Company pushed the 3D map orientation technology based on satellite image diagrm,
Google Earth, it also pushed the online war game GEWar taking Google Earth as the platform. The new edition map
service issued by US Microsoft could simulate Virtual Earth 3D, offer the vivid simulation of real world, establish
real and sensitive construction model through pictures and make users fly and enjoy the diorama in the browser. The
implementation of web virtual reality based on WEBVR is the result to effectively integrate social media and
user-generated content.
1. Characters and development foreground of the virtual community technology based on WEBVR
In existing WEBVR virtual community, users browse through user website and log in the system user main interface
through web browser, browse through the classification content according to the main control menu, interact with
community activity according to system information, and produce interactions with other virtual people in the activity.
85
The webpage issuance mainly depends on WEB3D technology and WEBGIS technology, and the integration part
adopts the system frame idea of the design mode and implements data transmission by virtue of XML document, and
uses XHTML and CSS standardization representation, and uses DOM to realize dynamic display and interaction, and
uses XML and XSTL to implement data exchange and data processing. The system structure of data exchange is seen in
Figure 1.
With the quick change of Internet, Internet continually goes deep into human livings and WEB3.0 will largely change
the Internet form of human living. As same as WEB2.0, WEB3.0 is not the technical innovation, but the idea
innovation, and guides the development and application of the technology. WEB3.0 will create new virtual community
which will be divided not by the terrain and border, but by the interests, language, topic, occupation and specialty.
Everyone could create a new Internet kingdom and be the king, and be the president through democratic election in the
Internet kingdom, and at that time, you will possess web citizens from all over the world.
As the substitute of WEB2.0, WEB3.0 is still based on WEB2.0, and is the Internet mode which realizes more
intelligent communication among humans and between human and machine. Based on the grasp to WEB3.0 concept, in
this article, we think that future web community should be new digital technology based on web virtual reality
technology and intelligent information customization, and develop to the directions of subsection, specialty and
compatibility.
(1) The information in the website could directly interact with information in other websites, and could be integrated
and used in multiple websites through the third party information platform.
(2) Users possess their own data in the Internet and the data can not use it in other websites.
(3) WEB3.0 is completely based on WEB, and can implement the function that complex system program can only
possess by browser.
2. Structure and characters of RIA system
RIA is the next generation web application which combines interactive user experience of desk application with the
deployment flexibility and cost analysis of traditional web application. The rich client technology in RIA connects with
existing back-end application server through offering the running environment which can load compiled client port
application (transferring by HTTP and the form of file) and using asynchronous client/server frame by client
application, and it is a sort of safe and updating and service-oriented model with good adaptability, and it simplifies and
improves the user interaction of WEB application, and the developed application can offer more abundant user
experiences with more interaction and response. At the same time, it can realize more visible and effective application
service of user experience with more responses, and it can develop the application combining the strong interaction and
the abundant content of desk software with the extent of web together.
The concept of rich includes two aspects, i.e. the rich of data model and the rich of user interface. The rich in
data means that the user interface could display and operate more complex data model embedding into the client port,
and could operate the computation of the client port and asynchronously send and receive data. Relative to traditional
HTML page, the advantage of this mode is that the program runs at the client port and the program can interact with the
server few when it interacts with users more. The complex data model balancing client port and server port can give you
larger space to establish more effective and interactive web application. The rich can also describe the completely
improved user interface, and HTML only offers users very limited interface control elements, and the user interface of
RIA offers flexible interface control elements which can better combine with data model. Traditional Internet model
uses the linear design and offers users some choices, and then users send choice results to the server, and this single
mode doesnt accord with the flexible and interactive requirement of application and users wills. Frequent server
requests and page fresh have many deficiencies such as slow page opening speed and reduced network bandwidth. If the
rich client interface is adopted, the part of application only receiving the requests could make corresponding change, but
the former server response would influence the whole interface. That means the interface will be decomposed into many
independent models which will make responding reactions to received information, and some of them will interact with
the server port, and some of them are communications among these models.
The applied character of RIA includes following aspects.
(1) Strong interaction. RIA offers abundant UI components, strengthens localization and uses distribution to manage the
components. Label guide, collapse column and tree catalog have approached to the development modes of AWT and
SWING.
(2) Direct management. Using RIA, user and interface interaction only need the part needing to be disposed. The
request/return model is not necessary for all user interface actions, and the computation of the client port can directly
realize the responses of users requests.
(3) Multiple-approach processing. RIA allows development personnel to put all contents in one interface and add
86
conversion and effects, and makes users more easily know the context and work flow of the application. It can mainly
deal with certain approach and easily be moved among approaches in the application.
(4) Client caching. Under the environment of RIA, data can be cached in the client port, and the user interface with
quick response speed and small times that data come and go the server can be realized, and the EMS memory
consumption of the server can be reduced, and the resource allocation will be more reasonable.
(5) Text independence. RIA text uses a sort of opening vector graph description language based on XML, so the text can
be searched by the web search engine or be inquired and compiled by the user browser.
(6) Platform independence. The language style is consistent for all client ports, so the code should not be rewrote for
different applications.
3. The virtual community frame based on WEBVR and RIA
RIA offers a strong technical platform which makes the computation ability of the client port can very approach to the
desk software system or the ability of the client computer in traditional C/S system, and it is fit for the traditional
N-layer development process. And it can also be integrated with residual environment to develop existing application
but not implement large numbers of modification. The virtual community frame based on WEBVR and RIA put forward
in the article is seen in Figure 2.
Under the system structure of WEBVR, the former so-called concept of single data center has not existed, and the
model and geography information may be distributed in any nodes of the web, so it is urgent to make users find and use
needed data in tremendous amount of Information Ocean. In addition, 3D web virtual community possesses large
numbers of data with different formats, and it is a very important problem to effectively manage these data and realize
information sharing. Facing this actuality, the space data storage, transmission and representation under the environment
of RIA can obtain satisfactory results. Because RIA server text uses a sort of opening vector graph description language
based on XML, so it can not only be used in the exact expression of data, but offer effective method of data
communication among asynchronous information systems. And most present browsers all support FlashPlayer, so the
client port needs not download any data processing components. For numerous non-professional users, they can
conveniently organize and issue their own space information resource through various universal RIA visualization
creation and editing tools and make WEBVR resources and other web resources can be shared conveniently by the
whole society and fully exert their own values. On the other hand, WEBVR based on RIA can make the map in front of
the client port possess more abundant interface expressive force and support complex image interaction, and make
WERVR become into real interactive system.
Furthermore, the technical characters of RIA also decide that it is fit for the large scale data-oriented application.
WEBVR is developed in the process converting from system-orientation to data-orientation, so whether for practical
meaning or for theoretical meaning, RIA has very important meaning for WEBVR.
4. The key technology of virtual community based on WEBVR and RIA
The 3D virtual community platform based on WEBVR and RIA adopts the bottom integration frame including
WEBVR, RIA and WEBGIS, and integrates web foregrounding platform module, background 3D scene map module,
virtual community construction module and compressive system integration module. Its total frame uses the idea of
design mode for reference, harmonizes the complexity of time and space, and enhances the natures of re-plantation and
extension. For the concrete implementation of the technology, the platform uses the technology of automatic modeling
which can automatically identify and acquire data and enhance the production efficiency. The platform also could
establish the multidimensional data pyramid model, classify and compress the multidimensional data of various kinds of
scene into corresponding data layers. For the data transmission between the client port and the server data, the platform
can optimize the transmission through the transfer algorithm and implement parallel transmission of multithreading.
And the platform could realize seamless integration and display multidimensional data by using the 3D panorama
imaging technology and the 3D model imaging engine technology through the expressive mode of 3D reality
simulation.
4.1 Seamless combination and low delay load based on browser non-insert multidimensional data
For the vector data, the platform establishes a set of vector data real-time reading, converting and rendering system
from the bottom layer, and it can translate the vector data into the data format supported by the browser for
transmission, i.e. the server port directly reads the vector data, and plots them to supportable format and sends them to
the client port. And then the platform connects the data produced from the foreground vector data resource with 3D data
resource. The platform uses exact mathematical algorithm and VR theory, considers the link compactness and the strain
of the effect representation, and finally achieve the effect of seamless link through multiple times operation, which can
not only inosculate with the reality but represent the art property.
After opening the browser, the system adopts pre-load technology with classified batch and parallel reading to the
87
multidimensional data. In the 3D space data transmission process, the single task program execution efficiency is low,
and if blocks happen when program receives data input, the program can keep on running after the program obtains
data. In the web environment, because the system usually receives and accepts data, if the blocks happen in the process
of receiving data, the program will be in the waiting state and can not implement any operation, and this web
application can not let users implement normal web interaction operations. The 3D virtual community platform based
on WEBVR and RIA can overcome these problems, and it can execute multiple relatively independent threads
simultaneously, establish the thread which could be used in 3D space data input, the thread which could be used in data
output and the thread which could be used for data processing at the background. If the input data thread is blocked
when receiving data, the output data thread and the data processing thread can still run. The design of multithreading 3D
space data web transmission would largely enhance the acquisition execution efficiency and the parallel processing
ability of 3D space data linking multidimensional data server through the web, and it is specially fit for the web
environment and multi-computer environment.
4.2 Real-time interaction between server and client 3D information
Utilize asynchronous document transmission technology to implement interaction between the operation layer and the
data layer, encapsulate the 3D information and the geographical data coding and transfer to the data layer, and through
the query computation of the data layer, and the final result will be feed back to the operation layer and implement
reverse coding. Implement web transmission taking XML document as the carrier to response users operations real
time.
4.3 The combination of Web community and 3D VR technology
Use Web3D to realize the VR display on the web, establish a 3D scene, and users can run in the scene from the first
visual angle. The interaction created between the scene and the controllers and the HD pictures will make users be
personally on the scene. In the 3D web virtual community, everyone will integrate his ID and behaviors in different web
activity spaces, and map a complete self in the virtual space, and everyone will have a his own multidimensional image,
and users can change his dress and body at will. Based on that, for the design of the client port, adopt RIA technology,
possibly offer IM, Blog, page customization, information customization and usual software interface, and make users
usage more convenient and fulfill individual demands.
4.4 The automatic 3D modeling technology through identifying and simulating acquired data
Based on the theory of pattern recognition, automatically classify the acquired data, and divide them into segment grain
data and overall arrangement image data, utilize multi-angle and multi-orientation acquired overall arrangement image
data to establish the outline models of various architectures in the scene and put up the position relation of the model
group. Correspond the segment grain data with the architecture surface at the corresponding position, and implement
pastes after optimized processing. This new technology could make the 3D model be quickly and automatically
simulated and established.
5. Implementation and expectation
In this article, we select the real-time interactive 3D simulation web community platform as the applied case. And the
platform possesses following five characters for the interactive experience.
(1) Possessing abundant interface expressive force and quick interactive speed.
(2) Offering multiple information representation and 3D scene visualization function.
(3) The information loading speed is quick and users waiting time is short.
(4) Click the measurement point on the WEB-3D Map, the detail geographical information can be displayed. And in the
information display column, HD pictures, 360 panorama graph and VR scene are especially fit for issuing to the
public.
(5) Realizing the connection with other governmental information databases, and users can directly interact with charts
and acquire deeper information.
The development tendencies such as specialty, segment and compatibility of the future web virtual community decide
the distributed Internet. The occurrence of RIA offers a sort of effective mechanism about information display,
organization, sharing and utilization for WEBVR. As a new tool, RIA has many parts lacking in clear definitions and
perfections to be developed further. We will continue to deeply study the technology of RIA. We hope the research of
RIA can guide the development and practice of WEBVR community, and further enhance the computation ability of the
client port, and fulfill human more and more intensive individuality demand. Otherwise, the enhancement of function
also induces many hidden safe troubles which should be improved in our future researches.
References
Bishop I D, Gimblett H R. (2000). Management of Recreational Areas: Geographic Information Systems, Autonomous
88
Agents and Virtual Reality. Environment and Planning B: Planning and Design. No.27(3). P.423-435.
Counsell J. (1998). 3D Built Form and Landscape from 2D Maps. Habitat. No.6(5). P.41-43.
Huang B, Lin H. Geo. (1999). VR: a Web-based Tool for Virtual Reality Presentation from 2D GIS Data. Computers &
Geosciences. No.25(10). P.1167-1175.
Luo, Yingwei, Wang, Xiaolin & Xu, Zhuoqun et at. (2004). Design and Implementation of a Hierarchical
Component-Based WebGIS. Chinese Journal of Computers. No.27(3). P.176-185.
Luo, Yingwei, Wang, Xiaolin & Xu, Zhuoqun. (2005). XML-Based Geospatial Metadata System. Chinese Journal of
Computers. No.28(7). P.1205-1212.
Takase Y, Sho N, Sone A, et al. (2003). Automatic Generation of 3-D City Models and Related Applications. Proc. of
International Workshop on Visualization and Animation of Reality-based 3D Models. Switzerland. No.2.
Takase Y, Sone A, Hatanaka T. (2004). A Development of 3D Urban Information System on Web. Proc. of Processing
and Visualization Using High-Resolution Images. Thailand.
Whyte J, Bouchlaghem N, Thorpe A. (2000). From CAD to Virtual Reality: Modeling Approaches, Data Exchange and
Interactive 3D Building Design Tools. Automation in Construction. No.10(1). P.43-55.
Winter bottom S J, Long D. (2006). From Abstract Digital Models to Rich Virtual Environments: Landscape Contexts
in Kilmartin Glen, Scotland. Journal of Archaeological Science. No.33(10). P.1256-1367.
Yang, Chongjun, Wang, Yuxiang & Wang, Xingling et al. (2001). Review of the Main Technologies of WebGIS.
Journal of Image and Graphics. No.6(9). P.886-894.
Zhang, Hong & Feng, Jiangfan et al. (2007). Research of WebGIS Based upon RIA. Geo-Information Science. No.9(4).
P.37-41.
Zhou, Wensheng, Maofeng & Hupeng. (2004). Study of Open WebGIS Based on XML. Journal of Image and
Graphics. No.29(10). P.1258-1264.
Zhou, Wensheng. (2002). Research of WebGIS Based on SVG. Journal of Image and Graphics. No.7(7). P.693-698.
Figure 1. The System Structure of the Network VR Based on Web-3D
Figure 2. The

e Virtual Comm

munity Frame

Based on WE

EBVR and RIA
Feb
A
bruary, 2009
89
90
Framework for Conference Management System

Abu Bakar Md Sultan, Abdul Azim Abdul Ghani & Lim Su Fang
Department of Information System
Universiti Putra Malaysia
43400 UPM Serdang, Malaysia
E-mail: abakar@fsktm.upm.edu.my, azim@fsktm.upm.edu.my
This research is sponsored by Research University Grant Scheme RUGS: 05-02-07-0346RU from Universiti Putra
Malaysia
Abstract
This paper presents a Web-based Conference Management System framework that is intended to support multiple
conferences and facilitate the conference management starting from the conference preparation until the attendance
tracking process during conference day. The framework derived from the analysis of several web-based related
applications. The framework is presents and the results are discussed.
Keywords: Conference Management, Web Application, System Framework
1. Introduction
Conference is a meeting for researchers to present, discuss, identify important problems and formulate future research
direction on specific themes. It provides a true learning environment. However, organizing a conference entails plenty
of time-consuming work. For instance, collecting paper submission from authors, assigning paper submitted to the
reviewers, notifying authors about whether the paper submitted is rejected or accepted. These processes will surely
consume a lot of precious time and complicated works if a conference is conducted manually. Unfortunately, there are
no reports from literatures that are seriously looking in how to produce an efficient system in managing the entire
process of conferences.
Lately, the need of systems for collaboration support has expanded, leading to their growing application in
organizational, communication, and cooperation processes. Hence, various applications have been developed for
conference organization to support or manage the workloads. The emergence of information communication technology
has changed the way the conferences suppose to organize. Thus this paper analyzes several online conferences and
proposes a framework for conference management system.
This paper presents a web-based Conference Management System framework that is intended to support multiple
conferences and facilitate the conference management starting from the conference preparation until the attendance
tracking process during conference day.
The rest of paper is organized as follows. In section 2 the analysis of features of current system and the solution are
discussed. An overview of a proposed solution is discussed in section 3. Section 4 includes the result and discussion.
Section 5 includes the conclusion and future suggestion.
2. Analysis of Existing Conference System
A Conference Management System shall include functions such as the online participant registration, paper submission,
paper reviewing, and notification process and so on. Through the study of the Conference Management System
available in the market, we study four most popular solutions for conference system which is ConfMan, CyberChair,
Confious and Chairman. The study revealed some differences between the systems. Details of the analysis are listed in
the table below:
The result from the Table 1 shows that each product has own advantages and disadvantages. Most of them are still
lacking throughout the entire conference management process.
As far as we are concerned, there is no research to produce a common framework for any Conference Management
System. Each system only developed when the conference is going to organize. The system developed not only includes
the main function of a conference but also to propose a common framework that allows the conference organizer to
91
create and customize the conference website. Hence, the main objective of the system is to reduce the effort and in the
meantime increase the efficiency and productivity in managing a conference.
3. Proposed Framework.
The old fashioned ways in organizing a conference will costs and consume a lot of precious time and effort. For this
reason, technical support tools and components are integrated to cope with the problems.
Basically, the system developed consists of three layers, Conference Management Dashboard, Conference Management
Platform (operation), and the database layer. There are five stakeholders for the system, namely the system
administrator, conference administrator, participant, presenter and reviewer. The Conference Management Dashboard is
managed by System Administrator to monitor the performance/status of all registered conference in the portal. The
operation of the Conference Management System consists of several modules which are the participant registration,
paper management, paper review manager, messaging module.
3.1 Paper Submission Process
For the paper submission process, apache common fileupload component is used to ease the process. This uploading
tool allows the presenters to submit their papers electronically with just a few mouse clicks. Apart from that, the papers
uploaded by presenter will be kept in a file directory that facilitate the paper management and reduce the possibility of
paper lost. A search engine is built-in in order to reduce the time taken in the searching paper or document process.
3.2 Notification Process
Conference Administrator uses the messaging module to notify the presenters about the papers acceptance decision.
Other than that, the messaging module will be integrated with SMS to provide e-mail to short messaging. This SMS
function is to ensure that the participants and reviewers are always updated with the conference information.
3.3 Paper Assignment Process
Paper assignment is proposed to assign the papers submitted to reviewers.
3.4 Paper Reviewing Process
The system developed allows reviewers to download and grade the papers assigned to them. Besides that, multiple files
download function also included to assist Conference Administrator in downloading all the papers submitted.
3.5 Conference Customization
Content Management System (CMS) is integrated in order to assist the conference organizer to customize the
conference website. This CMS allow the conference organizer to manage the creation, modification, and removal of
content from a Web site without needing the expertise of a Webmaster.
4. Result and Discussion
From the stimulation, paper submission, and paper assigning process can be done within 5 seconds. Other than that,
papers or documents are well-managed and errors caused by human are reduced.
5. Conclusion and Future Suggestion
The system developed is expected to reduce the cost, effort and time in managing a conference. It also promotes
paperless and environment friendly by reducing the paperwork involved in a conference. Our future effort is to propose
autonomous Conference Management System with the aims further reducing resources in managing conferences.
References
http://searchsoa.techtarget.com/sDefinition/0,,sid26_gci508916,00.html
http://www.easychair.org/
http://www.iec.org/online/tutorials/wire_sms/topic04.html
P. Halvorsen, K. Lund, V. Goebel, T. Plagemann, T Preuss & H. Hartmut, ConfMan: Integrated WWW and DBS
Support for Conference Organization, Poster Proceedings of The Eighth International World Wide Web Conference
(WWW'8), Toronto, Canada, May 1999, pp. 34 35.
V. Levacic, H. Bogunovic, S. Loncaric, B.Kuzmic, & B. Blazona: "Chairman: A Confrence Management System
", 28
th
International Convention MIPRO, Opatija, Croatia, May 30 - June 3, 2005
92
Table 1. Comparison of existing Conference Management System
Function ConfMan
[3]
CyberChair
[5]
Confious
[1]
Chairman [4]
Participant
Registration
Y Y
Paper
Submission
Y Y Y Y
Paper
Assignment
Y Y Y
Bidding
Process
Y
Paper
Downloading
Y Y
Paper
Reviewing
Y Y Y Y
Notification
Process
Y Y Y
Online PC
meeting
Y
Setup Phase Y Y
Table 2. System Framework
Conference Management Dashboard
P
o
r
t
a
l

R
e
g
i
s
t
r
a
t
i
o
n
,

C
u
s
t
o
m
i
z
a
t
i
o
n

a
n
d

P
u
b
l
i
s
h
i
n
g
P
a
r
t
i
c
i
p
a
n
t

R
e
g
i
s
t
r
a
t
i
o
n

P
a
p
e
r

M
a
n
a
g
e
m
e
n
t

P
a
p
e
r

R
e
v
i
e
w

M
a
n
a
g
e
r

I
n
t
e
g
r
a
t
e
d

M
e
s
s
a
g
i
n
g

S
y
s
t
e
m
P
r
o
g
r
a
m

S
c
h
e
d
u
l
e
r

F
i
n
a
n
c
e

S
y
s
t
e
m

A
t
t
e
n
d
a
n
c
e

S
y
s
t
e
m

Centralized Database
93
Figure 1. Screen Snapshot of Paper Submission Process
Figure 2. Screen Snapshot Notification Process
Figure 3. Screen Snapshot Paper Assigning Process
94
Figure 4. Screen Snapshot of Paper Reviewing Process
Figure 5. Screen Snapshot Customization Process
95
Design of the Closed Loop Speed Control System for DC Motor

Guoliang Zhong
College of Mechanical and Automotive Engineering
South China University of Technology
Guangzhou 510640, China
E-mail: zhgl2chl@126.com
College of Mechanical and Automotive Engineering
South China University of Technology
Guangzhou 510640, China
Liangzhong Jiang
E-mail: jianglzh88@163.com
The research is financed by the Guangzhou Municipal Science and Technology Key Projects (No. 2004A10403006).
(Sponsoring information)
Abstract
This article introduces the speed control principle of DC motor, expatiates on the speed control system taking
PIC16F877 SCM as the main control component, utilizes the characters of catching module, comparing module and
analog-to-digital conversion module in PIC16F877 SCM to be the trigger circuit, and gives the program flow chart. The
system has many advantages including simple structure, synchronization with the main circuit, stable shifting phase and
enough shifting phase range, the control angle of 10000 steps, stepless smooth control of motor, steep pulse front edge,
enough amplitude value, setting pulse width, good stability and anti-jamming, and cheap cost, and this speed control
system with good practical values can be realized easily.
Keywords: PIC16F877, DC motor speed control, Control circuit, PI control algorithm
1. Introduction
The quick development of electric technology makes the speed control of DC motor gradually translate from analog to
digital, and at present, the KZ-D towage system which extensively adopts the thyristor equipment (i.e. silicon controlled
thyristor, SCR) in the electrical towage control systems to supply power to electromotor has replaced cumbersome F-D
system of generator-electromotor, and especially the application of SCM technology makes the speed control
technology of DC motor enter into a new phase. In the DC governor system, there are many sorts of control circuit.
SCM has many advantages such as high performance, quick speed, small volume, cheap price and reliable stability,
extensive application and strong currency, and it can increase the ability of control and fulfill the requirement of
real-time control (quick reaction). The control circuit adopting analogy or digital circuit can be implemented by SCM.
In this article, we will introduce a sort of DC motor speed control system based on SCM PIC16F877.
2. Speed regulation principle of DC motor
In Figure 1, the armature voltage is
a
U
, the armature current is
a
I
, the total resistance of armature loop is
a
R
, the
motor constant is
a
C
, and the excitation flux is u. According to KVL equation, the rotate speed of the motor is
96
( ) ( ) ( ) ( ) ( )
0 1
0
1
60
( ) 1 1 ( 1) 0.84 0.63 1
1
a a a
a
a
a a a a
d d d
I
i
U I R
n
C
pN
pN
C
a
U I R U
T k T k a e k a e k T k e k e k
T
a Kp
T
a Kp
Tf T Td
=
u
=
~
= + = +
| |
= +
|
\ .
=
= +
Where, p is the pole-pairs, N is the number of turns. For the motor that the armature spur track number is a, the
motor constant
60
a
pN
C
a
=
, that means when the motor is confirmed, the value is fixed. But in
a a a
U I R
, because
a
R
is
only the winding resistance, so
a a
I R
is very small, and
a a a a
U I R U ~ . So it is obvious that when we change the
armature voltage, the rotate speed n changes with that.
3. System composition and work principle
3.1 Module frame of system hardware structure
The module block diagram of system hardware is seen in Figure 2.
3.2 Work principle of the system
The system is mainly composed by master switch, motor exciting circuit, thyristor speed control circuit (including
tachometer circuit), rectifying filter circuit, flat wave reactor and discharge circuit, energy consumption braking circuit,
and the system adopts the closed loop PI regulator to implement control. When the master switch closes, the
single-phase AC obtains continual current with small pulse through the control of thyristor speed control circuit, and
bridge rectifier, filtering and flat wave reactor for the motor, and at the same time, through the rectifying of exciting
circuit, AC makes the motor obtain excitation to begin to work. The speed in the regulation trigger circuit sets the
potentiometer RP1 to make the control angle output by PIC16F877 decrease when AN1 input voltage decreases, and the
flow angle of thyristor increases with that, and the output voltage of main circuit increase, and the motor speed
increases, and the output voltage of tachometer circuit increases, and the motor stably runs in the setting speed range
through the function of PI regulator.
4. Circuit designs of various parts in the system
4.1 Design of main circuit
The parameters of various components in the circuit are seen in Figure 3.
Press the start-up button SW, electrify the contactor KM loop, and KM normally open contact closes, and the normally
closed contact opens, and start the button self-lock, the main circuit connects, and the thyristor speed control circuit
controls the AC output through changing the control angle of bidirectional thyristor, and obtains the DC through bridge
rectifying and filtering, and at the same time, the motor obtains the excitation through rectifying of exciting circuit to
begin to work.
To limit the pulsation of DC, connect the flat wave reactor in the circuit, and the resistance R3 offers discharge loop for
the flat wave reactor when the master circuit suddenly breaks off.
To quicken braking and stopping, the energy consumption braking is adopted in the equipment, and the braking part is
composed by the resistance R4 and master circuit contactor normally closed contact.
The motor excitation is powered by the single rectifying circuit, and to prevent the uncontrollable high speed accident
induced by the excitation loss of motor, in the exciting circuit, serially connect the under-current relay KA, and the
action current can be regulated through the potentiometer RP.
4.2 Design of thyristor trigger circuit
The voltages at the point A and the point B in the main circuit change to 20V through the transformer, and after bridge
rectifying, the half signals about 100H occur at these two points, and the signals meet with NPN audion through
voltage-division R6 and R7 to amplify, produce zero passage pulse at the collector of the audion, and catch the zero
passage pulse ascending edge by CCP1 module and note the time of occurrence first, and catch the zero passage pulse
descending edge, and the time difference between both is the zero passage pulse width, and the half of the value is the
midpoint of pulse, and by this catching mode, we can exactly obtain the actual zero passage point of AC, and at the
97
same time, we can utilize ADC analog/digital conversion module to translate the simulation voltage of PIC16F877 pin
RA1/AN1 as the setting value of thyristor control angle (setting value of motor speed), change the setting value of
potentiometer RP1 and correspondingly change the setting value of thyristor control angle, and the output value of
tachometer circuit is input by the pin RA1/AN1 of PIC16F877, and the value is taken as the speed feedback value
through A/D conversion. The oscillation frequency of SCM in the system adopts 4MHz, and according to the character
of PIC16F877 order period, the resolution of thyristor control angle is the reciprocal of one fourth of SCM oscillation
frequency, i.e. 1s, for the half wave time of 10ms of the power, the control angle can achieve 10000 steps, which can
completely realize the stepless smooth control of motor.
4.3 Design of tachometer circuit
The tachometer circuit is composed by the optical coded disc accreting with motor rotor and the electric pulse
amplifying and shaping circuit. The frequency of electric pulse has fixed proportion with the rotate speed of motor, and
through amplifying and shaping, the electric pulse output by the optical coded disc is input from the pins RC0/T1CK1
of PIC16F877 as the standard TTL level, count by the TMR1 counter to compute the rotate speed, and compare this
rotate speed with the presetting rotate speed and obtain the difference value, and PIC16F877 implements PI operation to
this difference value to obtain the control increase, and send the thyristor control angle in CCP2 to change the effective
voltage of two ports of the motor, and finally control the rotate speed.
5. Software design
To obtain small super modulation of thyristor control angle, we design the speed closed control as the typical I system,
i.e. PI regulator, which is used to regulate the thyristor control angle time
d
T , and its control algorithm is
( ) ( ) ( ) ( ) ( )
0 1
0
( ) 1 1 ( 1) 0.84 0.63 1
1
d d d
I
i
T k T k a e k a e k T k e k e k
T
a Kp
T
= + = +
| |
= +
|
\ .
Where,
1
a Kp =
, Kp is the proportion coefficient of controller,
I
T
is the integral time constant, and
i
T
is the
sampling cycle.
Considering that the motor time constant of electromotor in the system is 0.12s, the warps couldnt be eliminated in
several sampling cycles, so we select 2ms as the tachometer sampling cycle in the system.
The software design module in the system mainly includes CCP1 ascending catching module, CCP1 descending
catching module, control angle setting value A/D conversion module, tachometer circuit pulse timing counting module,
PI regulator module and CCP2 comparison output module, and the program flow charts are seen in Figure 5 and Figure
6.
Suppose we obtain the zero-pass time T , and the thyristor control angle time is
d
T , so the comparison value which is
sent into CCP2 register CCPR2H:L is Tf T Td = + , and when the comparison is consistent, output high level in the pin
of CCP2 to make the thyristor connect, and modify the value of CCPR2H:L again according to the required trigger
pulse width value to sustain the output high level trigger pulse for certain time and return to low level again, so a
bidirectional thyristor trigger pulse output is completed.
6. Conclusions
The speed control system taking PIC16F877 SCM as the bidirectional thyristor trigger circuit designed in the article
possesses many characters such as simple structure, reliable running, wide regulation range, good current continuity and
quick response in the middle and small sized DC motor speed control system, and the rotate speed loop adopts PI
control algorithm, which can effectively restrain the super modulation of rotate speed, so it is a feasible design to adopt
the speed control system of SCM, and the running curve is seen in Figure 7.
References
Kang, Huaguang & Zou, Shoubin. (2000). Base of Electric Technology: Number Part (4th Edition). Beijing: Higher
Education Press.
Li, Xuehai. (2002). Practical Tutorial of PIC SCM. Beijing: Beijing University of Aeronautics and Astronautics Press.
Su, Jianhui. (2002). Study on the Photovoltaic Water Pump System and Its Control. Thesis of Doctors Degree of Hefei
University of Technology.
Zhang, Shen. (1996). Principle and Application of DC Brushless Motor. Beijing: Machine Industry Press.
Vol
98
l. 2, No. 1
Ma
sw
cir

aster
witch
rcuit
Excitin
circuit
motor

Figure
Figure 2. Blo
Thyristo
trigger cir
ng
of
r

e 1. Principle D
ck Diagram of
or
rcuit
Rec
filte

Diagram of DC
f Hardware Str
ctifying
er circuit
F
Co
C Motor
ructure Module
co
Flat wave react
and discharge
circuit
omputer and In
e
Energy
onsumption
braking
circuit
tor
e
nformation Sci
M
ience
Figure 3.
Figure 4.

Diagram of M
Thyristor Trig

Main Circuit
gger Circuit
Febbruary, 2009
99
100
Figure 5. Interrupt Program Flow Chart of CPP1 Module and CPP2 Module
Interrupt
entrance
Protect interrupt locale
W, STATUS, PCLATH
Clear CCP1 interrupt
symbol CCP1IF
(Pulse width value+
CCP2 value) as the
new value of CCP2
Set CCP2 comparison
output as the low
level mode
Whether CCP1 is
the ascending
edge catching
mode?
Set the next catching
of CCP1 as
descending edge
mode
Store the ascending
edge time caught by
CCP1
Compute CCP1
(ascending time+ the
difference between
zero-pass time with
ascending edge time) as
the zero-pass time T0
Compute
T0+Td as the Tf
of CCP2
Set CCP2
comparison
output as the
high level
mode
Recover interrupt locale W, STATUS,
PCLATH, interrupt returns
Set next
catching of
CCP1 as the
descending
edge mode
Compute
CCP1
((descending
edge time
ascending
time)/2) as
the difference
between
zero-pass
time with
ascending
edge time
Yes
No
No
Yes
No
For
CCP1interrupt
symbol
For CCP2
interrupt symbol
Clear CCP2 interrupt
symbol CCP2IF
Whether output
high level mode
for CCP2 mode?
101
Figure 6. Flow Chart of Main Program
Start
Set CCP1pin as input,
set CCP2 pin as output
Allow CCP1to catch
interrupt, CCP2
compares interrupt
Preset CCP1catching as
the ascending edge
mode
Set ADCON1, ANI pin
as input of analog signal
Set ADCON0, set self RC
oscillation, preset channel
AD module to start up
Interrupt peripheral
function module
Open entire interrupt
Transfer A/D
conversion subprogram
and save result to Vr
Transfer tachometer
circuit pulse timing
counter subprogram
and save result to Vr
Transfer PI regulator
subprogram and take
the result as the
control time Td
Vol
102
l. 2, No. 1
Figure 7. Ru

unning Curve
Co omputer and Innformation Sciience

Computer and Information Science

Uploaded by

Computer and Information Science

Uploaded by

Computer and Information Science February, 2009

A Framework for Selection of Information Systems Development

Model Checking the Inconsistency and Circularity

End-User Computing Satisfaction (EUCS)

Layer Model and Algorithm of Organization Authorization

Chi-Square Test for Anomaly Detection in XML Documents

Computer and Information Science February, 2009

Research of Sales Contract Management System Based on WEB

Complexity Analysis in Heterogeneous System

Pixel Based Temporal Analysis Using Chromatic Property

' ' = >

Fault Recovery Mechanisms in Utility Accrual

The Fusion of Remote Sensing Images Based on Lifting

Individual Differences, Perceived Ease of Use, and Perceived Usefulness

Research and Application of the 3D Virtual Community

Framework for Conference Management System

Design of the Closed Loop Speed Control System for DC Motor

You might also like