Agent Technology for e-Commerce
Chapter 6: Recommender Systems
Maria Fasli
[Link]
Agent Technology for e-Commerce
Recommender systems: The problem
Too much information: information overload – consumers have
too many options
A recommender system is a system which provides
recommendations to a user
Applications: Books, music CDs, movies. Even documents,
services and other products such as software games
Chapter 6 2
Agent Technology for e-Commerce
Recommender
Requester Recommender
Request for service
Sorted description
of P1,..Pn
Advertisement
of capabilities
Service delegation
Results of service request
Provider 1 Provider n
Chapter 6 3
Agent Technology for e-Commerce
Information needed
Information used for recommendations can come from different
sources:
browsing and searching data
purchase data
feedback explicitly provided by the users
textual comments
expert recommendations
demographic data
Chapter 6 4
Agent Technology for e-Commerce
Providing recommendations
Recommendations can take the following forms:
Attribute-based recommendations: based on syntactic attributes
of products (e.g. science fiction books)
Item-to-item correlation (as in shopping basket
recommendations)
User-to-user correlation (finding users with similar tastes)
Non-personalized recommendations (as in traditional stores, i.e.
dish of the day, generic book recommendations etc.)
Chapter 6 5
Agent Technology for e-Commerce
Recommendation technologies
Information retrieval (IR) systems:
allow users to express queries to retrieve information relevant to
a topic of interest or fulfil an information need
they are not useful in the actual recommendation process
they cannot capture any information about the users’ preferences
they cannot retrieve documents based on opinions or quality as
they are text-based
To address these issues two techniques have been developed:
Content-based filtering (Information filtering)
Collaborative-based filtering
Chapter 6 6
Agent Technology for e-Commerce
Content-based filtering
The system processes information from various sources and tries to
extract useful elements about its content
keyword-based search (keywords sometimes in boolean form)
semantic-information extraction by using associative networks of
keywords, or directed graphs of words
Chapter 6 7
Agent Technology for e-Commerce
Each user is assumed to act independently and the system
requires a profile of the user’s needs or preferences
The user has to provide information on her personal interests on
starting to use the system for the profile to be built
The profile includes information about the items of interest, i.e.
movies, books, CDs etc.
Content-based filtering techniques try to identify similar items
which are returned as recommendations
They do not depend on having other users in the system
Chapter 6 8
Agent Technology for e-Commerce
Issues
Pure content-based filtering systems are not capable of exploring
new items and topics
Over-specialization: one is restricted in viewing similar items
Difficult to apply in situations where the desirability of an item is
determined in part by aesthetic qualities that are difficult to
quantity – it is difficult to apply content-based analysis to such
items
Chapter 6 9
Agent Technology for e-Commerce
The user profiles
For the system to produce accurate recommendations, the user
has to provide constant feedback on the returned suggestions –
users do not like providing feedback
Consist entirely of ratings of items and topics of interest: the
fewer the ratings, the more limited the set of possible
recommendations
As the user’s interests change, these changes need to be tracked
Chapter 6 10
Agent Technology for e-Commerce
Collaborative filtering
Collaborative-based filtering systems can produce
recommendations by computing the similarity between a user’s
preferences and the preferences of other people
Such systems do not attempt to analyse or understand the content
of the items being recommended
They are able to suggest new items to user who have similar
preferences with others
Chapter 6 11
Agent Technology for e-Commerce
Basic mechanism
A large group of people’s preferences are registered
A subgroup of people is located whose preferences are similar of
the user who seeks the recommendation
An average of the preferences for that group is calculated
The resulting preference function is used to recommend options
to the user who seeks the recommendation
The concept of similarity needs to be defined in some way
Chapter 6 12
Agent Technology for e-Commerce
Example user-item matrix
What would be the recommendation for user D?
Chapter 6 13
Agent Technology for e-Commerce
Neighbourhood-based algorithms
Three steps
(i) The degree of similarity of the active user and the others in
the database is calculated (positive or negative)
(ii) A set of users is chosen as the basis for making the
prediction. This is determined based on the degree of similarity
and differs from system to system
(iii) The set of users chosen in the previous step is used to make
the recommendation. A user with high degree of similarity may
be assigned higher weight
Chapter 6 14
Agent Technology for e-Commerce
Pearson’s correlation coefficients
It reflects the degree of linear relationship between two variables
and ranges –1 to +1
The degree of correlation between an active user a and another
user u is:
Chapter 6 15
Agent Technology for e-Commerce
Next, the neighbourhood of users based on which the
recommendation will be provided is selected
The weighted average of the ratings of the neighbourhood of
users for the item of interest is then calculated as follows:
Chapter 6 16
Agent Technology for e-Commerce
Example
Using Pearson’s correlation coefficients:
wD,A= 0.9 wD,B= - 0.7 wD,C= 0
pD,Item4= 4.5
Chapter 6 17
Agent Technology for e-Commerce
Issues
A critical mass of users is needed in order to create a database of
preferences: first-rater or cold start problem
New items cannot be recommended until someone has rated them
The scarcity of ratings (the user profiles are sparse vectors of
ratings) also presents a problem
Recommendations will come from users with which the active
user shares ratings (or votes) – this presents a problem to
methods such as Pearson’s correlation coefficients; potential
solutions: default voting
Chapter 6 18
Agent Technology for e-Commerce
Scalability: in systems with a large number of items and users,
computation grows linearly; appropriate algorithms that scale up
are needed
Reliability, especially in reputation systems: content providers
inflate their ratings
Lack of transparency: the user is given no indication whether to
trust a recommendation – incorporating explanation systems
would help address this concern
Privacy – once a system has built your profile, who else can have
access to it?
Chapter 6 19
Agent Technology for e-Commerce
Combing collaborative and content-based filtering
The underlying idea is that the content is also taken into account
when attempting to identify similar users for collaborative
recommendations
A number of systems have been developed: Fab, Tango, the
Recommender system, GroupLens’ approach
Chapter 6 20
Agent Technology for e-Commerce
Recommender systems in e-commerce
Turning browsers into customers: they can stimulate the users’
needs (need identification stage)
Cross-selling: suggest additional products which may match the
user’s interests or current shopping basket
Personalization: personalized services, or the site can be
personalized to the user’s liking – unique shopping experience
Keeping customers informed
Retaining customer loyalty
Chapter 6 21
Agent Technology for e-Commerce
Personalization
Vendors can identify exactly who is visiting their store through
registration, cookies, spyware
Vendors can personalize their websites for their customers
They can keep track of preferences, actions, they can build
profiles of their users. These can be used for marketing
Vendors can measure the users’ desires – dynamic pricing
When the consumer is unaware, then problems arise, possible
breaches of the user’s privacy. Who else gains access to these
profiles?
Negative impact on consumer confidence
Chapter 6 22
