Machin

e
Learnin
g
 Chapter 1

Introduction to Machine Learning

 1.1 Need for Machine Learning?
• Business organizations use huge amount of data for their daily activities.
• Need to analyze data for taking decisions
Machine Learning has become so popular because of 3 reasons:
1. High volume of available data to manage
• Big companies such as facebook, Twitter and Youtube generate huge amount of data that grows at
phenomenal rate.
2. Cost of storage has reduced.
• The hardware cost has also dropped. Therefore, it is easier now to capture, process, store,
distribute, and transmit the digital information.
3. Availability of complex algorithms and tools
 Need for Machine Learning?
• Before starting the machine learning journey, let us establish these terms - data, information, knowledge,
intelligence, and wisdom.
 Need for Machine
Learning?
Data can be numbers or text that can be processed by a computer.
• Today, organizations are accumulating vast and growing amounts of data in different storage formats.
Processed data is called information.
• This includes patterns, associations, or relationships among data.
• For example, sales data can be analyzed to 'extract information’ like which is the fast selling product.
Condensed information is called knowledge.
•For example, the historical patterns and future trends obtained in the above sales data can be called
knowledge.
•Unless knowledge is extracted, data is of no use.
•Similarly, knowledge is not useful unless it is put into action.
An actionable form of knowledge is called intelligence. Computer systems have been successful till this
stage.
 The ultimate objective of knowledge pyramid is wisdom that represents the maturity of mind that is, so
far, exhibited only by humans.
 1.2 Machine Learning Explained
• “MACHINE LEARNING IS A FIELD OF STUDY THAT GIVES THE COMPUTERS TO LEARN WITHOUT BEING
EXPLICITLY PROGRAMMED”
 Machine Learning Explained
In conventional programming, after understanding the problem, a detailed design of the program
such as a flowchart or an algorithm needs to be created and converted into programs using a suitable
programming language.
This approach could be difficult for many real-world problems such as puzzles, games, and complex
image recognition applications.
Initially, artificial intelligence aims to understand these problems and develop general purpose rules
manually.
Then, these rules are formulated into logic and implemented in a program to create intelligent
systems.
This idea of developing intelligent systems by using logic and reasoning by converting an expert's
knowledge into a set of rules and programs is called an expert system.
As humans take decisions based on an experience, computers make models based on extracted
patterns in the input data and then use these data-filled models for prediction and to take decisions.
For computers, the “learnt model is equivalent to human experience”.
 Machine Learning Explained
 In statistical learning, the relationship between the input x and output y is modeled as a
function in the form y=f(x).
 Here, f is the learning function that maps the input x to output y.
 Learning of function f is the crucial aspect of forming a model in statistical learning.
 In machine learning, this is simply called mapping of input to output.

Formally: stated, a model is an explicit description of patterns within the data in the form of:
• Mathematical Equation

• Relational diagrams like Graphs/Trees

• Logical IF/ELSE rules
• Groupings called Clusters
 Another View of Machine Learning

Another pioneer of AI, Tom Mitchell's definition of machine learning states that,
"A computer program is said to learn from experience E, with respect to
task T and some performance measure P, if its performance on T measured
by P improves with experience E."

The important components of this definition are experience E, task T, and performance
measure P.
 Other Fields
• RELATION BETWEEN MACHINE LEARNING AND AI
 Machine Learning and AI
Contd…
The aim of AI is to develop intelligent agents. An agent can be a robot, humans, or any
autonomous systems.
Machine learning is the subbranch of AI, whose aim is to extract the patterns for prediction.
It is a broad field that includes learning from examples and other areas like reinforcement
learning.
Deep learning is a subbranch of machine learning. In deep learning, the models are
constructed using neural network technology.
Neural networks are based on the human neuron models.
Many neurons form a network connected with the activation functions that trigger further
neurons to perform tasks.
 Machine Learning and Data
Science
DATA SCIENCE IS AN “UMBRELLA TERM” COVERING FROM DATA COLLECTION TO DATA ANALYSIS.
 Machine Learning and Data
Science
• CHARACTERISTICS OF BIG DATA (3 V’s)

Big Data: Data Science concerns about collection of data. Big data is a field of data science that
deals with data’s following characteristics:
1. Volume: Huge amount of data is generated by big companies like Facebook, Twiitter,
YouTube.
2. Variety: Data is available in variety of forms like images, videos, and in different formats.
3. Velocity: It refers to the speed at which the data is generated and processed.
 Science
• DATA SCIENCE AND DATA MINING
 Data Mining: Like while mining the earth one gets into precious resources, it is often believed
that unearthing of the data produces hidden information.
 Nowadays, many consider that data mining and machine learning are same.
 There is no difference between these fields except that data mining aims to extract the
hidden patterns that are present in the data, whereas, machine learning aims to use it for
prediction.

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Science
• DATA SCIENCE AND DATA ANALYTICS / PATTERN RECOGNITION
Data Analytics: Another branch of data science.
 It aims to extract knowledge from crude data.
 There are different types of analytics.
 Predictive data analytics is used for making predictions.
 Machine learning is closely related to this branch of analytics and shares almost all
algorithms.
Pattern Recognition: It is an engineering field.
 It uses machine learning algorithms to extract the features for pattern analysis and pattern
classification.
 One can view pattern recognition as a specific application of machine learning.

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Machine Learning and Statistics
ROLE OF
STATISTICSIt is a branch of mathematics.
Statistics:
 Like machine learning (ML), it can learn from data.
 Statistical methods look for regularity in data called patterns.
 Initially, statistics sets a hypothesis and performs experiments to verify and validate the
hypothesis in order to find relationships among data.

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

Machine Learning Types

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

Labelled Data

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Unlabelled Data
DATA THAT IS NOT ASSOCIATED WITH LABELS IS CALLED UNLABELLED DATA.

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Supervised Learning
CLASSIFICATION

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Supervised Learning
KEY ALGORITHMS
Some of the key algorithms of classification are:
 Decision Tree
 Random Forest
 Support Vector Machines
 Naïve Bayes
 Artificial Neural Network and Deep Learning networks like CNN

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Supervised Learning
REGRESSION ALGORITHMS

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

Classification vs. Regression
Classification and regression are both types of supervised learning, but they serve different
purposes.

Feature Classification Regression

Assigns labels/categories to input Predicts continuous numerical

Definition
data values

Discrete (e.g., classes like "Spam" Continuous (e.g., temperature,

Output Type
or "Not Spam") price)

House price prediction, stock price

Email spam detection, image
Example Use Cases forecasting, temperature
recognition, medical diagnosis
prediction

Logistic Regression, Decision Trees, Linear Regression, Polynomial

Algorithms Random Forest, SVM, Neural Regression, Decision Trees, Neural
Networks Networks
 Unsupervised Learning
CLUSTERING IS A GROUPING PROCESS.

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Unsupervised Learning
KEY ALGORITHMS OF UNSUPERVISED LEARNING
 K-means algorithm
 Hierarchical algorithms

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

Key Differences

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Semi-supervised Learning
 Semi-supervised learning (SSL) is a machine learning approach that combines a small amount
of labeled data with a large amount of unlabeled data.
This method is useful when labeling data is expensive or time-consuming.
Example: Classifying Emails as Spam or Not Spam
Scenario: You are developing a machine learning model to classify emails as spam or not spam.
However, you have only a small set of labeled emails and a large set of unlabeled emails.
 First, train a model on the small labeled dataset.
 Use this trained model to predict labels for the unlabeled data (pseudo-labeling).
 Semi-supervised algorithms use unlabelled data by assigning a pseudo-label.
 Then, the labelled and pseudo-labelled dataset can be combined

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Reinforcement Learning
 Reinforcement learning mimics human beings.
 Like human beings use ears and eyes to perceive the world and take
actions.
 Reinforcement learning allows the agent to interact with the
environment to get rewards.
 The agent can be human, animal, robot, or any independent
program.
 The rewards enable the agent to gain experience. The agent aims to
maximize the reward.
 The reward can be positive or negative(punishment). When the
rewards are more, the behavior gets reinforced and learning
becomes possible.
© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED
 Challenges of Machine Learning

1. ILL-POSED PROBLEMS – PROBLEMS WHOSE SPECIFICATIONS ARE NOT CLEAR

2. HUGE DATA
3. HUGE COMPUTATION POWER
4. COMPLEXITY OF ALGORITHMS
5. BIAS-VARIANCE

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Machine Learning Process
MACHINE LEARNING/DATA MINING PROCESS

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Process
1. Understanding the business: this step involves understanding the objectives and
requirements of the business organization.
Also involves the formulation of the problem statement for the data mining process.
2. Understanding the data: It involves the steps like data collection, study of the characteristics
of the data, formulation of hypothesis and matching of patterns to the selected hypothesis.
3. Preparation of data(preprocessing): Producing the final dataset by cleaning the raw data,
handling the missing values, eliminating redundant data.
4. Modelling: Involves training of the data mining\ML algorithm to obtain the pattern(useful
information).
5. Evaluate: Involves the evaluation of the model result using statistical analysis and
visualization methods.
The performance of the classifier is determined by evaluating the accuracy of the classifier.
6. Deployment: Involves the deployment of results of the data mining algorithm to improve the
existing process or for a new situation.
 Machine Learning Applications
MACHINE LEARNING MAJOR APPLICATIONS

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Machine Learning Applications

MACHINE LEARNING MAJOR APPLICATIONS

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Summary

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED

 Summary

© OXFORD UNIVERSITY PRESS 2021. ALL RIGHTS RESERVED