MACHINE LEARNING

1. Course Details
Program Name BTECH COMPUTER SCIENCE AND INFORMATION TECHNOLOGY
Qualification Pack
Course Name MACHINE LEARNING Course Code BTCS0505
Version No 1.0 Version Update date
Pre-requisite Programming (Python), Linear Algebra
Course Outcome CO1. Define machine learning and differentiate among supervised, unsupervised
learning and reinforcement learning. Compare machine learning techniques such as
Classification, Regression, Clustering, Association Rule mining. Give example
applications for each..
CO2. Explain various algorithms under each ML technique and interpret the
associated procedure in mathematical terms
CO3. Discuss neural network as a deep learning technique. Describe deep learning
and explain standard algorithms used in Deep Learning.
CO4. Evaluate the performance of algorithms/ models using metrics such as
confusion matrix, accuracy, recall, ROC curves and PR curves. employ procedures
such as cross validation
CO5. Demonstrate the use of the algorithms in real-world problem by
implementation, optimize the models learned and report the accuracy, recall and
precision and other performance measures.
Total Credits / 4/1:1:1:1
L:T:P:S
Teaching & Teaching Scheme Examination Scheme (Marks)
Examination
Scheme
Hours (Sessions in L T P S Th Pr Th Pr. Skill
term, should match Internal Internal Term Term Assessment
with hrs mentioned end end
in Syllabus) 50 40 20
15 15 30 15 50 40

2. COURSE DESCRIPTION
Sr. No Module/Units Learning outcomes Instructional
Objectives
1 Unit 1 Unit 1: LO1: Explain what machine learning,
Introduction to how is it employed in different
Machine Learning application and compare diverse types
of machine learning techniques

2 Unit 2: Supervised LO2: Define and discuss terms

Learning Hypothesis, Model, Feature Space, Cost
Function & parameter learning with
respect to supervised learning
algorithms

LO3: Explain the mathematical intuition

behind (regression and classification –
supervised learning) algorithms; linear
regression, logistic regression, decision
trees and SVM.

LO4: Measure the performance of

algorithms and models using metrics
like confusion matrix, accuracy, Fβ
measure, recall and precision

LO5: Implement (supervised learning)

algorithm models, check performance
and compare accuracy of models &
algorithms

3 Unit 3: Unsupervised LO3: Explain the mathematical intuition

Learning behind (unsupervised learning )
algorithms; frequent item-set, K-Means
clustering

LO3: Explain Reinforcement learning

and relevant techniques in RL

LO4: Measure the performance of

(clustering) algorithms and models
using metrics like confusion matrix,
accuracy, Fβ measure, recall and
precision

LO5: Implement (unsupervised learning)

algorithm models, check performance
and compare accuracy of models &
 algorithms

4 Unit 4: Neural LO3: Explain the mathematical intuition

Networks behind (neural networks/perceptron)
algorithms; CNN, Backpropagation,

L05: Implement (CNN as an image

classifier) algorithm models, check
performance and compare accuracy of
models & algorithms

5 Unit 5:Real World LO6 Use Machine learning techniques

Machine Learning to solve real world problems by
identification of correct algorithm

LO7 Design and analyze with ML

models through experiments.

3. SYLLABUS
Sr. No Module/Units Detailed Topic Wise Syllabus References

1 Unit 1: Introduction to Introduction

Machine Learning
 2 - What is Machine Learning
Theory Duration (05
HR): 3 - Types and Applications of
ML
Practical Duration (10
4 - AI vs ML
HR):
5 - Essential Math for ML and AI

2 Unit 2: Supervised
Learning 6 - Introduction to Supervised
Learning
Theory Duration
(05:00): 7 – Regression Technique -
Linear Regression
Practical Duration 8 - CLogistic Regression
(10.00):
9 - Support Vector Machines

10 – Decision Trees

11 - Model Selection Procedures

12 – Implementation of
Algorithms Using Popular
Programming Language

3 Unit 3:Unsupervised 14 - Introduction to

Learning Unsupervised Learning
Theory Duration
(05:00): 15 - Association Rules

16 - Cluster Analysis
Practical Duration
(10.00): 17 - Reinforcement Learning

18 - Bonus! KMeans Clustering

Project

4 Unit 4: Neural 19 - Introduction to Neural

Networks Networks
Theory Duration
(05:00): 20 - The Perceptron

21 - The Backpropagation
Practical Duration
Algorithm
(10.00):
 22 - Training Procedures

Convolutional Neural Network

5 Unit 5:Real World 24 - Introduction to Real World

Machine Learning ML

Theory Duration 25 - Choosing an Algorithm

(05:00):
26 - Design and Analysis of ML
Experiments
Practical Duration
(10:00): 27 - Common Software for ML
Minor Project
 4. Learning Resources
Sr Module/ Text Books References/ Online
. Units Resources (Link)
N
o
1 UNIT – I 1. V.K. Jain, Machine Learning, Khanna https://www.youtube.com/
Introduction Publishing House playlist?
to Machine 2. Vinod Chandra S.S., Artificial list=PL1xHD4vteKYVpaIiy295pg6_S
Learning Intelligence & Machine Learning, PHI Y5qznc77
3. Tom Mitchel, Machine Learning, NPTEL – MOOC Machine Learning
McGraw-Hill 2016 Lecture 1-4 Week 1
Practical
Duration https://www.youtube.com/
(06:00): playlist?list=PLLssT5z_DsK-
h9vYZkQkYNWcItqhlRJLN
Machine Learning Andrew NG
Lecture 1.1 – 1.3

2 Unit 2: 1. V.K. Jain, Machine Learning, Khanna https://www.youtube.com/

Supervised Publishing House playlist?
Learning 2. Vinod Chandra S.S., Artificial list=PL1xHD4vteKYVpaIiy295pg6_S
Intelligence & Machine Learning, PHI Y5qznc77
3. Tom Mitchel, Machine Learning, NPTEL – MOOC Machine Learning
Practical McGraw-Hill 2016 Lecture 14-16, 21- 24
Duration https://www.youtube.com/
(12.00): playlist?list=PLLssT5z_DsK-
h9vYZkQkYNWcItqhlRJLN
Machine Learning Andrew NG
Lecture 2.1 – 2.7, 6.2-6.5

3 Unit 1. V.K. Jain, Machine Learning, Khanna https://www.youtube.com/

3:Unsupervis Publishing House playlist?
ed Learning 2. Vinod Chandra S.S., Artificial list=PL1xHD4vteKYVpaIiy295pg6_S
Intelligence & Machine Learning, PHI Y5qznc77
NPTEL – MOOC Machine Learning
 Practical 3. Tom Mitchel, Machine Learning, 2016 Lecture 65-66
Duration McGraw-Hill
(09.00):
4 Unit 4: 1. V.K. Jain, Machine Learning, Khanna https://
Neural Publishing House www.deeplearningbook.org/
Networks 2. Vinod Chandra S.S., Artificial
Intelligence & Machine Learning, PHI
3. Tom Mitchel, Machine Learning,
Practical McGraw-Hill.
Duration
(09.00):
5 Unit 5:Real Student is open to refer any material available
World depending on the project chosen
Machine
Learning
Practical
Duration
(09.00):
 5. Learning Outcome Achievement (Assignments / Activities)

Case Study analysis

Learning Outcome

Group Discussions

Other (Pl explain)

Prototype making
Lab Experiment
Written Test

Presentation

Role Play
Research
Practical

Seminar
Tutorial
Project
Sr. Learning
ID

No Module/Unit Outcome
Knowledg
e/Skill
/ Attitude

1 Introduction LO1 ✓ ✓ ✓ ✓ ✓
LO2 ✓ ✓ ✓
LO3 ✓ ✓
2 Supervised
LO4 ✓ ✓ ✓ ✓ ✓
Learning
LO5 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
LO3 ✓ ✓ ✓ ✓ ✓
Unsupervised
3 LO4 ✓ ✓ ✓ ✓ ✓
Learning
LO5 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Neural LO3 ✓ ✓ ✓ ✓
4
Network LO5 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Real World LO6 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
5
Problem LO7 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
6. Evaluation System
Description Allotted marks
Internal Theory 50
Internal Practical 40
Term end Theory 50
Term end Practical 40
Skill Assessment 20
TOTAL 200

7. Weightage of Units for Examination (Faculty to suggest)

Unit % Weightage
Introduction 20
Supervised Learning 20
Unsupervised Learning 20
Neural Networks 20
Real World Problems 20

8. Teaching Plan (Classroom Lectures)

Session Week
Module/Unit Topic
Number Number
1 Introduction .Introduction, Techniques, Week 1
2 Introduction Compare techniques, Applications Week 2
3 Supervised Learning Supervised Learning, Process, Terminologies Week 3
4 Supervised Learning Linear Regression, Cost Function, Hypothesis Week 4
5 Supervised Learning Logistic Regression, Cost Function, Hypothesis Week 5
Decision trees, Information Gain, entropy, Gini
6 Supervised Learning Week 6
Index
Support Vector Machines, Intuition, Hypothesis,
7 Supervised Learning Week 7
Cost function
Performance Metrics – Confusion Matrix,
8 Supervised Learning Week 8
Precision, Recall, Accuracy, ROC and PR curves
9 Unsupervised Learning Definition, Techniques, Frequent Item Set Mining Week 9
10 Unsupervised Learning Discussion of K- means algorithm Week 10
Performance Metrics: Elbow Method, Silhouette
11 Unsupervised Learning Week 11
Analysis, RAND index, Mutual Information
Introduction to Neural networks and Perceptron
12 Neural Networks Week 12
based algorithm
13 Neural Networks Backpropagation algorithm Week 13
14 Neural Networks CNN as deep leaning for image classification Week 14
 9. Practical Plan
Session Equipm
Module/ Week
Numbe Description of Experiments ent
Unit Number
r /Tools
Introduction
1,2,3, to Python / Data Manipulation with Pandas Python Week 1,2,3
Pandas
Supervised
4,5,6 Computation of Confusion Matrix Week 4,5,6
Learning
Supervised Testing and Tuning performance with multiple
7,8,9 Week 7,8,9
Learning algorithms
Unsupervised Week
10,11,12 Implementing K Means Clustering Algoritm
Learning 10,11,12
Neural Week
13,14,15 Implementing Image Classification with CNN
Networks 13,14,15

10. Skill Plan

Equipm
Session Week
Module/Unit Topic ent
Number Number
/Tools
Supervised Operating
1,2,3 Linear Regression 1-3
learning System
Supervised Operating
4,5,6 Logistic Regression 4-6
learning System
Supervised WAMP
7,8,9 ROC curves 7-9
learning Server
Supervised Suitable
10,11,12 Tuning SVM parameters 10-12
learning Editor
Unsupervised Suitable
13,14,15 Computing different distance measured. 13-15
learning Editor