Nirma University

Institute of Technology
Computer Science and Engineering Department

Course Policy Template

2CSOE51 Machine Learning [2 2 - 3]

B.Tech. (CSE)

Semester: V, Academic Year: 2022-23, Term: Odd

Course Code & Name : 2CSOE51 – Machine Learning

Credit Details : 3
Course Faculty : Dr. Preeti Kathiria (Course Coordinator)
Contact No. & Email : [email protected]
Office : New building 5th Floor
Visiting Hours : Monday to Friday - 8:45 to 4:00,
Saturday (Odd) - 8:45 to 4:45
Course Site : https://lms.nirmauni.ac.in/course/view.php?id=434

Introduction to Course:
Machine learning is the science of getting computers to act without being explicitly
programmed. In the past decade, machine learning has given us self-driving cars,
practical speech recognition, effective web search, and a vastly improved understanding
of the human genome.

Course Learning Outcomes:

At the end of the course, students will be able to –

1. Comprehend statistical methods as basis of machine learning domain

2. Learn variety of learning algorithms for appropriate applications
3. Identify various machine learning techniques to solve problems in applicable
domains
 Program Outcomes:

PO1 : an ability to apply knowledge of mathematics, science and engineering in practice

PO2 : an ability to identify, critically analyze, formulate and solve engineering problems with
comprehensive knowledge in the area of specialization

PO3 : an ability to select modern engineering tools and techniques and use them with appropriate skills

PO4 : an ability to design a system and process to meet desired needs within realistic constraints such as
health, safety, security and manufacturability

PO5 : an ability to contribute by research and innovation to solve engineering problems

PO6 : an ability to understand the impact of engineering solutions in a contemporary, global, economical,
environmental, and societal context for sustainable development

PO7 : an ability to function professionally with ethical responsibility as an individual as well as in

multidisciplinary teams with positive attitude

PO8 : an ability to communicate effectively

PO9 : an ability to appreciate the importance of goal setting and to recognize the need for life-long reflective
learning

Program Specific Outcomes:

PSO 1: To apply the theoretical concepts of computer engineering and practical

knowledge in analysis, design and development of computing systems and
interdisciplinary applications
PSO 2: To work as a socially responsible professional by applying computer engineering
principles and management practices

Program Educational Objectives:

PEO I: To prepare graduates who will be successful professionals in industry,
government, academia, research, entrepreneurial pursuit and consulting firms.

PEO II: To prepare graduates who will contribute to society as broadly educated,
expressive, ethical and responsible citizens with proven expertise.

PEO III: To prepare graduates who will achieve peer-recognition; as an individual or in a

team; through demonstration of good analytical, design and implementation skills.

PEO IV: To prepare graduates who will thrive to pursue life-long learning to fulfill their
goals.

Mapping of COs to POs and PSOs

PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PO12 PSO1 PSO2

CO1 3 2 1 0 0 0 0 0 0 0 0 0 3 0
CO2 2 0 0 3 0 0 0 0 0 0 1 0 3 0
CO3 0 3 0 0 2 0 0 0 1 0 0 1 3 2

Syllabus

Syllabus: Teaching
Hours
Unit I 03

Probability and Statistics: Empirical Probability, Theoretical Probability,

Joint Probability, Bayes’ Theorem, Descriptive Statistics, Measure of
Center, Measure of Variability, Measure of Position

Unit II 08

Regression Techniques: Basic concepts and applications of Regression,

Simple Linear Regression – Gradient Descent and Normal Equation
Method, Non-Linear Regression, Linear Regression with Regularization,
Hyper-parameters tuning, Loss Functions, Evaluation Measures for
Regression Techniques

Unit III 11

Classification Techniques: Basics of supervised learning, Naïve Bayes

Classification, Fitting Multivariate Bernoulli Distribution, Gaussian
Distribution and Multinomial Distribution, K-Nearest Neighbours,
Decision trees, Random Forest

Support Vector Machines: Hard Margin and Soft Margin, Kernels and
Kernel Trick, Evaluation Measures for Classification Techniques

Unit IV 04

Artificial Neural Networks: Biological Neurons and Biological Neural

Networks, Perceptron Learning, Activation Functions, Multilayer
Perceptron, Back-propagation Neural Networks

Unit V 04

Clustering: Basics of unsupervised learning, Hierarchical Agglomerative

Clustering, k-means Algorithm, Self-Organizing Maps

Self-Study:
The self-study contents will be declared at the commencement of semester. Around
10% of the questions will be asked from self-study contents.

Laboratory Work:
Laboratory work will be based on applications of the above syllabus with minimum
10 experiments to be incorporated.

Suggested Readings:
1. Tom Mitchell, Machine Learning, TMH
2. C. Bishop, Pattern Recognition and Machine Learning, Springer
 3. R. O. Duda, P. E. Hart and D. G. Stork, Pattern Classification and Scene Analysis, Wiley
4. Kishan Mehrotra, Chilukuri Mohan and Sanjay Ranka, Elements of Artificial Neural
Networks, Penram International
5. Rajjan Shinghal, Pattern Recognition, Techniques and Applications, OXFORD
6. Athem Ealpaydin, Introduction to Machine Learning, PHI

Component wise Continuous Evaluation & Semester End Examination

weightage:

Course Code 2CSOE51 Semester 7 Year 2021 Credit 3

Lectures/we 2 Practicals 2 Tutorials 0
ek /week /week

Exam Weightage Exam Hours

CE LPW SEE SEE Practical
0.4 0.2 0.4 3 -

Breakup of CE
Unit 1 Unit 2 Unit 3
Quiz1 and Quiz2 Sessional Assignment

Inter Component 0.3 0.4 0.3

Weightage

Lesson Plan
Sr. Topics Hours CLOs
No
.
1 Introduction: [02] CLO 1
• Theory and practices in machine learning
• Overview of machine learning along with its definition and [01]
its application area
• Types of machine learning techniques
• Significance of Model Training [01]
• Importance of Data Visualization

2 Probability and Statistics: [02]

• Empirical Probability, Theoretical Probability, Joint [01]
Probability
• Bayes’ Theorem, Descriptive Statistics, [01]
• Measure of Center, Measure of Variability, Measure of
Position
3 Supervised learning for Prediction [08] CLO1, CLO2
• Basic concepts and applications of Regression, Simple [02]
Linear Regression – Gradient Descent and Normal Equation
Method
• Multiple Linear Regression, Non-Linear Regression [02]
• Linear Regression with Regularization, Hyper-parameters [01]
tuning
• Loss Functions [01]
• Evaluation Measures for Regression Techniques [02]
4 Classification Techniques [10] CLO 1, CLO
• Basics of supervised learning [01] 2
• Naïve Bayes Classification [02]
• Fitting Multivariate Bernoulli Distribution, [01]
• Gaussian Distribution and Multinomial Distribution
• K-Nearest Neighbours [02]
• Decision trees: ID3,C4.5, CART [02]
• Random Forest
• Support Vector Machines: Hard Margin and Soft Margin, [02]
Kernels and Kernel Trick, Evaluation Measures for
Classification Techniques
5 Artificial Neural Networks [04] CLO1, CLO2
• Biological Neurons and Networks [01]
• Artificial Neuron Model (Perceptron Learning) [01]
• Activation Functions (Linear, step, ramp, log sigmoid and
tan sigmoid, ReLU, Leaky ReLU)
• Multilayer Perceptron
• Feed Forward ANN
• Back Propagation Neural Networks [02]
• Various Loss Functions (Binary and Categorical cross-
entropy)
6 Unsupervised Learning : Clustering [04] CLO1,CLO2
• Basics of unsupervised learning [01]
• K-means Clustering [02]
• Hierarchical Agglomerative clustering [01]
• Self-organizing maps
Total 30

List of Practical

Sr. Week List of Experiments Mapped

NO. No. with CO
1 1,2 Introduction to Python and Numpy (4 Hrs) 3
2 3,4 Introduction to Pandas, Matplotlib and Sklearn (4 Hrs) 3
3 5,6 Simple and Multiple Linear Regression using Gradient 3
Descent & Normal Equation Method (without using
sklearn or equivalent library for both) (4 Hrs)
 4 7 Linear Regression with Regularization (without using 3
sklearn or equivalent library) and Simple and Multiple
Linear Regression with and without regularization using
Sklearn (2 Hrs)
5 8 Naïve-Bayes – Multivariate Bernoulli, Multinomial and 3
Gaussian using sklearn (2 Hrs)
6 9 Decision Trees – ID3, C4.5 using sklearn (2 Hrs) 3
7 10 Support Vector Classification using sklearn (2 Hrs) 3
8 11 AND gate using Perceptron Learning (self- 3
implementation) (2 Hrs)
9 12,13 Ex-OR Gate/any other problem using Backpropagation 3
Neural Networks (self-implementation) (4 Hrs)
10 14,15 K-means using sklearn (4 Hrs) 3
Total 30

1. Course Assessment Schemes

(Course with Laboratory component)

COMPONENTS WEIGHTAGE

Course Details Continuous Evaluation (CE) LPW

No of Quiz Sessional Special Assignment * Lab Term

Assignme 1 /Tutorials Work End
Course nts and Exam
2 Exam
Subject Subject Name Coordinator No. of
Code Practic
als
WT WT Type of No of WT WT WT
Assignmen Tutorial
t s

2CSOE5 Machine Dr. Preeti 01 0.3 0.4 Project + - 0.3 10 0.75 0.25
1 Learning Kathiria Other
(Computat
ional
Coding)

Teaching-learning methodology: (Proposed)

• Lectures: Use of Black board, PPT, Discussion, Case Studies
• Practical: Use of python for implementing and doing
Active learning techniques:(Proposed)
• Flipped classroom
• Active discussions

Types of Special/Innovative Assignments, Term Papers, mini Projects

etc.
• Satellite image classification using feature extraction
• Deep feature Based image classification
• Reinforcement learning for robotic arm.

Course Material: (In the website)

• Course Policy
• PPTs, Notes, other Material
• Assignments, Tutorials, Lab Manuals
• Question bank
• Web-links, Blogs, Video Lectures, Journals
• Animations /Simulations, Software
• Advanced topics
• Industries/Organizations

Course Outcome Attainment:

• Use of formal evaluation components of continuous evaluation, laboratory work,
semester end examination.
• Informal feedback during course conduction.