Vivekanand Education Society’s

Institute of Technology,
Chembur, Mumbai

ARTIFICIAL NEURAL NETWORK AND FUZZY LOGIC

FA C U LT Y N A M E : A S H W I N I S AWA N T
A S S I S TA N T P R O F E S S O R
D E PA RT M E N T O F E L E C T R O N I C S & T E L E C O M M U N I C AT I O N E N G I N E E R I N G
 Introduction to
Artificial Neural Network and
Fuzzy Logic
Faculty Name: Ashwini Sawant

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 2

 Things to Know:
Course code: ECC604
Credits: 03
Course Course Name Teaching Scheme (Contact Hours) Credits Assigned
Code
Theory Practical Tutorial Theory Practical Tutorial Total
Artificial
ECC604 Neural Networks 03 -- -- 03 -- -- 03
and Fuzzy Logic

Course Course Name Examination Scheme

Code Theory Marks Exam Term Work Practical and Total
Internal Assessment End Sem. Exam. Oral
Duration
Test1 Test2 Avg.
(Hrs.)
Artificial
ECC604 Neural 20 20 20 80 03 -- -- 100
Networks and
Fuzzy Logic

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 3

 Course Pre-requisite and Objectives
Course Prerequisites:
1. Basic linear Algebra
2. Engineering Mathematics-I to IV

Course Objectives:
1. To introduce the concepts and understanding of artificial neural networks
2. To provide adequate knowledge about supervised and unsupervised neural networks
3. To introduce neural network design concepts
4. To expose neural networks based methods to solve real world complex problems
5. To study the architecture of CNN and its application in image classification.
6. To introduce fuzzy logic and fuzzy inference systems

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 4

 Course Outcomes
After successful completion of the course, the student will be able to:
1. Comprehend the concepts of biological neurons and artificial neurons
2. Analyze the feed-forward and feedback neural networks and their learning algorithms.
3. Comprehend the neural network training and design concepts
4. Build a simple CNN model and apply in image classification
5. Analyze the application of neural networks and fuzzy logic to real world problems.

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 5

SYLLABUS
Module 1. Introduction to Neural Networks and their Basic
Concepts (07 Hours)
1.1 Biological neuron and Artificial neuron, McCulloch-Pitts Model, Activation
Function, various types of Activation Functions and types of Neural Network

Architectures, Prerequisites for Training of Neural Networks. Linearly Separable and Linearly Non-
Separable Systems with examples, Concepts of Supervised Learning, Unsupervised Learning, and
Reinforcement Learning.
Brief survey of applications of Neural Networks.

Expected questions on: Definitions, Numerical and Derivation

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 6

 Module 1:- Introduction to Neural Networks
& their Basic Concepts
Hrs Marks Books Outcome
Topics T: Text Book R: Mapping
Reference

Biological neuron and Artificial 7 20 T1: Sivanandam- Ch1, 1, 2

neuron, McCulloch-Pitts Model. Ch2 and related
( 8 to 10 numerical (excluding
Activation Function, Various types of Activation Marks Hebb N/W)
Functions*. numerical
) T3: Rajsekaran* R7:
Types of Neural Network Mehrotra R8: Zurada
Architectures, Prerequisites for
Training of Neural Networks.
Linearly Separable and Linearly
Non-Separable Systems with
examples. Sivanandam reprint 2013
Concepts of Supervised Learning,
Unsupervised Learning, and
Reinforcement Learning.
Brief survey of applications of
Neural Networks.
Module 2. Supervised Learning Neural Networks
(07 Hours)
2.1. Perceptron - Single Layer Perceptron, Multilayer Perceptron and
their Architecture. Error Functions: Mean Square Error and Sum
Squared Error. Gradient Descent, Generalized delta rule, Error back
propagation, Stopping Criterions for Training.

Expected questions on: Definitions, Numerical

and Derivation
21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 8
 Module 2:-Supervised Learning Neural
Networks

Topics Hrs Marks Books Outcome

Mapping
Perceptron - Single Layer 7 20 T1 Sivanandam: Ch.3 2, 3
Perceptron, 49-73 and related
Multilayer Perceptron and their ( 8 to 10 numerical
Marks
Architecture.
numerical) T2: Haykin
Error Functions: Mean Square T3: Rajsekaran
Error and Sum Squared Error. R7: Mehrotra
R8: Zurada
Gradient Descent,
R3: Bishop
Generalized delta rule,
Error back propagation.
Stopping Criterions for Training
(T2, related topic in ch4)
Module 3. Unsupervised Learning Neural Networks (7 Hours)
3.1 Competitive Learning Network – Kohonen Self-Organizing
Networks – Architecture, Training Algorithm, Discrete Hopfield
Network- Hopfield Matrix, Testing Algorithm, K-Means Clustering
Algorithm.

Expected questions on: Definitions, Numerical

and Derivation
21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 10
 Module 3:-Unsupervised Learning Neural
Networks
Topics Hrs Marks Books Outcome
Mapping
7 20 T1: Sivanandam, 2, 3
Competitive Learning Network – Ch5 pp 155-161
Kohonen Self-Organizing Networks – ( 8 to 10 and related
Architecture, Training Algorithm.* Marks numericals
numerical)

Discrete Hopfield Network- Hopfield T1: Sivanandam,

Ch4 pp 110-114
Matrix, Testing Algorithm
and related
numericals
T2: Haykin, R3:
K-Means Clustering Algorithm. Bishop, (Related
portion in Ch5)
 Module 4. Algorithms of Neural Networks (04 Hours)
4.1 Basic concept of Machine Learning, Support Vector
Machine (SVM) - Introduction and SVM based Binary
Classifier, LMS Algorithm.
Expected questions on: Definitions, Numerical and Derivation

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 12

Module 4:- Algorithms of Neural Networks
Topics Hrs Marks Books Outcome
Mapping
Basic concept of Machine Learning, 4 15 T2: Haykin, Ch. 6 3
Support Vector Machine (SVM),
Introduction and SVM based Binary
Classifier.
LMS Algorithm. T2: Haykin, Ch.3
Module 5. Convolution Neural Network (CNN) (07 Hours)
5.1. Basic concept of Deep Learning, Convolution Operation,
Overview of CNN
Architecture, Input layer, Convolution layers, Pooling layers,
Padding, Strided Convolutions, Rectified Linear Unit (ReLU),
One Layer of a Convolutional
Network, Fully Connected Layers, Complex Image
Classification using CNN.
Expected questions on: Definitions, Numerical and Derivation
21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 14
 Module 5: Convolutional Neural Networks
Topics Hrs Marks Books Outcome
Mapping

Basic concept of Deep Learning 7 20 T4: Mohit Sewak: 4, 5

Ch2, 3 (and
Convolution Operation applications)
Overview of CNN Architecture,
Input layer, Convolution layers, R5: Ian Goodfellow:
Pooling layers, Padding, Strided Ch9
Convolutions, Rectified Linear Unit
(ReLU)
One Layer of a Convolutional
Network, Fully Connected Layers
Complex Image Classification using
CNN.
 Module 6. Introduction to Fuzzy Inference System (7 Hours)
6.1 Introduction to Fuzzy Logic, Fuzzy Rules, Fuzzy Properties - Operations,

Membership Functions, Fuzzification - Membership Value Assignments using

Intuition Method, Defuzzification Methods -- Mean of Maxima and Centroid

(Centre of Area) Methods, Fuzzy Inference System with reference to Mamdani
Model, Brief Review of Applications of Fuzzy Logic to Speed Control of DC
Motor and Washing Machine.
Expected questions on: Definitions, Numerical and Derivation

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 16

 Module 6:- Introduction to Fuzzy
Inference System
Topics Hours Marks Books Outcome
mapping
Introduction to Fuzzy Logic, Fuzzy T1: Sivanandam
Rules, Fuzzy Properties, Ch 7*,
Operations*. Fuzzy properties,
operations and
Membership Functions, 07 25
related numerical
5
Fuzzification - Membership Value Ch9 **
Assignments using Intuition (10 to 15
Marks
Method. **
numerical)
Defuzzification Methods -- Mean of Ch10***
Maxima and Centroid (Centre of
Area) Methods.***
Fuzzy Inference System with Ch12****
reference to Mamdani Model.**** T5: Ross,
Brief Review of Applications of R9: Kosko
Fuzzy Logic to Speed Control of DC
Motor and Washing Machine.

Total 39 Lectures
Textbooks
1. S. N. Sivanandam and S. N. Deepa, Introduction to Soft Computing, Wiley
India Publications, 3rd Edition.
2. Simon Haykin, Neural Networks and Learning Machines, Pearson Prentice
Hall, 3rd Edition
3. S. Rajasekaran and G. A. Vijayalakshmi Pai, Neural Networks, Fuzzy Logic,
and Genetic Algorithms, PHI Learning Pvt. Ltd, 2003.
4. Practical Convolutional Neural Networks by Mohit Sewak, Md. Rezaul Karim,
Pradeep Pujari, Packt Publishing, 2018.
5. Timothy J. Ross, Fuzzy Logic with Engineering Applications, Wiley India
Publications, 3rd Edition.

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 18

Reference Books
1. Hagan, Demuth, and Beale, Neural Network Design, Thomson Learning, 2nd Edition.
2. Simon Haykin, Neural Network- A Comprehensive Foundation, Pearson Education, 2nd Edition.
3. Christopher M. Bishop, Neural Networks for Pattern Recognition, Oxford University Press, 2005.
4. William W. Hsieh, Machine Learning Methods in the Environmental Sciences: Neural Network and
Kernels, Cambridge University Press, 2009.
5. Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press, 2016
6. S. N. Sivanandam, S. Sumathi, and S. N. Deepa, Introduction to Neural Network using Matlab, Tata
McGraw-Hill Publications, 2006.
7. Mehrotra Kishan, Mohan C. K. Ranka Sanjay, Elements of Artificial Neural Networks, Penram
International Publishing Pvt. Ltd, 2nd Edition.
8. J. M. Zurada, Introduction to Artificial Neural Systems, Jaico Publishers, 2006.
9. Bart Kosko, Neural Networks and Fuzzy Systems, Pearson Education, 2007.
21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 19
List of Suggested Experiments to be conducted in IPMV Laboratory
(ECL 603):
1.Classification of Non-linearly Separable Binary Pattern using
Multilayer Perceptron Neural Network.
2.Pattern Clustering using K-means Algorithm.
3.Binary Pattern Restoration using Discrete Hopfield Neural Network.
4.Image Classification using Support Vector Machine.
5.Object Recognition using Convolutional Neural Network.
6.Design Fuzzy Controller for Washing Machine

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 20

Artificial neural network:
https://www.youtube.com/watch?v=vpOLiDyhNUA

Current and Future Applications for Neural Networks

https://www.youtube.com/watch?v=zzcujMwKlWg

Neural Network & Its business applications

https://www.youtube.com/watch?v=9HJSMHUParQ

21/02/2022 ASHWINI SAWANT, ASSISTANT PROFESSOR, VESIT 21