Scribd
Upload
1K views

Soft Computing Syllabus

This document outlines the course objectives, outcomes, topics, and evaluation scheme for a course on Soft Computing Techniques. The course introduces neural networks, fuzzy logic, and hybrid systems as tools for solving complex real-world problems. It aims to provide an understanding of these techniques and their application areas. Topics covered include artificial neural networks, fuzzy logic, adaptive neuro-fuzzy inference systems, and fuzzy control systems. Students will be evaluated through in-semester and end-semester exams.

Uploaded by

Anil V. Walke
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
1K views

Soft Computing Syllabus

This document outlines the course objectives, outcomes, topics, and evaluation scheme for a course on Soft Computing Techniques. The course introduces neural networks, fuzzy logic, and hybrid systems as tools for solving complex real-world problems. It aims to provide an understanding of these techniques and their application areas. Topics covered include artificial neural networks, fuzzy logic, adaptive neuro-fuzzy inference systems, and fuzzy control systems. Students will be evaluated through in-semester and end-semester exams.

Uploaded by

Anil V. Walke
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

SOFT COMPUTING TECHNIQUES(404191)

Teaching Scheme: Examination Scheme:


Lectures: 3Hrs/ Week In Semester Assessment:
Phase I : 30
End Semester Examination:
Phase II: 70
Course Objectives:
Introduce a relatively new computing paradigm for creating intelligent machines useful
for solving complex real world problems.
Insight into the tools that make up the soft computing technique: fuzzy logic, artificial
neural networks and hybrid systemsTechniques.
To create awareness of the application areas of soft computing technique
Provide alternative solutions to the conventional problem solving techniques in
image/signal processing, pattern recognition/classification, control system

Course Outcomes:
Having successfully completing the course students will be able to
use a new tool /tools to solve a wide variety of real world problems
find an alternate solution , which may offer more adaptability, resilience and optimization
Identify the suitable antenna for a given communication system
Gain knowledge of soft computing domain which opens up a whole new career option
Tackle real world research problems

Unit I : Artificial Neural Network -I 8L


Biological neuron, Artificial neuron model, concept of bias and threshold , McCulloch Pits
Neuron Model , implementation of logical AND, OR, XOR functions Soft Topologies of neural
networks, learning paradigms: supervised, unsupervised, reinforcement, Linear neuron model :
concept of error energy , gradient descent algorithm and application of linear neuron for linear
regression, Activation functions : binary , bipolar (linear, signup, log sigmoid, tan sigmoid)
Learning mechanisms: Hebbian, Delta Rule o Perceptron and its limitations Draft

Unit II : Artificial Neural Network-II 8L


Multilayer perceptron (MLP) and back propagation algorithm o Application of MLP for
classification and regression o Self organizing Feature Maps, k means clustering o Learning
vector quantization Radial Basis Function networks: Covers theorem, mapping functions
(Gaussian, Multi-quadrics, Inverse multiquadrics, Application of RBFN for classification and
regression o Hopfield network, associative memories.

Unit III : Fuzzy Logic -I 6L


Concept of Fuzzy number, fuzzy set theory(continuous, discrete) o Operations on fuzzy sets,
Fuzzy membership functions (core ,boundary ,support) , primary and composite linguistic terms ,
Concept of fuzzy relation, composition operation (T-norm,Tconorm) o Fuzzy ifthen rules.

Unit IV : Fuzzy Logic -II 6L


Fuzzification , Membership Value Assignment techniques, Defuzzification ( Maxmembership
principle, Centroid method, Weighted average method), Concept of fuzzy inference, Implication
rules- DienesRescher Implication, Mamdani Implication, Zadeh Implication, Fuzzy Inference
systems Mamdani fuzzy model , Sugeno fuzzy model , Tsukamoto fuzzy model,
Implementation of a simple twoinput single output FIS employing Mamdani model Computing.

Unit V : Fuzzy Control Systems 6L


CONTROL SYSTEM DESIGN PROBLEM 1.5, Control (Decision) Surface, Assumptions in a
Fuzzy Control System Design V, Fuzzy Logic Controllers Soft o Comparison with traditional
PID control, advantages of FLC, Architecture of a FLC: Mamdani Type , Example Aircraft
landing control problem.

Unit VI : Adaptive Neuro-Fuzzy Inference Systems(ANFIS) 6L


ANFIS architecture, Hybrid Learning Algorithm, Advantages and Limitations of ANFIS
Application of ANFIS/CANFIS for regression

Text Books
1. Fundamentals of Neural Networks: Architectures, Algorithms And Applications,
LaureneFausett, Pearson Education, Inc, 2008 .
2. Fuzzy Logic With Engineering Applications, Third Edition Thomas, Timothy Ross,
John Wiley & Sons,2010
3. Neuro- Fuzzy and Soft Computing, J.S. Jang, C.T. Sun, E. Mizutani, PHI Learning
Private Limited.
4. Principles of Soft Computing , S. N. Sivanandam, S. N. Deepa, John Wiley & Sons,
2007
Reference Books
1. Introduction to the theory of neural computation, John Hertz, Anders Krogh, Richard
Palmer, Addison Wesley Publishing Company, 1991
2. Neural NetworksA comprehensive foundation,, Simon Haykin,Prentice Hall
International Inc-1999
3. Neural and Adaptive Systems: Fundamentals through Simulations, Jos C. Principe
Neil R. Euliano , W. Curt Lefebvre, John-Wiley & Sons, 2000
4. Pattern Classification, Peter E. Hart, David G. Stork Richard O.Duda,Second
Edition,2000
5. Pattern Recognition, SergiosTheodoridis , KonstantinosKoutroumbas, Fourth Edition,
Academic Press, 2008
6. A First Course in Fuzzy Logic, Third Edition, Hung T. Nguyen, Elbert A. Walker,
Taylor & Francis Group, LLC, 2008
7. Introduction to Fuzzy Logic using MATLAB, S. N. Sivanandam , S.Sumathi, S. N.
Deepa, Springer Verlag, 2007
Practical Sessions: (Use MATLAB / OCTAVE/ SCILAB /any appropriate open source
software.)(any 8 experiments)
1. Implement simple logic network using MP neuron model
2. Implement a simple linear regressor with a single neuron model
3. Implement and test MLP trained with back-propagation algorithm
4. Implement and test RBF network
5. Implement SOFM for character recognition
6. Implement fuzzy membership functions (triangular, trapezoidal, gbell, PI, Gamma,
Gaussian)
7. Implement defuzzyfication (Max-membership principle, Centroid method, Weighted
average method)
8. Implement FIS with Mamdani inferencing mechanism
9. A small project: may include classification or regression problem, using any soft
computing technique studied earlier

You might also like