Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
PRNN syllabus.pdf
1. BM016 PATTERN RECOGNITION AND NEURAL
NETWORKS
3 0 2 4
Prerequisite Nil
Course
Objectives
The course will introduce the student to
1. The fundamentals of pattern recognition and its application.
2. Implement several supervised and unsupervised algorithms suitable for
pattern classification.
3. Analyse computational methods such as linear discriminant functions and
nearest neighbour rule. Implement basic neural network architectures and
learning algorithms,
4. Apply in pattern recognition, image processing, and computer vision
5. Use the Pattern and Neural Classifiers for classification applications.
Course
Outcomes
On successful completion of the course, the student will be able to:
1. Implement the fundamentals of pattern recognition and neural networks.
(2,3)
2. Design and apply different pattern recognition techniques to the
applications of interest (3,4)
Note: Numbers given in the parenthesis refer to Graduate Attributes required
by NBA.
UNIT I INTRODUCTION TO PATTERN RECOGNITION AND SUPERVISED
LEARNING
Overview of Pattern recognition, Types of Pattern recognition, Parametric and
Nonparametric approach, Bayesian classifier, Discriminant function, non-parametric density
estimation, histograms, kernels, window estimators, k- nearest neighbor classifier, estimation
of error rates.
UNIT II UNSUPERVISED LEARNING AND CLUSTERING ANALYSIS
Unsupervised learning- Hierarchial clustering- Single-linkage Algorithm, Complete –
linkage Algorithm, Average-linkage algorithm and Ward’s method.Partitional clustering-
Forgy’s Algorithm, k-means algorithm and Isodata Algorithm
UNIT III INTRODUCTION TO SIMPLE NEURAL NETWORK
Elementary neurophysiology and biological neural network- Artificial neural
network-Architecture, biases and thresholds, Hebb net, Perceptron, Adaline and Madaline.
2. UNIT IV BACK PROPAGATION AND ASSOCIATIVE MEMORY
Back propagation network, generalized delta rule, Bidirectional Associative memory
Hopfield Network
UNIT V NEURAL NETWORKS BASED ON COMPETITIVE LEARNING
Kohonen Self organizing map, Learning Vector Quantisation, Counter Propagation network.
TEXT BOOKS
1. Duda R.O. Hart P.G, “Pattern Classification and scene analysis”, Wiley Edition 2000.
2. Hagan, Demuth and Beale, “Neural network design”, Vikas Publishing House Pvt Ltd., New Delhi,
2002 .
REFERENCES
1.Freeman J.A., and Skapura B.M, “Neural Networks, Algorithms, Applications and Programming
Techniques”, Addison - Wesley, 2003.
2.Earl Gose, Richard Johnsonbaugh Steve Jost, “Pattern Recognition and Image Analysis”, Prentice Hall
of India Pvt Ltd., New Delhi, 1999.
3.Robert Schalkoff, “Pattern recognition, Statistical, Structural and neural approaches” John Wiley and
Sons (Asia) Pvt Ltd., Singapore, 2005.
4.Laurene Fausett, “Fundamentals of neural networks- Architectures, algorithms and applications”,
Prentice Hall, 1994