GUJARAT TECHNOLOGICAL UNIVERSITY
Bachelor of Engineering
Subject Code: 3171613
PATTERN RECOGNITION
B.E. 7th Semester
Type of course: Elective
Prerequisite: Familiarity with linear algebra, probability, random process, and statistics
Rationale: Pattern recognition techniques are used to design automated systems that improve their own
performance through experience. This course covers the methodologies, technologies, and algorithms of
statistical pattern recognition from a variety of perspectives.
Teaching and Examination Scheme:
Teaching Scheme Credits Examination Marks Total
L T P C Theory Marks Practical Marks Marks
ESE (E) PA (M) ESE (V) PA (I)
3 0 2 4 70 30 30 20 150
Syllabus:
Sr. Content Total
No. Hrs
1 Basics of Probability, Random Processes and Linear Algebra (recap): Probability: 06
independence of events, conditional and joint probability, Bayes theorem Random Processes:
Stationary and non-stationary processes, Expectation, Autocorrelation, Cross-Correlation,
spectra.
2 Linear Algebra: Inner product, outer product, inverses, eigen values, eigen vectors, singular 03
values, singular vectors.
3 Bayes Decision Theory: Minimum-error-rate classification. Classifiers, Discriminant 03
functions, Decision surfaces. Normal density and discriminant functions. Discrete features.
4 Parameter Estimation Methods: Maximum-Likelihood Estimation: Gaussian case. 12
Maximum a Posteriori estimation. Bayesian estimation: Gaussian case. Unsupervised
learning and clustering - Criterion functions for clustering. Algorithms for clustering: K-
Means, Hierarchical and other methods. Cluster validation. Gaussian mixture models,
Expectation-Maximization method for parameter estimation. Maximum entropy estimation.
Sequential Pattern Recognition. Hidden Markov Models (HMMs). Discrete HMMs.
Continuous HMMs. Nonparametric techniques for density estimation, K-Nearest Neighbour
method.
5 Dimensionality reduction: Principal component analysis - it relationship to Eigen analysis. 08
Fisher discriminant analysis - Generalized Eigen analysis. Eigen vectors/Singular vectors as
dictionaries. Factor Analysis, Total variability space - a dictionary learning methods. Non
negative matrix factorization - a dictionary learning method.
6 Linear discriminant functions: Gradient descent procedures, Perceptron, Support vector 03
machines - a brief introduction.
7 Artificial neural networks: Multilayer perceptron – feed forward neural network. A brief 04
introduction to deep neural networks, convolutional neural networks, recurrent neural
networks.
8 Non-metric methods for pattern classification: Non-numeric data or nominal data. 03
Decision trees: Classification and Regression Trees (CART).
Page 1 of 2
w.e.f. AY 2018-19
GUJARAT TECHNOLOGICAL UNIVERSITY
Bachelor of Engineering
Subject Code: 3171613
Suggested Specification table with Marks (Theory):
Distribution of Theory Marks
R Level U Level A Level N Level E Level C Level
10 25 25 20 15 5
Legends: R: Remembrance; U: Understanding; A: Application, N: Analyze and E: Evaluate C:
Create and above Levels (Revised Bloom’s Taxonomy)
Note: This specification table shall be treated as a general guideline for students and teachers. The actual
distribution of marks in the question paper may vary slightly from above table.
Reference Books:
1) R.O.Duda, P.E.Hart and D.G.Stork, Pattern Classification, John Wiley, 2001
2) S.Theodoridis and K.Koutroumbas, Pattern Recognition, 4/e, Academic Press, 2009
3) C.M.Bishop, Pattern Recognition and Machine Learning, Springer, 2006
4) Morton Nadier and Eric Smith P., Pattern Recognition Engineering, John Wiley & Sons, New York, 1993.
5) Robert J. Schalkoff, Pattern Recognition: Statistical, Structural and Neural Approaches, John Wiley & Sons
Inc., New York, 2007.
6) Tom Mitchell, Machine Learning, McGraw-Hill
7) Tou and Gonzales, Pattern Recognition Principles, Wesley Publication Company, London, 1974.
Course Outcome:
After learning the course, the students should be able to:
Sr. CO Statement Marks
No. % Weightage
1 Know the major approaches in statistical and syntactic pattern recognition. 25
2 Understand pattern recognition theories, such as Bayes classifier, linear 25
discriminant analysis
3 Become aware of the theoretical issues involved in pattern recognition system 25
design such as the curse of dimensionality.
4 Apply pattern recognition techniques in practical problems 25
List of Experiments and Design based Problems (DP)/Open Ended Problem:
Minimum 10 practical’s based on the content of the syllabus.
One Design Based/Open Ended Problem to be assigned in a group of 3 students.
List of Open Source Software/learning website:
• https://nptel.ac.in/courses/117/105/117105101/
• https://nptel.ac.in/courses/117/108/117108048/
• http://videolectures.net/Top/Computer_Science/Machine_Learning/Pattern_Recognition/
• https://www.youtube.com/results?search_query=pattern+recognition
Page 2 of 2
w.e.f. AY 2018-19
