| Abstract: |
COURSE OUTLINE
Introduction and mathematical preliminaries - What is pattern recognition?, Clustering vs. Classification; Applications; Linear Algebra, vector spaces, probability theory, estimation techniques.
Classification: Bayes decision rule, Error probability, Error rate, Minimum distance classifier, Mahalanobis distance; K-NN Classifier, Linear discriminant functions and Non-linear decision boundaries.
Fisher’s LDA, Single and Multilayer perceptron, training set and test sets, standardization and normalization.
Clustering: Different distance functions and similarity measures, Minimum within cluster distance criterion, K-means clustering, single linkage and complete linkage clustering, MST, medoids, DBSCAN, Visualization of datasets, existence of unique clusters or no clusters.
Feature selection: Problem statement and Uses, Probabilistic separability based criterion functions, interclass distance based criterion functions, Branch and bound algorithm, sequential forward/backward selection algorithms, (l,r) algorithm.
Feature Extraction: PCA, Kernel PCA.
Recent advances in PR: Structural PR, SVMs, FCM, Soft-computing and Neuro-fuzzy.
|
Electronics & Communication Engineering
Pattern Recognition
