E E 666 - Pattern Recognition

Decision-theoretic pattern recognition and classification. Supervised learning and training algorithms, perceptions, reward and punishment, potential functions, linear discriminants. Bayesian learning, parametric and nonparametric classification, Bayes and Fisher classifiers. Unsupervised learning and clustering; maximum-distance, Kmeans, and Isodata algorithms, graph-theoretic approach. Feature selection through clustering transformation, entropy minimization, Karhunen-Loeve expansion. Principles of syntactic pattern recognition; formal language theory, recognition grammars, learning, and geometrical inference.

Credits: (3)

Ohio University

Ohio University

E E 666 - Pattern Recognition