The goal of this course is to introduce the basic concepts and techniques used in the field of pattern recognition (PR). Broadly speaking, PR is a science that concerns the classification (or recognition) of measurements. It has many important applications, for example, document analysis, face recognition, fingerprint identification, speech recognition, medical diagnosis, data mining, and information retrieval.
The outline of this course is given below.
I. Pattern Recognition Overview
II. Bayesian Decision Theory
III. Supervised Learning Using Parametric Approaches
IV. Supervised Learning Using Non-parametric Approaches
V. Linear Discriminant Functions
VI. Unsupervised Learning and Clustering
VII. Syntactic Pattern Recognition
VIII. Neural Pattern Recognition
IX. Stochastic Methods
二、教 科 書：R. Duda, P. Hart, D. Stork, `Pattern Classification and Scene Analysis,` second edition, John Wiley and Sons, 2000.
Term Project: 30%