Introduction to Pattern Recognition: A Matlab Approach

CHAPTER

2

Classifiers Based on Cost

Function Optimization

2.1 INTRODUCTION

This chapter deals with techniques and algorithms that “emancipate” from the Bayes decision theory

rationale. The focus is on the direct design of a discriminant function/decision surface that separates the

classes in some optimal sense according to an adopted criterion. The techniques that are built around the

optimal Bayesian classifier rely on the estimation of the pdf functions describing the data distribution

in each class. However, in general this turns out to be a difficult task, especially in high-dimensional

spaces. Alternatively, one may focus on designing a decision sur ...