Comparison of Classifiers in Small Training Sample Size Situations for Pattern Recognition

Yoshihiko HAMAMOTO  Shunji UCHIMURA  Shingo TOMITA  

IEICE TRANSACTIONS on Information and Systems   Vol.E77-D   No.3   pp.355-357
Publication Date: 1994/03/25
Online ISSN: 
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Image Processing, Computer Graphics and Pattern Recognition
pattern recognition,  classifier,  error probability,  small training sample size,  

Full Text: PDF>>
Buy this Article

The main problem in statistical pattern recognition is to design a classifier. Many researchers point out that a finite number of training samples causes the practical difficulties and constraints in designing a classifier. However, very little is known about the performance of a classifier in small training sample size situations. In this paper, we compare the classification performance of the well-known classifiers (k-NN, Parzen, Fisher's linear, Quadratic, Modified quadratic, Euclidean distance classifiers) when the number of training samples is small.