Fuzzy discriminant analysis in fuzzy groups

Junzo Watada, Hideo Tanaka, Kiyoji Asai

Research output: Contribution to journalArticlepeer-review

24 Citations (Scopus)

Abstract

This paper deals with discriminant problems to classify samples with fuzzy multi-attribute into fuzzy groups. In this problem, the objective is to determine the linear discriminant function that provides the maximum separation of fuzzy groups in a real space. For this purpose, we define the fuzzy variance ratio and employ maximization of the fuzzy variance ratio as a criterion. Furthermore, a partial correlation coefficient in the fuzzy groups is defined to estimate the influence of each attribute itself on the discrimination between fuzzy groups.

Original languageEnglish
Pages (from-to)261-271
Number of pages11
JournalFuzzy Sets and Systems
Volume19
Issue number3
DOIs
Publication statusPublished - 1986
Externally publishedYes

Keywords

  • Classification
  • Fuzzy discriminant analysis
  • Fuzzy mean
  • Fuzzy partial correlation coefficient
  • Fuzzy variance

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Information Systems and Management
  • Statistics, Probability and Uncertainty
  • Electrical and Electronic Engineering
  • Statistics and Probability

Fingerprint Dive into the research topics of 'Fuzzy discriminant analysis in fuzzy groups'. Together they form a unique fingerprint.

Cite this