APPROXIMATE REASONING APPROACH TO PATTERN-RECOGNITION

Authors
RAY KS GHOSHAL J
Citation
Ks. Ray et J. Ghoshal, APPROXIMATE REASONING APPROACH TO PATTERN-RECOGNITION, Fuzzy sets and systems, 77(2), 1996, pp. 125-150
Citations number
10
Categorie Soggetti
Computer Sciences, Special Topics","System Science",Mathematics,"Statistic & Probability",Mathematics,"Computer Science Theory & Methods
Journal title
Fuzzy sets and systems → ACNP
ISSN journal
01650114
Volume
77
Issue
2
Year of publication
1996
Pages
125 - 150
Database
ISI
SICI code
0165-0114(1996)77:2<125:ARATP>2.0.ZU;2-R
Abstract
Approximate reasoning approach to pattern recognition consists of ling uistic rules tied together by means of two concepts: fuzzy implication s and a compositional rule of inference. In this paper, first we study the applicability of different interpretations of fuzzy implication t o pattern recognition problem and subsequently compare their performan ces over a set of synthetic data. Finally, we use the most applicable (according to our study) interpretations of implication for vowel reco gnition of three different Indian languages and we obtain very promisi ng results.