Presentation 2011-03-09
On Classification Function of Radial Basis ART Maps
Yoko ENOSAWA, Haruna MATSUSHITA, Toshimichi SAITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we present a new application of radial basis adaptive resonance theory Maps (RB-ARTMAP) to classification problems. The RB-ARTMAP can classify a set of analog input data by circle-shaped categories. The RB-ARTMAP uses a new distance that measures the similarities of an input and categories. The RB-ARTMAP has an simpler learning algorithm than other ART systems. Using basic three benchmarks, we have confirmed that our algorithm is competitive at classification function with Fuzzy-ARTMAP and support vector machine (SVM).
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Adaptive Resonance Theory(ART) / clustering / Support Vector Machine(SVM) / supervised learning
Paper # NC2010-201
Date of Issue

Conference Information
Committee NC
Conference Date 2011/2/28(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) On Classification Function of Radial Basis ART Maps
Sub Title (in English)
Keyword(1) Adaptive Resonance Theory(ART)
Keyword(2) clustering
Keyword(3) Support Vector Machine(SVM)
Keyword(4) supervised learning
1st Author's Name Yoko ENOSAWA
1st Author's Affiliation Department of Electrical and Electronics Engineering, Hosei University()
2nd Author's Name Haruna MATSUSHITA
2nd Author's Affiliation Department of Reliability-based Information Systems Engineering, Kagawa University
3rd Author's Name Toshimichi SAITO
3rd Author's Affiliation Department of Electrical and Electronics Engineering, Hosei University
Date 2011-03-09
Paper # NC2010-201
Volume (vol) vol.110
Number (no) 461
Page pp.pp.-
#Pages 4
Date of Issue