Presentation 1997/3/18
Study of a classification method by self-organizing feature map algorithm
Mizue KAYAMA, Toshio OKAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The main purpose of this research is to improve the classification performance of the competitive learning on a neural network. To fulfill this purpose, a classification algorithm was proposed, and a classifying system embedded the peculiar algorithm has been developed. We conducted the experiments for checking the validity of the result obtained from our system. In this paper, we propose the classification algorithm which has robustness to classification accuracy for noise. In order to guarantee the power mentioned above, we contrive the mechanism which decide boundary lines in competitive layer of neural network based on each weight value of all units.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Neural network / Classification / Learning for neighborhood units / Self-organizing feature map
Paper # AI96-43,KBSE96-33
Date of Issue

Conference Information
Committee AI
Conference Date 1997/3/18(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 Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Study of a classification method by self-organizing feature map algorithm
Sub Title (in English)
Keyword(1) Neural network
Keyword(2) Classification
Keyword(3) Learning for neighborhood units
Keyword(4) Self-organizing feature map
1st Author's Name Mizue KAYAMA
1st Author's Affiliation The Graduate School of Information Systems,University of Electro-Communications()
2nd Author's Name Toshio OKAMOTO
2nd Author's Affiliation The Graduate School of Information Systems,University of Electro-Communications
Date 1997/3/18
Paper # AI96-43,KBSE96-33
Volume (vol) vol.96
Number (no) 594
Page pp.pp.-
#Pages 8
Date of Issue