Project/Area Number |
11680450
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
社会システム工学
|
Research Institution | Okayama Prefectural University |
Principal Investigator |
KANAGAWA Akihiro Faculty of Computer Science and System Engineering, Associate Professor, 情報工学部, 助教授 (70204534)
|
Co-Investigator(Kenkyū-buntansha) |
辛 忠 岡山県工業技術センター, 室長
KAWABATA Hiroaki Faculty of computer Science and System Engineering, Professor, 情報工学部, 教授 (70081271)
TAKAHASHI Hiromitsu Faculty of computer Science and System Engineering, Professor, 情報工学部, 教授 (30109889)
ZHANG Zhong Department of System Engineering, Industrial Technology Center of Okayama Prefecture
章 忠 岡山県工業技術センター, 室長
|
Project Period (FY) |
1999 – 2000
|
Project Status |
Completed (Fiscal Year 2000)
|
Budget Amount *help |
¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2000: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1999: ¥1,900,000 (Direct Cost: ¥1,900,000)
|
Keywords | Associative memory / Classification / Cellular neural networks / Associatron / Bitmap pattern / Soft computing / ニューラルネットワーク / バイナリーニューラルネット / ウェーブレット変換 / バイナリニューラルネット |
Research Abstract |
This investigation deals with classification problems, such as diagnosis, in which classes are defined by categorical forms. Recently, soft computing techniques including learning algorithm in neural networks and fuzzy expert systems are applied to the classification problems. We have already presented a prototype of associative classification model using cellular neural networks(CNN). That is categories or classes are expressed by bitmap patterns consist of pixels which have three levels. Classification is done by association from the given data expressed by bitmap pattern. The problem is, firstly, which associative memory is suit for this classification method. We investigate two kind of associative memories, one is the associatron, and the other is extended CNN.The associatron is a model for a neural network proposed by Nakano. It embeds Hebb's law using a method to strengthen the synapse connection that interprets that the synapse connection between simultaneously excited cells grows stronger. Associative classification using associatron yields an outstanding result in the problem of iris. On the contrary, extended CNN shows remakable effects for diagnosing of liver disease.
|