DocumentCode
2845265
Title
An Improved ART1 neural network algorithm for character recognition
Author
Li, Peng ; Xianxi, Ma
Author_Institution
Sch. of Commun. & Control Eng., Jiangnan Univ., Wuxi, China
fYear
2010
fDate
26-28 May 2010
Firstpage
2946
Lastpage
2949
Abstract
The paper indicates the shortage of standard ART1 neural network, and an improved calculating method of similarity is presented. The corresponding place value of two vectors at the same time is considered in this method. The method avoids the different result of ART1 neural network because of inputting different sequence. In order to solve the pattern excursion problem of ART1 neural network, the principle of minority subordinate to majority is proposed to reduce the appeared problem. They improve the applicative effect of ART1 neural network.
Keywords
ART neural nets; character recognition; vectors; ART1 neural network; adaptive resonance theory; character recognition; inputting different sequence; minority subordinate principle; pattern excursion; vector; Biological neural networks; Character recognition; Clustering algorithms; Control engineering; Humans; Neural networks; Pattern recognition; Stability; Switches; Testing; ART1 neural network; Character process; Pattern recognition;
fLanguage
English
Publisher
ieee
Conference_Titel
Control and Decision Conference (CCDC), 2010 Chinese
Conference_Location
Xuzhou
Print_ISBN
978-1-4244-5181-4
Electronic_ISBN
978-1-4244-5182-1
Type
conf
DOI
10.1109/CCDC.2010.5498672
Filename
5498672
Link To Document