改进的ART1网络及其在英文字符识别中的应用  

An improved ART1 neural network used in English characters recognize

作　　者：余鹏飞[1] 刘兵[2] 

机构地区：[1]云南大学信息学院,云南昆明650091 [2]云南大学软件学院,云南昆明650091

出　　处：《云南大学学报（自然科学版）》2004年第5期397-400,共4页Journal of Yunnan University(Natural Sciences Edition)

基　　金：云南省教育厅科学研究基金资助项目 ( 0 2QY0 78) .

摘　　要：通过实验发现传统的ART1人工神经网络的分类结果和输入样本的顺序有关 .分析并指出了原因 ,给出了相应的改进算法 .最后 。The traditional adaptive resonance theory 1(ART1) neural network is widely used in pattern recognition,but the classified result of ART1 is sensitive to the sequence of the input patterns.If the sequence changed,the result changed too.Its limitation is analyzed and the cause of it is that traditional ART1 neural network ignore the effect of each input pattern's zero element on the result of classify.A new improved ART1 neural network is proposed and it can solve the problem mentioned above.Experiment results show the improved ART1 neural network is effective on English characters recognition.

关 键 词：ART1网络 英文字符 模式识别 人工神经网络 前馈连接权 反馈连接权 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TP183[自动化与计算机技术—计算机科学与技术]