首页 | 本学科首页   官方微博 | 高级检索  
     检索      

Application of Partially Connected Neural Network
作者姓名:李刚  赵林
作者单位:Glorious Sun School of Business & Management,Donghua University, Shanghai 200051 , China
摘    要:This paper focuses mainly on application of Partially Connected Backpropagation Neural Network (PCBP) instead of typical Fully Connected Neural Network (FCBP). The initial neural network is fully connected, after training with sample data using cross-entropy as error function, a clustering method is employed to cluster weights between inputs to hidden layer and from hidden to output layer, and connections that are relatively unnecessary are deleted, thus the initial network becomes a PCBP network. Then PCBP can be used in prediction or data mining by training PCBP with data that comes from database. At the end of this paper, several experiments are conducted to illustrate the effects of PCBP using Iris data set.

关 键 词:FCBP  PCBP  神经网络  计算机技术
文章编号:1672-5220(2007)02-0260-04
修稿时间:2006-08-20

Application of Partially Connected Neural Network
LI Gang,ZHAO Lin.Application of Partially Connected Neural Network[J].Journal of Donghua University,2007,24(2):260-263.
Authors:LI Gang  ZHAO Lin
Abstract:This paper focuses mainly on application of Partially Connected Backpropagation Neural Network (PCBP) instead of typical Fully Connected Neural Network (FCBP). The initial neural network is fully connected, after training with sample data using cross-entropy as error function, a clustering method is employed to cluster weights between inputs to hidden layer and from hidden to output layer, and connections that are relatively unnecessary are deleted, thus the initial network becomes a PCBP network.Then PCBP can be used in prediction or data mining by training PCBP with data that comes from database. At the end of this paper, several experiments are conducted to illustrate the effects of PCBP using Iris data set.
Keywords:neural network  FCBP  PCBP  Pruning
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号