RBF神经网络在Web分类挖掘中的应用  

Applying RBF Neural Network to Web Classification Mining

作　　者：艾尼瓦尔.努尔买买提 

机构地区：[1]和田师范专科学校,新疆和田848000

出　　处：《长春师范学院学报（自然科学版）》2009年第2期32-35,共4页Journal of Changchun Teachers College

摘　　要：为了有效地组织和分析大量WEB信息,本文设计了WEB分类发掘系统。BP网络应用广泛,但也有许多不足之处。因此,提出了用RBFNN(径向基函数神经网络)分类WEB页面信息的方法。分类系统框架主要包括RBF(径向基函数)分类器、评估模型及数据预处理。用Macro-Fi作为分类效果的评估标准,实验结果证实,RBFNN分类器比BPNN分类器更有效、更准确。并且用相同的分类器对不同类进行分类,分析了不同分类精度。To effectively organize and analyze massive Web information, this paper has designed a Web classification mining system. BP network has lots of disadvantages, so this paper has proposed a method that uses RBFNN （Radial Basis Function Neural Network） to classify the text information in Web pages. In this paper, the model of classification system mainly includes RBF （Radial Basis Function） classifier,estimate model and data pretreatment. Using Macro- Fi as evaluation standard of classification performance, experimental results can verify that RBFNN classification has better classification accuracy and is more efficient than BPNN algorithm. This article also analyzes the results of different classification accuracy, using the same classifier to classify different classes.

关 键 词：BP网络 WEB分类 径向基函数 特征选择 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]