Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (4): 170-172.

• 数据库与信息处理 • Previous Articles     Next Articles

The Research and Implementation of Web Text Clustering Algorithm WTCA

  

  • Received:2006-09-12 Revised:1900-01-01 Online:2007-02-01 Published:2007-02-01

Web文本聚类算法WTCA的研究与实现

郑煜 钱榕   

  1. 东北林业大学数学系 北京科技大学信息工程学院
  • 通讯作者: 钱榕

Abstract: In this paper , we present a new algorithm of Web text clustering mining —— WTCA. This algorithm includes the training stage and the clustering stage of SOM network. It can distinguish the most meaningful features from the Concept Space without the evaluation function. The algorithm has been applied to the Modern Long-distance Education Net. It can automatically congregate the text information of education field, which is collected from education sites and help people to browse the important information quickly by information navigation mechanism and acquire useful knowledge.

Key words: Web Text Mining, Text Clustering, nonstructural data mining, Self-Organization Feature Mapping

摘要: 本文提出了一种新的Web文本聚类算法WTCA ——基于自组织特征映射神经网络(SOM)的聚类算法。该算法分为训练SOM网络及聚类分析两个阶段,具有自稳定性,无须外界给出评价函数;能够识别概念空间中最有意义的特征,抗噪音能力强。该算法应用到现代远程教育网,可以对各类远程教育站点上收集的文本资料信息自动进行聚类分析;从海量Web文本信息源中快速有效地获取重要的知识。

关键词: Web文本挖掘, 文本聚类, 非结构化数据挖掘结构模型, 自组织特征映射