一种对BBS语料进行话题提取的聚类算法  

A CLUSTERING ALGORITHM FOR TOPIC DETECTION AND TRACKING IN BBS

作　　者：李卓尔[1] 胡运发[1] 

机构地区：[1]复旦大学计算机信息与技术系,上海200433

出　　处：《计算机应用与软件》2008年第8期1-3,共3页Computer Applications and Software

基　　金：国家自然科学基金项目(60173027)

摘　　要：基于BBS语料的话题提取主要是从大量的BBS论坛讨论信息中,将正在或近期讨论的各种话题提取出来。在自主开发的一套话题提取系统中采用了一个原始聚类算法,能够对真实的BBS语料进行有效话题提取。随后将语料中的关联信息引入到原始聚类算法中进行改进,提高了算法的性能,取得了良好的效果。Topic detection and tracking in BBS is mainly to detect the topics being discussed or had been discussed recently from a host of related information from BBS. The topic detection system with a clustering algorithm is effective for topic detection and tracking in BBS. Im- provement is made on the related information of BBS,which is introduced to the original clustering algorithm. The performance of the algorithm is improved, and better results are achieved.

关 键 词：BBS 话题提取 关联信息 聚类算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TS744[自动化与计算机技术—计算机科学与技术]