一种基于语义相似度的群智能文本聚类的新方法  

New method of hybrid intelligent text clustering based on semantic similarity

作　　者：陶红[1] 周永梅[1] 高尚[1] 

机构地区：[1]江苏科技大学计算机科学与工程学院,江苏镇江212003

出　　处：《计算机应用研究》2012年第2期482-484,532,共4页Application Research of Computers

基　　金：人工智能四川省重点实验室开放基金资助项目(2009RY001);"青蓝工程"资助项目

摘　　要：针对基于VSM(vector space model)的文本聚类算法忽略了词之间的语义信息和各维度之间的关系,导致文本的相似度计算不够精确,提出了一种基于语义相似度的群智能文本聚类的新方法。该方法融合了模拟退火算法的全局搜索和蚁群算法的正反馈能力。其思路是,首先从语义上分析文本,利用K-均值算法进行文本聚类,再根据K-均值算法的结果,使用蚁群和模拟退火算法进行调整聚类。测试结果表明这种算法能够提高聚类精度和召回率,也验证了混合算法的正确性。The problem with the text clustering algorithm based on vector space model（VSM） is that semantic information between words and the link between the various dimensions are overlooked,resulting in inaccuracy in the text similarity calculation,this paper proposed a hybrid intelligent algorithm based on computing the text semantic similarity.This algorithm combined the good global search capability of simulated annealing algorithm and the good positive feedback ability of ant colony algorithm.It extended the algorithm to analyze the text according to its semantic,then used K-means clustering to seed the initial solution and the ant colony algorithm and simulated annealing algorithm to adjust the initial cluster.Through the result,this algorithm can improve the clustering precision and recall rate and the efficiency of the hybrid algorithm is verified.

关 键 词：文本聚类 语义相似度 K-均值算法 蚁群算法 模拟退火算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]