极大熵球面K均值文本聚类分析

计算机科学与探索 ›› 2007, Vol. 1 ›› Issue (3): 331-339.

极大熵球面K均值文本聚类分析

修宇^1,3,王士同^1,2+,朱林¹,宗成庆²

1.江南大学信息工程学院,江苏无锡 214036
2.中科院自动化研究所模式识别国家重点实验室,北京 100080
3.安徽工程科技学院计算机科学与工程系,安徽芜湖 241000

收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-10-20 发布日期:2007-10-20
通讯作者: 修宇

Maximum-entropy sphere K-means document clustering analysis

XIU Yu^1,3,WANG Shitong^1,2+,ZHU Lin¹,ZONG Chengqing²

1.School of Information Engineering,Jiangnan University,Wuxi,Jiangsu 214036,China
2.National Laboratory of Pattern Recognition,Institute of Automation,Chinese Academy of Sciences,Beijing 100080,China
3.Department of Computer Science and Engineering,Anhui University of Technology and Science,Wuhu,Anhui 241000,China

Received:1900-01-01 Revised:1900-01-01 Online:2007-10-20 Published:2007-10-20
Contact: XIU Yu

摘要/Abstract

摘要： 提出了一种基于极大熵理论的球面K均值文本聚类算法ME-SPKM。该算法利用了传统文本聚类算法SPKmeans中使用的余弦相似度度量,进而引入极大熵理论构造了适合文本聚类的极大熵目标函数。对文本数据的实验证明了极大熵球面K均值文本聚类算法取得了比传统文本聚类算法更好的聚类效果。

关键词: 极大熵, 文本聚类, 球面K均值

Abstract: A maximum-entropy version of the spherical K-means document clustering algorithm ME-SPKM is presented based on the well-known maximum-entropy. The proposed method uses the cosine similarity which is adopted by the typical text clustering algorithm SPKmeans,then constructs a maximum-entropy-based objective function. Experimental results demonstrate that the maximum-entropy spherical K-means ME-SPKM can achieve better clustering results than traditional clustering approaches in text clustering.

Key words: maximum-entropy, document clustering, spherical K-means

修宇1,3 ,王士同1,2+ ,朱林1 ,宗成庆2 . 极大熵球面K均值文本聚类分析[J]. 计算机科学与探索, 2007, 1(3): 331-339.

XIU Yu^1,3,WANG Shitong^1,2+,ZHU Lin¹,ZONG Chengqing². Maximum-entropy sphere K-means document clustering analysis[J]. Journal of Frontiers of Computer Science and Technology, 2007, 1(3): 331-339.