基于改进仿射传播聚类的文本自适应模糊聚类  

Research on Adaptive Text Fuzzy Clustering Based on Improved Affinity Propagation Clustering

作　　者：田胜利[1] 熊德兰[1] 

机构地区：[1]许昌学院计算机科学与技术学院,河南许昌461000

出　　处：《小型微型计算机系统》2010年第11期2287-2290,共4页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60475040)资助;河南省教育厅自然科学基金项目(2008B520031)资助

摘　　要：仿射传播聚类有与FCM、K-均值聚类方法等效的聚类功能、并且不需要事先确定初始聚类中心.但它也有一个缺陷就是有时在聚类过程中聚类的数目会出现摇摆,最终影响聚类结果.针对此提出了动态调整阻尼系数以消除摇摆及当调整方法失效时自动跳出摇摆等技术,成功改进了仿射传播聚类算法.并把改进后的仿射传播聚类应用于自适应文本模糊聚类中,取得了较好的聚类效果.Affinity propagation （AP）clustering has a same ability with FCM, K-means clustering and without selected the initial cluster center and the cluster number, but AP clustering has a limitation：the number of cluster often swings when clustering even impact the result of convergence. For the paper proposes an improved Affinity propagation （imAP） clustering to overcome this limitation suecessfully,that dynamic adjustment damping factors eliminate sway and escaping from sway when the adjustment method failed. Applying imAP to adaptive text fuzzy clustering. The results of experment indicates that imAP obtains good clustering effect, and indicate the results of the experiments.

关 键 词：自然语言处理 文本聚类 模糊聚类 改进的仿射传播聚类 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]