基于稀疏表示的近邻传播聚类算法  

Affinity Propagation Clustering Based on Sparse Representation

作　　者：胡晨晓[1] 邹显春[1] 陈武[1] 杨阳[1] 

机构地区：[1]西南大学计算机与信息科学学院,重庆400715

出　　处：《西南大学学报（自然科学版）》2014年第5期220-224,共5页Journal of Southwest University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61003203)

摘　　要：借助稀疏表示具有能较好刻画样本之间相似度的特点,提出一种基于稀疏表示的近邻传播聚类算法.仿真实验表明,本聚类算法较基于其它距离度量的算法能获得更好的聚类效果.Affinity propagation clustering is an efficient clustering algorithm based on the information propagation between neighborhood nodes.It does not require the input distance matrix to be symmetric nor each element of the matrix to be positive.Its performance is largely dependent on the distance metrics, thus it is possible to boost its performance by adapting more reliable distance metrics.Given the advantages of sparse representation in more faithful measuring the similarity between two samples,we propose an affinity clustering algorithm based on sparse representation.The experimental study on several datasets shows that the proposed algorithm performs better than the algorithms based on other distance metrics.

关 键 词：稀疏表示 近邻传播 聚类 距离度量 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]