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.
出处
《西南大学学报（自然科学版）》
CAS
CSCD
北大核心
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)
关键词
稀疏表示
近邻传播
聚类
距离度量
sparse representation
affinity propagation
clustering
distance metric
