A Collaborative Kernel Fuzzy Clustering

摘要：

Collaborative clustering implementing on several subsets can be processed together with an objective function, which improves the clustering performance by collaborating partition matrices among different feature subsets. Kernel-based clustering can map the observed data to a higher dimensional feature space with a kernel function, which achieves more clearly classification by enlarging the difference among samples. Here an improved algorithm of collaborative kernel fuzzy c-means clustering (CKFCM) was proposed, in which the idea of collaboration was introduced into kernel fuzzy c-means clustering (KFCM). Taking both the advantages of kernel methods and collaboration, CKFCM makes full use of the collaborative relation among different subsets based on KFCM. The results obtained on the benchmark datasets show that CKFCM is more effective than the two basic algorithms.

关键词： kernel methods fuzzy c-means clustering kernel fuzzy c-means clustering collaborative clustering

作者: GAO Cui-Fang WU Xiao-Jun

作者单位: School of Information Engineering, Jiangnan University, Wuxi, 214122

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 763-766

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）