摘要
将核方法的思想推广到模糊C-均值算法,提出一种改进的模糊核C-均值算法,改进的模糊核C-均值算法较以前的模糊核C-均值方法有更好的鲁棒性,不但可以在有野值存在的情况下得到较好的聚类结果,而且因为放松的隶属度条件,使最终聚类结果对预先确定的聚类数目不十分敏感。改进的模糊核C-均值算法在多种数据结构条件下可以有效地进行聚类。
In this paper, the kernel method is extended to the fuzzy C-Means algorithm, and an improved fuzzy kernel C-Means algorithm is proposed, in contrast to former fuzzy kernel C-Means method, the improved fuzzy kernel C-Means algorithm has better robustness, it not only can get better clustering effect at the circumstance of outliers existing, but also can make the final clustering results less sensitive to the number of beforehand clustering data because of relaxed subjection degree condition. The results of experiments show that the improved fuzzy kernel C-Means clustering algorithm can effectively cluster on data with diversiform structures.
出处
《微电子学与计算机》
CSCD
北大核心
2006年第12期141-143,146,共4页
Microelectronics & Computer
基金
国家自然科学基金项目(10571109)
关键词
聚类分析
模糊C-均值
核方法
无监督学习
Clustering analysis, Fuzzy C-Means algorithm, Kernel-based method, Unsupervised learning
