FCM算法中参数的优选方法及实例应用
FCM算法中参数的优选方法及实例应用
【A New Validity Function for Fuzzy Clustering】
以前在 cinc 2009 发表过的一篇文章,可以自适应的给出FCM的最佳聚类数(经典的FCM算法需要预先给定聚类数目才可以)。
[img]file:///C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\TempPic\}K7))03J{JS48637EJU[UF0.tmp[/img]
======================================
A New Validity Function for Fuzzy Clustering
Wuhan, China June 06-June 07
Yang Li
Fusheng Yu
This paper first gives a new validity function for fuzzy clustering, then presents a method of the optimal selecting of the cluster number in the standard fuzzy c-means clustering algorithm, and finally outlines the fuzzy c-means clustering algorithm with parameters self-adapted. Experimental results carried on synthetic data set and data set based on actual background illustrate the performance of the new validity function and the corresponding fuzzy clustering algorithm.
ADDITIONAL INFORMATION
Index Terms: Fuzzy C-Means; fuzzy clustering analysis; cluster number; clustering validity function
Citation: Yang Li, Fusheng Yu, "A New Validity Function for Fuzzy Clustering," cinc, vol. 1, pp.462-465, 2009 International Conference on Computational Intelligence and Natural Computing, 2009
Abstract:This paper gives a method of the optimal choice for fuzzy weighting exponent m and the numbers c. Then the Fuzzy c-Means clustering algorithm with parameters self-adepted is presented in this paper. At last expermental results with artificial data and data based on actual background illustrates the effectiveness of the algorithm.