优化初始中心的模糊C-均值(FCM)算法  

Optimized initial centers for fuzzy C-means algorithm

作　　者：江克勤[1] 施培蓓[2] 

机构地区：[1]安庆师范学院计算机与信息学院,安徽安庆246011 [2]合肥师范学院现代教育技术中心,安徽合肥230601

出　　处：《合肥工业大学学报（自然科学版）》2009年第5期762-764,768,共4页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(60773128);安庆市科学技术局重点科技计划资助项目(2003-50)

摘　　要：文章针对模糊C-均值(FCM)算法对初始中心敏感的缺点,通过计算样本的权重,提出基于权重的初始中心选取算法,选取有代表性的样本作为初始聚类中心,给出优化初始聚类中心的FCM算法。与传统算法比较,改进算法可以得到较稳定的结果,并且提高了聚类的准确率;实验证明了改进算法的有效性。The fuzzy C-means（FCM） algorithm is sensitive to initial clustering centers. This paper introduces an initial center selecting algorithm based on the weight through computing the sample weight and selecting representative samples as initial clustering centers. The optimized initial centers are presented for the FCM algorithm. In comparison with traditional algorithms, the presented algorithm can get more steady cluster results, and the clustering accuracy is also improved. Experiment shows the effectiveness of the improved algorithm.

关 键 词：聚类 模糊C均值 权重 聚类中心 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]