FCM算法中权重指数m的研究
Study of Weighting Exponent m in a FCM Algorithm
模糊c均值算法(FCM)是最常用的聚类算法之一。使用模糊c均值算法时，选取恰当的权重指数（m）是非常重要的。本文通过推导证明，得出权重指数m不仅决定FCM算法的聚类模糊程度和样本在类间的分享程度，而且还影响目标函数的凹凸性及算法的收敛性
he fuzzy c-means algorithm (FCM) is one of widely used clustering algorithms. It is very important to select an appropriate fuzziness index m when implementing the FCM. This paper proved from the deduction that the Weighting exponent m not only decide the clustering fuzzy degree of the FCM algorithm and the sharing degree of Exemplars between classes，but also affect the concavity- convexity of the object function and the convergence of the algorithm.
许磊、吴晓娟
计算技术、计算机技术
权重指数硬C均值算法模糊c均值算法划分熵
Weighting exponentHCM algorithmFCM algorithmPartition entropy
许磊,吴晓娟.FCM算法中权重指数m的研究[EB/OL].(2006-03-21)[2025-08-16].http://www.paper.edu.cn/releasepaper/content/200603-368.点此复制
评论