Abstract: The fuzzy C-means clustering algorithm has the shortages including its sensitivity for data of outlier and noise and its distributed imbalanced samples，this paper presents an improved algorithm：By improving the subject function，the impacts of outlier are eliminated，and in order to differentiate the different effects of different samples for knowledge discovery，every sample holds a quantificational weight to improve clustering results of noise and distributed imbalanced samples.The experimental results show that the modified algorithm is more robust and has higher clustering accuracy.

Key words: Fuzzy C-Means（FCM）, weights, clustering

摘要： 为解决模糊C-均值（FCM）聚类算法对噪声和孤立点数据敏感、样本分布不均衡的问题，提出了具体的改进和提高的方法：改进隶属度函数，以消除孤立点对聚类结果的影响；为每个样本点赋予一个定量的权值，以区分不同的样本点对于知识发现的不同作用，改善噪音和分布不均衡的样本集的聚类结果。实验结果表明该算法具有更好的健壮性和聚类效果。

关键词: 模糊C-均值, 权值, 聚类