Abstract: Fuzzy C-Means(FCM) clustering algorithm can not deal with updated data, so this paper presents Incremental Clustering Algorithm Based on Adaptive FCM namely AIFCM. It combines the density with data set, and can automatically decide the prototypes of the clusters and the number of the clusters when the cluster is splitting. It can deal with updated data filter the outliers and satisfy with the requirement of users.

Key words: cluster analysis, Fuzzy C-Means(FCM) algorithm, incremental clustering, Adaptive Incremental Fuzzy C-Means(AIFCM) algorithm

摘要： 针对模糊C-均值(FCM)算法不能很好地处理更新数据的缺点，提出基于FCM的自适应增量式聚类算法AIFCM。该算法结合密度和集合的思想，给出一种自动确定聚类初始中心的方法，能在聚类过程中动态改变聚类结果数，改善聚类的质量，减少人为的主观因素，获得比较符合用户需求的聚类结果，并能在原有聚类结果的基础上简单有效地处理更新数据，过滤噪声数据，较好地避免大量重复计算。

关键词: 聚类分析, 模糊C-均值算法, 增量式聚类, AIFCM算法

CLC Number: 

• TP301.6