Fuzzy C-Means Clustering Algorithm Based on Incomplete Data
Zhiping Jia, Zhiqiang Yu, Chenghui Zhang · 2006
In order to solve the problem that the traditional fuzzy c-means(FCM) clustering algorithm can not directly act on incomplete data, a modified algorithm IDFCM(Incomplete Data FCM) based on the FCM algorithm is proposed. The IDFCM algorithm takes the percentage of incomplete data in datasets and its effect on clustering analysis into consideration. Finally, the experimental clustering results of IRIS data and mobile distributed inspected data of the ocean are given, which can clearly prove that the IDFCM algorithm is very efficient for clustering incomplete data.