模糊C-均值(FCM)聚类算法的改进

Improvement of Fuzzy C-Means Clustering Algorithm

针对目前模糊C-均值聚类算法不适用于有噪声和样本不均衡等问题,借助改进算法AFCM和WFCM的思想,提出另一种新的聚类算法。它是AFCM和WAFCM结合的一种算法,但有着更好的健壮性和聚类效果。

Aiming at fuzzy C-means clustering algorithm having considerable trouble in a noisy environment and in accuracy with a large number of different sample sized clusters, a new clustering algorithm has been proposed, with the help of the thought of improvement＇s clustering algorithm-AFCM and WFCM. The algorithm is one combining AFCM and WFCM, while it is more robust and has a better clustering result than two others.