本文提出了一个在模糊聚类中判别聚类有效性的新指标。该指标可有效地对类间有交叠或有多孤立点的情况做出准确的判定。文中基于模糊C-均值聚类算法(FCM),应用多组的测试数据对其进行了性能分析,并与当前较广泛使用且较具代表性的某些相关指标进行了深入的比较。实验结果表明,该指标函数的判定性能是优越的,它可以自动地确定聚类的最佳个数。 In this paper, we propose a new index to identify the validity of clustering in fuzzy clustering. This indicator can effectively make accurate judgments about whether there is overlapping or more isolated points between classes. Based on fuzzy C-means clustering algorithm (FCM), this paper analyzes the performance of multi-group FCM by using multiple sets of test data, and makes an in-depth comparison with some of the more commonly used and representative indicators. Experimental results show that the performance of the indicator function is superior, which can automatically determine the optimal number of clusters.