一种新的基于Fuzzy c-means的高效自适应截集算法

A New Effective Section Set Adaptive Algorithm Based on Fuzzy c-means

提出了一种新的模糊聚类方法-自适应截集算法。该方法克服了聚类数目c要求预先确定、局部最优、分类不确定等弱点,对算法结构加以改进,增加聚类有效性问题的分析,在聚类过程中可动态调整聚类数目。针对时间消耗问题,利用模糊截集提高分类识别的速度。经实验表明,本算法可以提高聚类算法的可靠程度和分类识别的正确性。

This paper presents a new section set adaptive FCM algorithm. The algorithm solves the weakness of local optimality, unsure classification and numbers of clustering ascertained previously. And it improves on the architecture of FCM＇s algorithm, enhances the analysis for effective clustering. During the process of cluster, it may adjust numbers of cluster dynamically. Finally, it uses the method of section set decreasing the time of classification. By experiments, the algorithm can improve dependability of clustering and correctness of classification.