一种具有自适应参数的基于密度加权的粗糙K-均值算法

Rough K-means clustering algorithm with self-adaptive parameter and weighted-density

针对原始粗糙K-均值聚类算法采用固定权重和阈值的缺陷,提出一种改进的粗糙K-均值算法。改进的算法根据K-均值聚类算法的特点,在基于密度加权的K-means算法基础上,对固定经验权重和固定阈值这两种参数进行改进,提出了一种自适应参数方法。实验结果表明,该算法降低了迭代次数,聚类结果更为精确。

It proposed an improved Rough K-means Clustering algorithm to overcome the defect of fixed weight and threshold in the original Rough K-means clustering algorithm.Based on the feature of K-means algorithm,the improved Rough K-means Clustering algorithm with weighted-density integrated self-adaptive weight and threshold as well.The experiments showed that the new algorithm reduced the iteration times and improved the clustering accuracy.