中文摘要：

传统的模糊C均值聚类（FCM）算法具有简单、稳定和高效等特点,但在噪声点较多的情况下容易受噪声影响,使得算法效率降低。文章结合变精度粗糙集模型,提出一种改进的FCM算法,该算法利用变精度粗糙集模型刻画不确定集合上近似集和下近似集的原理,将经过聚类算法后的类簇边缘范围中的对象根据变精度粗糙集的阈值特性划分为正域、负域、边界域三个部分,使得聚类的准确率得到提升。仿真实验结果表明该算法使得聚类结果更加清晰,在边界域较模糊的情况下聚类准确率比传统FCM算法有一定的提高。

英文摘要：

The traditional fuzzy C-means algorithm （FCM） is simple, stable and efficient, however, it is easy to be affected by the noise, which makes the efficiency of the algorithm reduced. An improved FCM algorithm based on variable precision rough sets model is introduced. The upper and lower approximation sets of the variable precision rough sets model are used in the algorithm. According to the different threshold of the variable precision rough sets, the objects in the edge range are divided into three regions, positive region, negative region and boundary region,and the accuracy of algorithm is then improveal. The experimental results show that the algorithm makes the clustering results more clear, and the clustering accuracy is better than that of the traditional FCM algorithm in the case of fuzzy boundary region.