一种改进的粗K均值聚类算法  

An improved rough-means clustering algorithm

作　　者：王莉[1,2] 周献中[1] 沈捷[2] 

机构地区：[1]南京大学工程管理学院,南京210093 [2]南京工业大学自动化与电气工程学院,南京210009

出　　处：《控制与决策》2012年第11期1711-1714,1719,共5页Control and Decision

基　　金：国家自然科学基金项目(70971062);东南大学复杂工程系统测量与控制教育部重点实验室开放课题(2010A004)

摘　　要：Lingras提出的粗K均值聚类算法易受随机初始聚类中心和离群点的影响,可能出现一致性和无法收敛的聚类结果.对此,提出一种改进的粗K均值算法,选择潜能最大的K个对象作为初始的聚类中心,根据数据对象与聚类中心的相对距离来确定其上下近似归属,使边界区域的划分更合理.定义了广义分类正确率,该指标同时考虑了下近似集和边界区域中的对象,评价算法性能更准确.仿真实验结果表明,该算法分类正确率高,收敛速度快,能够克服离群点的不利影响.Rough-means clustering algorithm proposed by Lingras is sensitive to the initial centers of the cluster and outliers and may result in identical clustering and non-convergence.In this paper,an improved rough-means clustering algorithm is proposed.The objects with maximum potentials are chosen as initial centers.The absolute distance between object and center of clusters is considered to decide whether a data object belongs to the lower or upper approximation set of a cluster,so the division of boundary area is more reasonable.General classification accuracy considering the objects in lower approximation set and boundary area is defined for rough-means clustering algorithm,and it is more appropriate for evaluating rough means clustering.The simulation results show that,the proposed algorithm has the advantages of high classification accuracy and fast convergence,and can also avoid the bad influence of outlier.

关 键 词：聚类 粗糙集 粗k均值 广义分类正确率 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]