ROUGH K-MEANS CLUSTER WITH ADAPTIVE PARAMETERS

摘要：

In this paper, we firstly analyze Lingras algorithm with respect to its objective-function, numerical stability of the clusters.Then we point out its shortcoming in adjusting thethree coefficients W1, Wu and ε.To tackle this problem, arough k-means clustering method is finally presented with adaptive parameters.This algorithm is used in a testing sample and obtains a less error clustering rate.

关键词： Clustering algorithm Rough sets Rough k-means Adaptive parameters

作者: TAO ZHOU YAN-NING ZHANG HE-JING YUAN HUI-LING LU

作者单位: School of Computer, Northwestern Polytechnical University, Xian Shaanxi, 710072, China;Department o School of Computer, Northwestern Polytechnical University, Xian Shaanxi, 710072, China Department of Comp, Shaanxi University of technology, Hanzhong Shaanxi, 723000, China

会议类型: 国际会议

会议名称: 2007 International Conference on Machine Learning and Cybernetics(IEEE第六届机器学习与控制论国际会议)

会议地点: 香港

会议语种:英文

页码: 3063-3068

在线出版日期: 2007-08-19（万方平台首次上网日期，不代表论文的发表时间）