A new method to calculate weights of attributes in spectral clustering algorithms

doi:10.1109/ICM.2011.213

LZUFE-KMS

	A new method to calculate weights of attributes in spectral clustering algorithms
	Li, Zhendong 1; Sun, Wei 2
	2011
会议名称	2011 International Conference on Information Technology, Computer Engineering and Management Sciences, ICM 2011
会议录名称	Proceedings - 2011 International Conference of Information Technology, Computer Engineering and Management Sciences, ICM 2011
卷号	2
页码	58-60
会议日期	2011-09-24 - 2011-09-25
会议地点	Nanjing, Jiangsu, China
出版者	IEEE Computer Society
摘要	In order to improve the spectral clustering algorithms, there is one algorithm that when calculate the distance formula as similar matrix of spectral clustering, considering the weights of different attributes, but the methods to calculate weights of attributes are of significant limitation, affecting the clustering results. Consequently, inspired by the MIV method, we construct a method named MDIV that is to calculate the weights of attributes by the average distance change from attribute values change. For illustration, two UCI data sets are utilized to show the feasibility of the MDIV in calculating weights of attributes of spectral clustering algorithms. Empirical results show that attributes weights determined by MDIV improve the spectral clustering algorithm more significantly than by factor analysis. That is to say, weights of attributes calculated by the MDIV method are more realistic than other common methods. The MDIV method can effectively reduce human intervention in calculating weights of attributes, and obtain more objective results to improve accuracy of clustering. So, to a certain extent, MDIV can be a general method to calculate weights of attributes in clustering. © 2011 IEEE.
关键词	Factor analysis Multivariant analysis Attribute values Average Distance Clustering results Distance formula Human intervention MIDV Spectral clustering Spectral clustering algorithms
DOI	10.1109/ICM.2011.213
收录类别	EI
语种	英语
EI入藏号	20120614754137
文献类型	会议论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/9836
专题	兰州财经大学
作者单位	1.School of Information Engineering, Lanzhou University of Finance and Economics, Lanzhou, China; 2.School of Statistics, Lanzhou University of Finance and Economics, Lanzhou, China
推荐引用方式 GB/T 7714	Li, Zhendong,Sun, Wei. A new method to calculate weights of attributes in spectral clustering algorithms[C]:IEEE Computer Society,2011:58-60.