基于主成分分析和变异系数的改进的亲和力传播聚类算法

549

Copyright © 2014 Nescience Enterprises Ltd.

An Improved Affinity Propagation

Clustering Algorithm Based on

Principal Component Analysis and

Variation Coefficient

Limin Wang

School of Management Science and Information Engineering,

Jilin University of Finance and Economics, Changchun, Jilin, China

Institute of Software,

Changchun University of Technology, Changchun, Jilin, China

E-mail: hanxvming@163.com

*Corresponding author: hanxvming@163.com

Na Huang

School of Information Management and Engineering,

Shanghai University of Finance and Economics, Shanghai, China

E-mail:sapruna@126.com

Xintong Guo

School of Management Science and Information Engineering,

principal component analysis into AP algorithm and combined the algorithm with the

coefficient of variation. Initially, a dimension reduction was performed on the original data

with a focus on keeping most information. After, the improved AP algorithm was applied

for clustering. The clustering experiments were done by using the Wine, Pima and

Ionosphere datasets in the UCI database. The simulation results show that the proposed

algorithm has superior clustering performance when compared to the traditional AP

algorithm. In addition, the proposed algorithm was applied to the clustering analysis for the

listed companies, and some satisfactory results were also obtained. This method

unmistakably has great application potential; for instance, it is enriching studies on

intelligence theory, providing a novel reference tool to assist governments in making

School of Management Science and Information Engineering, Jilin University of Finance