Constraint Rules and Matching Micro-clusters Based Affinity Propagation Clustering Algorithm

Li-min WANG¹, You ZHOU², Xu-ming HAN³, Yi-zhang WANG⁴*, Jing-lin YU¹, Shuai WANG⁵
¹School of Internet Finance and Information Engineering, Guangdong University of Finance,
Guangzhou, 510521, China
wlm_new@163.com, yujinglin@gduf.edu.cn
² College of Computer Science and Technology, Jilin University, 130012, Changchun, China
zyou@jlu.edu.cn
³ College of Information Science Technology, Jinan University, Guangzhou, 510632, China
hanxvming@163.com
⁴ College of Information Engineering, Yangzhou University, 225127, Yangzhou, China
wyzhang_new@sina.com (*Corresponding author)
⁵ Business School, Jilin University, 130012, Changchun, China
wang.s@qq.com

Abstract: The performance of original affinity propagation (AP) clustering algorithm is greatly influenced by an important parameter: preference (median of similarities between data points), and it may be difficult to identify complex structure data. To address the afore-mentioned issues, this paper proposes two novel methods namely the constraint rules-based affinity propagation (CRAP) and matching micro-clusters hierarchical clustering algorithm (MMHC). The CRAP algorithm can obtain better results by searching the optimal preference value by means of the constraint rules-based search algorithm (CRS). The MMHC algorithm initially takes results of AP as micro-clusters, then they are matched in order to achieve the right partitions of complex structure data. Experimental results demonstrate that the improved clustering algorithm performs better than AP.

Keywords: Affinity propagation, Constraint rules, Micro-clusters hierarchical clustering.

>>FULL TEXT: PDF

CITE THIS PAPER AS:
Li-min WANG, You ZHOU, Xu-ming HAN, Yi-zhang WANG, Jing-lin YU, Shuai WANG, Constraint Rules and Matching Micro-clusters Based Affinity Propagation Clustering Algorithm, Studies in Informatics and Control, ISSN 1220-1766, vol. 29(3), pp. 353-362, 2020. https://doi.org/10.24846/v29i3y202008