Multi-relational data semi-supervised K-means clustering algorithm

Authors:
Zhanguo Xia;Wentao Zhang;Shiyu Cai;Shixiong Xia
Affiliations:
School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu, China;School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu, China;School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu, China;School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu, China
Venue:
AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part I
Year:
2011

Citing 2
Cited 0

CrossClus: user-guided multi-relational clustering

Data Mining and Knowledge Discovery
Stochastic and iterative techniques for relational data clustering

Stochastic and iterative techniques for relational data clustering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Based on the traditional K-means clustering algorithm, a new semi-supervised K-means clustering algorithm (MMK-means) is proposed in this paper, in which use semi-supervised learning method to solve the problem of clustering on multi-relational data set. In order to improve the quality of clustering results, the algorithm making full use of the various relationships between objects and attributes to guide the choice of the marked data, and use these relationships to the initial center of clusters. Experimental results on Financial Data database verify the accuracy and effectiveness of the algorithm.