On rough set based fuzzy clustering for graph data

Data clustering refers to partition the original data set into some subsets such that every vertex belongs to one or more subsets at the same time. For graph data that composed by attribute information of vertices as well as structural information between vertices, how to make an efficient clustering is not an easy thing. In this paper, we propose a novel method of how to partition graph data into some overlapping subgraph data in aspect of rough set theory. At first, we introduce a detailed description about the global similarity measurement of vertices. After that, an objective-function oriented optimization model is constructed in terms of updating fuzzy membership degree and cluster center that based on the theory of rough set. Obviously, the determined cluster is no longer a fuzzy set, but a rough set, that is to say, the cluster is expressed by the upper approximation set and lower approximation set. Finally, eleven real-world graph data and four synthetic graph data are applied to verify the validity of the proposed fuzzy clustering algorithm. The experimental results show that our algorithm is better than existing clustering approach to some extent.

Automated summary

This paper introduces a novel method of graph data clustering based on rough set theory. The method partitions the graph data into overlapping subgraph data by constructing an optimization model and updating fuzzy membership degree and cluster center. Experimental results demonstrate that the proposed method outperforms existing clustering approaches to some extent.