Attribute-driven community search

TLDR: This paper develops an efficient greedy algorithmic framework to iteratively remove nodes with the least popular attributes, and shrink the graph into an ATC, and builds an elegant index to maintain k-truss structure and attribute information, and proposes efficient query processing algorithms.

Abstract: Recently, community search over graphs has gained significant interest In applications such as analysis of protein-protein interaction (PPI) networks, citation graphs, and collaboration networks, nodes tend to have attributes Unfortunately, most previous community search algorithms ignore attributes and result in communities with poor cohesion wrt their node attributes In this paper, we study the problem of attribute-driven community search, that is, given an undirected graph G where nodes are associated with attributes, and an input query Q consisting of nodes Vq and attributes Wq, find the communities containing Vq, in which most community members are densely inter-connected and have similar attributes We formulate this problem as finding attributed truss communities (ATC), ie, finding connected and close k-truss subgraphs containing Vq, with the largest attribute relevance score We design a framework of desirable properties that good score function should satisfy We show that the problem is NP-hard However, we develop an efficient greedy algorithmic framework to iteratively remove nodes with the least popular attributes, and shrink the graph into an ATC In addition, we also build an elegant index to maintain k-truss structure and attribute information, and propose efficient query processing algorithms Extensive experiments on large real-world networks with ground-truth communities show that our algorithms significantly outperform the state of the art and demonstrates their efficiency and effectiveness