Community Detection in Bipartite Networks Using HellRank Centrality Measure
Subject Areas : ICTعلی خسروزاده 1 , علی موقر 2 , محمدمهدی گیلانیان صادقی 3 , حمیدرضا ماهیار 4
1 -
2 -
3 -
4 -
Keywords:
Abstract :
Community structure is a common and important feature in many complex networks, including bipartite networks. In recent years, community detection has received attention in many fields and many methods have been proposed for this purpose, but the heavy consumption of time in some methods limits their use in large-scale networks. There are methods with lower time complexity, but they are mostly non-deterministic, which greatly reduces their applicability in the real world. The usual approach that is adopted to community detection in bipartite networks is to first construct a unipartite projection of the network and then communities detect in that projection using methods related to unipartite networks, but these projections inherently lose information. In this paper, based on the bipartite modularity measure that quantifies the strength of partitions in bipartite networks and using the HellRank centrality measure, a quick and deterministic method for community detection from bipartite networks directly and without need to projection, proposed. The proposed method is inspired by the voting process in election activities in the social society and simulates it.
