An approach for discovering user similarity in social networks based on the Bayesian network and MapReduce

摘要

Adopting Bayesian network (BN) as the effective framework for representing and inferring dependencies and uncertainties among variables, in this paper, we established a BN-based model to discover user similarities in social networks. First, we built a BN to describe the direct similarity relationships between users, called social user BN and abbreviated as SUBN. Second, we proposed a distributed storage method based on Hbase to store the SUBN and support the efficient probabilistic inferences. Consequently, we proposed a SUBN-based method to find indirect similarity relationships between users. Experimental results show the efficiency and accuracy of our method.