Consistent parameter estimation in general stochastic block models with overlaps

This paper considers the parameter estimation problem in Stochastic Block Model with Overlaps (SBMO), which is a quite general instance of random graph model allowing for overlapping community structure. We present the new algorithm successive projection overlapping clustering (SPOC) which combines the ideas of spectral clustering and geometric approach for separable non-negative matrix factorization. The proposed algorithm is provably consistent under SBMO with general conditions on the parameters of the model. SPOC is also shown to perform well experimentally in comparison to other algorithms.