Mixed membership distribution-free model

arXiv:2112.04389v5 Announce Type: replace-cross
Abstract: We consider the problem of community detection in overlapping weighted networks, where nodes can belong to multiple communities and edge weights can be finite real numbers. To model such complex networks, we propose a general framework - the mixed membership distribution-free (MMDF) model. MMDF has no distribution constraints of edge weights and can be viewed as generalizations of some previous models, including the well-known mixed membership stochastic blockmodels. Especially, overlapping signed networks with latent community …

abstract arxiv communities community constraints cs.lg cs.si detection distribution edge framework free general mixed multiple networks nodes numbers physics.soc-ph stat.ml type