Variational EM algorithm for the Stochastic Block Model

An implementation of an EM algorithm for the Stochastic Block Model. Project done during my master’s degree.

This project explores graph clustering using the Stochastic Block Model (SBM) and the EM algorithm. The SBM is a statistical model that identifies clusters in a graph by modeling the probability of edges between nodes based on their cluster memberships. The EM algorithm is used to estimate the parameters of the SBM, resulting in a clustering of the nodes.

Jean-Jacques Daudin, Franck Picard, Stéphane Robin. A mixture model for random graphs. [Research Report] RR-5840, INRIA. 2006, pp.19. ffinria-00070186

• The code is available on github.
• The project report is also available on github.