Contextual Stochastic Block Model

The contextual stochastic block model (cSBM) is a statistical model for analyzing networks with node attributes, aiming to improve community detection by incorporating both network structure and node features. Current research focuses on developing optimal inference algorithms, often based on belief propagation or iterative clustering, and evaluating their performance against graph neural networks (GNNs) on synthetic cSBM datasets. This work is significant for advancing our understanding of how to effectively leverage node attributes in network analysis, leading to improved algorithms for tasks like community detection and node classification in various applications.