Model selection in overlapping stochastic block models

DOI10.1214/14-EJS903zbMATH Open1349.62276arXiv1405.2722OpenAlexW2963775172MaRDI QIDQ2015159FDOQ2015159

Authors: Pierre Latouche, Etienne Birmelé, Christophe Ambroise 

Publication date: 23 June 2014

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: Networks are a commonly used mathematical model to describe the rich set of interactions between objects of interest. Many clustering methods have been developed in order to partition such structures, among which several rely on underlying probabilistic models, typically mixture models. The relevant hidden structure may however show overlapping groups in several applications. The Overlapping Stochastic Block Model (2011) has been developed to take this phenomenon into account. Nevertheless, the problem of the choice of the number of classes in the inference step is still open. To tackle this issue, we consider the proposed model in a Bayesian framework and develop a new criterion based on a non asymptotic approximation of the marginal log-likelihood. We describe how the criterion can be computed through a variational Bayes EM algorithm, and demonstrate its efficiency by running it on both simulated and real data.

Full work available at URL: https://arxiv.org/abs/1405.2722

Recommendations

• Overlapping stochastic block models with application to the French political blogosphere
• Variational Bayesian inference and complexity control for stochastic block models
• Likelihood-based model selection for stochastic block models
• Non-parametric overlapping community detection
• Stochastic blockmodels with a growing number of classes

zbMATH Keywords

graph clusteringmodel selectionrandom graph modelsglobal and local variational techniquesoverlapping stochastic block models

Mathematics Subject Classification ID

Bayesian inference (62F15) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Graph minors (05C83)

Cites Work

• A nonparametric view of network models and Newman–Girvan and other modularities
• Statistical mechanics of complex networks
• Title not available (Why is that?)
• Estimation and Prediction for Stochastic Blockstructures
• Pattern recognition and machine learning.
• Exact and Monte Carlo calculations of integrated likelihoods for the latent class model
• Estimation and prediction for stochastic blockmodels for graphs with latent block structure
• Mixed membership stochastic blockmodels
• Title not available (Why is that?)
• Community structure in social and biological networks
• Stochastic Blockmodels for Directed Graphs
• Cluster Inference by Using Transitivity Indices in Empirical Graphs
• Overlapping stochastic block models with application to the French political blogosphere
• Uncovering latent structure in valued graphs: a variational approach
• 10.1162/jmlr.2003.3.4-5.993
• Fast online graph clustering via Erdős-Rényi mixture
• Accuracy of variational estimates for random graph mixture models

Cited In (9)

• The dynamic random subgraph model for the clustering of evolving networks
• Hierarchical Network Models for Exchangeable Structured Interaction Processes
• Modeling the social media relationships of Irish politicians using a generalized latent space stochastic blockmodel
• Likelihood-based model selection for stochastic block models
• Weighted stochastic block model
• Goodness of Fit of Logistic Regression Models for Random Graphs
• Overlapping stochastic block models with application to the French political blogosphere
• Estimation and selection for the latent block model on categorical data
• Detecting Overlapping Communities in Networks Using Spectral Methods

Uses Software

• StOCNET
• latentnet
• PRMLT

This page was built for publication: Model selection in overlapping stochastic block models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2015159)