Extended stochastic block models with application to criminal networks

DOI10.1214/21-AOAS1595zbMATH Open1498.62323arXiv2007.08569OpenAlexW3124950460MaRDI QIDQ2080752FDOQ2080752

David Dunson, Daniele Durante, Tommaso Rigon, Sirio Legramanti

Publication date: 10 October 2022

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Reliably learning group structures among nodes in network data is challenging in several applications. We are particularly motivated by studying covert networks that encode relationships among criminals. These data are subject to measurement errors, and exhibit a complex combination of an unknown number of core-periphery, assortative and disassortative structures that may unveil key architectures of the criminal organization. The coexistence of these noisy block patterns limits the reliability of routinely-used community detection algorithms, and requires extensions of model-based solutions to realistically characterize the node partition process, incorporate information from node attributes, and provide improved strategies for estimation and uncertainty quantification. To cover these gaps, we develop a new class of extended stochastic block models (ESBM) that infer groups of nodes having common connectivity patterns via Gibbs-type priors on the partition process. This choice encompasses many realistic priors for criminal networks, covering solutions with fixed, random and infinite number of possible groups, and facilitates the inclusion of node attributes in a principled manner. Among the new alternatives in our class, we focus on the Gnedin process as a realistic prior that allows the number of groups to be finite, random and subject to a reinforcement process coherent with criminal networks. A collapsed Gibbs sampler is proposed for the whole ESBM class, and refined strategies for estimation, prediction, uncertainty quantification and model selection are outlined. The ESBM performance is illustrated in realistic simulations and in an application to an Italian mafia network, where we unveil key complex block structures, mostly hidden from state-of-the-art alternatives.

Full work available at URL: https://arxiv.org/abs/2007.08569

zbMATH Keywords

networkBayesian nonparametricsproduct partition modelGibbs-type prior

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics to social sciences (62P25) Social networks; opinion dynamics (91D30)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Understanding predictive information criteria for Bayesian models
• Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
• Estimation and Prediction for Stochastic Blockstructures
• Bayesian Measures of Model Complexity and Fit
• Model selection and clustering in stochastic block models based on the exact integrated complete data likelihood
• Spectral clustering and the high-dimensional stochastic blockmodel
• Pseudo-likelihood methods for community detection in large sparse networks
• Hierarchical clustering with discrete latent variable models and the integrated classification likelihood
• Bayesian Clustering and Product Partition Models
• Likelihood-based model selection for stochastic block models
• Network cross-validation by edge sampling
• Network Cross-Validation for Determining the Number of Communities in Network Data
• Bayesian cluster analysis: point estimation and credible balls (with discussion)
• Estimating the marginal likelihood using the arithmetic mean identity
• Consistency of community detection in networks under degree-corrected stochastic block models
• Consistency of spectral clustering in stochastic block models
• Mixed membership stochastic blockmodels
• Bayesian Nonparametric Estimation of the Probability of Discovering New Species
• Asymptotic normality of maximum likelihood and its variational approximation for stochastic blockmodels
• Community structure in social and biological networks
• Community Detection and Stochastic Block Models
• A BAYESIAN APPROACH TO MODELING STOCHASTIC BLOCKSTRUCTURES WITH COVARIATES
• Bayes Factors
• Global spectral clustering in dynamic networks
• Probabilistic Community Detection With Unknown Number of Communities
• Estimating the number of communities by spectral methods
• A mixture of experts latent position cluster model for social network data
• Comparing clusterings -- an information based distance
• Fast unfolding of communities in large networks
• An asymptotic analysis of a class of discrete nonparametric priors
• Inconsistency of Pitman-Yor process mixtures for the number of components
• Mixture Models With a Prior on the Number of Components
• Community detection in networks with node features
• A species sampling model with finitely many types
• A Consistent Adjacency Spectral Embedding for Stochastic Blockmodel Graphs
• Bayesian nonparametric estimators derived from conditional Gibbs structures
• Role of normalization in spectral clustering for stochastic blockmodels
• Controlling the Reinforcement in Bayesian Non-Parametric Mixture Models
• Statistical inference on random dot product graphs: a survey
• Covariate-assisted spectral clustering
• A new Monte Carlo method for estimating marginal likelihoods
• Analysis of spectral clustering algorithms for community detection: the general bipartite setting
• Identifying overlapping terrorist cells from the Noordin Top actor-event network
• The hierarchy of block models
• Bayesian testing for exogenous partition structures in stochastic block models
• Multiresolution Network Models
• A Block Model for Node Popularity in Networks with Community Structure

Cited In (6)

• Finite-dimensional Discrete Random Structures and Bayesian Clustering
• Dynamic Stochastic Blockmodel Regression for Network Data: Application to International Militarized Conflicts
• Extended stochastic block model with spatial covariates for weighted brain networks
• Bayesian mixture models (in)consistency for the number of clusters
• Horizon-\(K\) farsightedness in criminal networks
• Bayesian learning of graph substructures

This page was built for publication: Extended stochastic block models with application to criminal networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2080752)