Detecting structural changes in longitudinal network data

DOI10.1214/19-BA1147zbMATH Open1436.62401arXiv1811.01384WikidataQ128386586 ScholiaQ128386586MaRDI QIDQ2297236FDOQ2297236

Authors: Jong Hee Park, Yunkyu Sohn 

Publication date: 18 February 2020

Published in: Bayesian Analysis (Search for Journal in Brave)

Abstract: Dynamic modeling of longitudinal networks has been an increasingly important topic in applied research. While longitudinal network data commonly exhibit dramatic changes in its structures, existing methods have largely focused on modeling smooth topological changes over time. In this paper, we develop a hidden Markov multilinear tensor model (HMTM) that combines the multilinear tensor regression model (Hoff 2011) with a hidden Markov model using Bayesian inference. We model changes in network structure as shifts in discrete states yielding particular sets of network generating parameters. Our simulation results demonstrate that the proposed method correctly detects the number, locations, and types of changes in latent node characteristics. We apply the proposed method to international military alliance networks to find structural changes in the coalition structure among nations.

Full work available at URL: https://arxiv.org/abs/1811.01384

Recommendations

• Multiple change points detection and clustering in dynamic networks
• Latent space models for dynamic networks
• A mixed effects model for longitudinal relational and network data, with applications to international trade and conflict
• Latent space approaches to community detection in dynamic networks
• Model-based clustering of time-evolving networks through temporal exponential-family random graph models

zbMATH Keywords

Bayesian inferencehidden Markov modelWAICmilitary alliancenetwork latent spacetopological changes over time

Mathematics Subject Classification ID

Bayesian inference (62F15) Directional data; spatial statistics (62H11) Markov processes: estimation; hidden Markov models (62M05) Applications of statistics (62P99) Functional data analysis (62R10) Topological data analysis (62R40)

Cites Work

• Understanding predictive information criteria for Bayesian models
• Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
• Spectral clustering and the high-dimensional stochastic blockmodel
• Covariance structure of the Gibbs sampler with applications to the comparisons of estimators and augmentation schemes
• A survey of statistical network models
• Bayesian compressed regression
• Pattern recognition and machine learning.
• Likelihood ratio tests and singularities
• Estimation and comparison of multiple change-point models
• Consistency of community detection in networks under degree-corrected stochastic block models
• Machine learning. A probabilistic perspective
• Marginal Likelihood from the Gibbs Output
• Title not available (Why is that?)
• Discrete temporal models of social networks
• Random graph models for temporal processes in social networks*
• A mixed effects model for longitudinal relational and network data, with applications to international trade and conflict
• Partially Collapsed Gibbs Samplers
• A Multilinear Singular Value Decomposition
• Hierarchical multilinear models for multiway data
• Model Averaging and Dimension Selection for the Singular Value Decomposition
• Networks. An introduction.
• Multilinear tensor regression for longitudinal relational data
• Title not available (Why is that?)
• Tensor decompositions and sparse log-linear models
• Title not available (Why is that?)
• Bayesian anomaly detection methods for social networks
• Inference for nonconjugate Bayesian Models using the Gibbs sampler
• Dealing with reciprocity in dynamic stochastic block models
• Reconstructing evolving signalling networks by hidden Markov nested effects models

Cited In (6)

• Multiple change points detection and clustering in dynamic networks
• Joint modeling of longitudinal relational data and exogenous variables
• A latent space model and Hotelling's T² control chart to monitor the networks of Covid-19 symptoms
• Network trees: a method for recursively partitioning covariance structures
• A mixed effects model for longitudinal relational and network data, with applications to international trade and conflict
• Sequential network change detection with its applications to ad impact relation analysis

Uses Software

• rstiefel
• PMTK
• PRMLT

This page was built for publication: Detecting structural changes in longitudinal network data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2297236)