Latent self-exciting point process model for spatial-temporal networks

DOI10.3934/DCDSB.2014.19.1335zbMATH Open1327.93266arXiv1302.2671OpenAlexW2063569667MaRDI QIDQ478267FDOQ478267

Authors: Yoon-Sik Cho, Aram Galstyan, Jeffrey Brantingham, George E. Tita 

Publication date: 3 December 2014

Published in: Discrete and Continuous Dynamical Systems. Series B (Search for Journal in Brave)

Abstract: We propose a latent self-exciting point process model that describes geographically distributed interactions between pairs of entities. In contrast to most existing approaches that assume fully observable interactions, here we consider a scenario where certain interaction events lack information about participants. Instead, this information needs to be inferred from the available observations. We develop an efficient approximate algorithm based on variational expectation-maximization to infer unknown participants in an event given the location and the time of the event. We validate the model on synthetic as well as real-world data, and obtain very promising results on the identity-inference task. We also use our model to predict the timing and participants of future events, and demonstrate that it compares favorably with baseline approaches.

Full work available at URL: https://arxiv.org/abs/1302.2671

Recommendations

• A review of self-exciting spatio-temporal point processes and their applications
• Markov-modulated Hawkes processes for modeling sporadic and bursty event occurrences in social interactions
• scientific article; zbMATH DE number 7055492
• Point-process models of social network interactions: Parameter estimation and missing data recovery
• Multivariate spatiotemporal Hawkes processes and network reconstruction

zbMATH Keywords

clusteringdynamic network data analysisexpectation-maximization (EM)-algorithmlearning and inferenceself-exciting modelspatial-temporal point process

Mathematics Subject Classification ID

Decentralized systems (93A14) Discrete event control/observation systems (93C65)

Cites Work

• Simulation of nonhomogeneous poisson processes by thinning
• Space-time point-process models for earthquake occurrences
• Self-exciting point process modeling of crime
• Statistics for spatio-temporal data
• Mixed membership stochastic blockmodels
• Empirical Analysis of an Evolving Social Network
• Discrete temporal models of social networks
• Self-exciting hurdle models for terrorist activity
• Affine point processes and portfolio credit risk
• Point processes and queues. Martingale dynamics
• Point Process Modelling for Directed Interaction Networks

Cited In (15)

• Fast estimation of multivariate spatiotemporal Hawkes processes and network reconstruction
• Inferring social structure from continuous-time interaction data
• A modeling socialization point process sequence prediction algorithm
• A review of self-exciting spatio-temporal point processes and their applications
• Comment on ``A review of self-exciting spatio-temporal point processes and their applications by Alex Reinhart
• Rejoinder on: ``A review of self-exciting spatio-temporal point processes and their applications
• Tracking Dynamic Point Processes on Networks
• Preface to special issue on ``Mathematics of Social Systems
• Multivariate spatiotemporal Hawkes processes and network reconstruction
• Online estimation and community detection of network point processes for event streams
• Hidden Markov models for the activity profile of terrorist groups
• First- and Second-Order Characteristics of Spatio-Temporal Point Processes on Linear Networks
• Point-process models of social network interactions: Parameter estimation and missing data recovery
• A map of approaches to temporal networks
• A Metric Space for Point Process Excitations

This page was built for publication: Latent self-exciting point process model for spatial-temporal networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q478267)