Bayesian nonparametrics for Sparse Dynamic Networks
Preprint
Abstract
We propose a Bayesian nonparametric prior for time-varying networks. To each node of the network is associated a positive parameter, modeling the sociability of that node. Sociabilities are assumed to evolve over time, and are modeled via a dynamic point process model. The model is able to (a) capture smooth evolution of the interaction between nodes, allowing edges to appear/disappear over time (b) capture long term evolution of the sociabilities of the nodes (c) and yield sparse graphs, where the number of edges grows subquadratically with the number of nodes. The evolution of the sociabilities is described by a tractable time-varying gamma process. We provide some theoretical insights into the model and apply it to three real world datasets.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Sparse Graphs Using Exchangeable Random MeasuresJournal of the Royal Statistical Society Series B: Statistical Methodology, 2017
- Bayesian Models of Graphs, Arrays and Other Exchangeable Random StructuresIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- Detecting communities and their evolutions in dynamic social networks—a Bayesian approachMachine Learning, 2010
- Posterior Analysis for Normalized Random Measures with Independent IncrementsScandinavian Journal of Statistics, 2009
- Bayesian Poisson process partition calculus with an application to Bayesian Lévy moving averagesThe Annals of Statistics, 2005
- Constructing Stationary Time Series Models Using Auxiliary Variables With ApplicationsJournal of the American Statistical Association, 2005
- Exchangeable random measures in the planeJournal of Theoretical Probability, 1990
- Stochastic evolution equations and related measure processesJournal of Multivariate Analysis, 1975
- A limit theorem of branching processes and continuous state branching processesKyoto Journal of Mathematics, 1968