Bayesian anomaly detection methods for social networks
Authors:
- Nicholas A. HeardNHNicholas A. Heard
- David J. WestonDWDavid J. Weston
- Kiriaki PlataniotiKPKiriaki Platanioti
- David J. HandDHDavid J. Hand
Preprint
Abstract
Learning the network structure of a large graph is computationally demanding, and dynamically monitoring the network over time for any changes in structure threatens to be more challenging still. This paper presents a two-stage method for anomaly detection in dynamic graphs: the first stage uses simple, conjugate Bayesian models for discrete time counting processes to track the pairwise links of all nodes in the graph to assess normality of behavior; the second stage applies standard network inference tools on a greatly reduced subset of potentially anomalous nodes. The utility of the method is demonstrated on simulated and real data sets.Keywords
All Related Versions
- Version 1, 2010-11-08, ArXiv
- Published version: The Annals of Applied Statistics, 4 (2), 645.
This publication has 4 references indexed in Scilit:
- Scan Statistics on Enron GraphsComputational and Mathematical Organization Theory, 2005
- Logit Models and Logistic Regressions for Social Networks: I. An Introduction to Markov Graphs and pPsychometrika, 1996
- Specification and testing of some modified count data modelsJournal of Econometrics, 1986
- A Bayesian Analysis of Some Nonparametric ProblemsThe Annals of Statistics, 1973