Locality statistics for anomaly detection in time series of graphs

Abstract

The ability to detect change-points in a dynamic network or a time series of graphs is an increasingly important task in many applications of the emerging discipline of graph signal processing. This paper formulates change-point detection as a hypothesis testing problem in terms of a generative latent position model, focusing on the special case of the Stochastic Block Model time series. We analyze two classes of scan statistics, based on distinct underlying locality statistics presented in the literature. Our main contribution is the derivation of the limiting distributions and power characteristics of the competing scan statistics. Performance is compared theoretically, on synthetic data, and on the Enron email corpus. We demonstrate that both statistics are admissible in one simple setting, while one of the statistics is inadmissible a second setting.

15 pages, 6 figures

References in corpus (2)

• Bayesian anomaly detection methods for social networks
• Anomaly Detection in Time Series of Graphs using Fusion of Graph Invariants

Cited by in corpus (16)

• Optimal Change Point Detection and Localization in Sparse Dynamic Networks
• A nonparametric two-sample hypothesis testing problem for random dot product graphs
• The blessing of transitivity in sparse and stochastic networks
• Sequential change-point detection based on nearest neighbors
• Change point localization in dependent dynamic nonparametric random dot product graphs
• Online Change Point Detection for Weighted and Directed Random Dot Product Graphs
• Change-Point Detection in Dynamic Networks with Missing Links
• Modelling and prediction of financial trading networks: An application to the NYMEX natural gas futures market
• The Importance of Being Correlated: Implications of Dependence in Joint Spectral Inference across Multiple Networks
• Multiple Network Embedding for Anomaly Detection in Time Series of Graphs
• Active Community Detection in Massive Graphs
• Asymptotic Distribution-Free Change-Point Detection for Multivariate and non-Euclidean Data
• F-FADE: Frequency Factorization for Anomaly Detection in Edge Streams
• Network Analysis with the Enron Email Corpus
• FlashGraph: Processing Billion-Node Graphs on an Array of Commodity SSDs
• Latent Space Model for Higher-order Networks and Generalized Tensor Decomposition