IEEE transactions on signal processing, 2014-02, Vol.62 (3), p.703-717
2014
Details
- Autor(en) / Beteiligte
- Titel
- Locality Statistics for Anomaly Detection in Time Series of Graphs
- Ist Teil von
- IEEE transactions on signal processing, 2014-02, Vol.62 (3), p.703-717
- Ort / Verlag
- New York, NY: IEEE
- Erscheinungsjahr
- 2014
- Link zum Volltext
- Quelle
- IEEE/IET Electronic Library
- Beschreibungen/Notizen
- 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 properties and power characteristics of the competing scan statistics. Performance is compared theoretically, on synthetic data, and empirically, on the Enron email corpus.
- Sprache
- Englisch
- Identifikatoren
- ISSN: 1053-587XeISSN: 1941-0476DOI: 10.1109/TSP.2013.2294594Titel-ID: cdi_pascalfrancis_primary_28203549
- Format
- –
- Schlagworte
- Anomaly detection, Applied sciences, Communities, Constraining, Derivation, Detection, estimation, filtering, equalization, prediction, Electronic mail, Exact sciences and technology, Graphs, Information, signal and communications theory, Limiting, scan statistics, Signal and communications theory, Signal processing, Signal, noise, Statistics, Stochastic processes, Tasks, Telecommunications and information theory, Testing, Time series, Time series analysis, time series of graphs, Transaction processing