Nonparametric Anomaly Detection on Time Series of Graphs
| Autor: | Ivor Cribben, Dorcas Ofori-Boateng, Yulia R. Gel |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
Statistics and Probability
Dynamic network analysis Theoretical computer science Series (mathematics) Computer science 05 social sciences Univariate Nonparametric statistics Asymptotic distribution 01 natural sciences 03 medical and health sciences 010104 statistics & probability 0302 clinical medicine 0502 economics and business Test statistic Change points Discrete Mathematics and Combinatorics Anomaly detection 0101 mathematics Statistics Probability and Uncertainty Algorithm 030217 neurology & neurosurgery Change detection 050205 econometrics |
| DOI: | 10.1101/2019.12.15.876730 |
| Popis: | Identifying change points and/or anomalies in dynamic network structures has become increasingly popular across various domains, from neuroscience to telecommunication to finance. One of the particular objectives of the anomaly detection task from the neuroscience perspective is the reconstruction of the dynamic manner of brain region interactions. However, most statistical methods for detecting anomalies have the following unrealistic limitation for brain studies and beyond: that is, network snapshots at different time points are assumed to be independent. To circumvent this limitation, we propose a distribution-free framework for anomaly detection in dynamic networks. First, we present each network snapshot of the data as a linear object and find its respective univariate characterization via local and global network topological summaries. Second, we adopt a change point detection method for (weakly) dependent time series based on efficient scores, and enhance the finite sample properties of change point method by approximating the asymptotic distribution of the test statistic using the sieve bootstrap. We apply our method to simulated and to real data, particularly, two functional magnetic resonance imaging (fMRI) data sets and the Enron communication graph. We find that our new method delivers impressively accurate and realistic results in terms of identifying locations of true change points compared to the results reported by competing approaches. The new method promises to offer a deeper insight into the large-scale characterizations and functional dynamics of the brain and, more generally, into intrinsic structure of complex dynamic networks. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|