Detecting Anomalies in Time-Varying Networks Using Tensor Decomposition

Anna Sapienza, Andre Panisson, Joseph Wu, Laetitia Gauvin, Ciro Cattuto

Risultato della ricerca: Capitolo in libro/report/atti di convegnoContributo a conferenzapeer review

Abstract

New data sources from sensor networks and Internet-of-Things applications promise a wealth of interaction data that can be naturally represented as time-varying networks. This brings forth new challenges for the identificationand removal of time-varying graph anomalies that entail complex correlations of topological features and temporal activity patterns. Here we present an anomaly detection approach for temporal graph data, based on an iterative tensor decomposition and masking procedure. We test this approach using high-resolution social network data from wearable proximity sensors. The dataset includes metadata that allow to independently build a ground truth, used to validate the anomaly detection method. Our approach achieves high accuracy in identifying meso-scale network anomalies due to sensor wearing protocol, proving the practical viability of the method for a real-world application.

Lingua originaleInglese
Titolo della pubblicazione ospiteProceedings - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015
EditorXindong Wu, Alexander Tuzhilin, Hui Xiong, Jennifer G. Dy, Charu Aggarwal, Zhi-Hua Zhou, Peng Cui
EditoreInstitute of Electrical and Electronics Engineers Inc.
Pagine516-523
Numero di pagine8
ISBN (elettronico)9781467384926
DOI
Stato di pubblicazionePubblicato - 29 gen 2016
Pubblicato esternamente
Evento15th IEEE International Conference on Data Mining Workshop, ICDMW 2015 - Atlantic City, United States
Durata: 14 nov 201517 nov 2015

Serie di pubblicazioni

NomeProceedings - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???15th IEEE International Conference on Data Mining Workshop, ICDMW 2015
Paese/TerritorioUnited States
CittàAtlantic City
Periodo14/11/1517/11/15

Fingerprint

Entra nei temi di ricerca di 'Detecting Anomalies in Time-Varying Networks Using Tensor Decomposition'. Insieme formano una fingerprint unica.

Cita questo