Statistical inference on attributed random graphs: Fusion of graph features and content

Abstract: Fusion of information from graph features and content can provide superior inference for an anomaly detection task, compared to the corresponding content-only or graph feature-only statistics. In this paper, we design and execute an experiment on a time series of attributed graphs extracted from the Enron email corpus which demonstrates the benefit of fusion. The experiment is based on injecting a controlled anomaly into the real data and measuring its detectability. 1 

• Type: Journal
• Year: 2010
• Where: CSDA
• Authors: John Grothendieck, Carey E. Priebe, Allen L. Gorin
• Keywords: Anomaly Detection, CSDA 2010, Enron Email Corpus, Graph Feature-only Statistics