Graph Mining

Anomaly Detection in Graphs

Contact Student: Vik Jakkula,


Anomalies are those rare abnormal occurances in the data. We are investing approaches to identify anomalies in graph based  datasets.
Anomaly detection can be used in vast fields including but not limited to network intrusion, fraud detection, health monitoring, sensor networks and more.

Relevant Publications:

- William Eberle and Lawrence Holder. "Anomaly Detection in Data Represented as Graphs." Intelligent Data Analysis: An International Journal. Volume 11, Number 6, pp. 663-689. 2007.

- William Eberle and Lawrence Holder. "Discovering Structural Anomalies in Graph-Based Data" Mining Graphs and Complex Structures Workshop, IEEE International Conference on Data Mining (ICDM), October 2007.

- William Eberle, Lawrence Holder and Jeffrey Graves. “Insider Threat Detection Using a Graph-based Approach,” Journal of Applied Security Research, Volume 6, Issue 1, pp. 32-81, January 2011.