Policy Modeling and Anomaly Detection in ABAC Policies.

Sensitive data is available online through web and distributed protocols that highlight the need for access control mechanisms. System designers write access control policies to represent conditions on accessing data. Access control policies can contain anomalies (redundancy, inconsistency, irrelevancy, and incompleteness) that can lead to security vulnerabilities. Detecting anomalies in large and complex policies is challenging due to the lack of effective analysis mechanisms and tools. In this paper, we introduce a formal tree-based policy modeling technique to represent, update, and analyze access control policies. Based on the proposed formal policy modeling, we propose an anomaly detection technique. Our approach focuses on Attribute Based Access Control (ABAC) policies as they are widely adopted. Also, they can provide high flexibility and enhance security and information sharing. The effectiveness of our policy modeling and anomaly detection technique has been demonstrated through experimental evaluation.