Attribute-Based Access Control Policy Generation Approach from Access Logs Based on the CatBoost

Attribute-based access control (ABAC) has higher flexibility and better scalability than traditional can be used for fine-grained of large-scale information systems. Although ABAC depict a dynamic, complex policy, it is costly, tedious, error-prone to manually define. Therefore, worth studying how construct an policy efficiently accurately. This paper proposes generation approach based on the CatBoost algorithm automatically learn policies from historical logs. First, we perform weighted reconstruction attributes mined. Second, provide rule extraction algorithm, pruning optimization among which algorithms are improve accuracy generated policies. In addition, present new quality indicator measure simplicity Finally, results experiment conducted validate verify its feasibility effectiveness.