A Policy Rule Dissimilarity Evaluation Approach Based on Fuzzy Theory

Policy conflict detection is an important and difficult technique in policy research. A key step in policy conflict detection is to compare the relationship between policies. Existing approaches to address the problem of policy comparison are mainly based on logical reasoning and Boolean function comparison. Such approaches are computationally expensive and don't scale well for large heterogeneous distributed environments. Consequently a lightweight and effective approach is needed to improve the efficiency of policy relationship evaluation. Considering that policy rule comparison is the basis of policy comparison, we introduce the concept of rule dissimilarity in this paper and apply fuzzy theory to analyzing and computing rule dissimilarity to address the rule relationship comparison firstly. Rule dissimilarity measure provides a lightweight approach to pre-compile a large amount of rules and only return the most similar rules for further policy relationship evaluation and policy conflict detection. Detailed algorithms are presented for the rule dissimilarly computation. Results of our case study demonstrate the efficiency and practical value of our approach.