Statistical and Fuzzy Approach for Database Security

A new type of database anomaly is described by addressing the concept of cumulated anomaly in this paper. Dubiety-determining model (DDM), which is a detection model basing on statistical and fuzzy set theories for cumulated anomaly, is proposed. DDM can measure the dubiety degree of each database transaction quantitatively. Software system architecture to support the DDM for monitoring database transactions is designed. We also implemented the system and tested it. Our experimental results show that the DDM method is feasible and effective.