LEARNING PROBABILISTIC USER BEHAVIOR MODELS FROM DATABASE ACCESS LOG

Abstract

Client behavior modeling is a standout amongst the most imperative and fascinating issues should have been understood when creating and abusing present day programming frameworks. By client behavior modeling we mean finding examples of client action and developing prescient models dependent on point of reference behavior data. These models allow gauging next client activity based on the present movement. Essentially such technique was arranged to the commercial applications in proposal frameworks. At present time the region of its application is altogether more extensive. These methods assume an incredible job in PC security frameworks, where they are utilized for detecting malicious or unqualified client activities. In this paper proposed to learning probabilistic user behavior model approach used into database log process and their experimental results are checked into better performance of existing methods.