Research On Intrusion Protective Mechanism Of Database Based On User Behavior Mining

More and more important data stores in the Database Systems,the data is very attractive to some intruder because of its high importance and value,so the Database Systems has high possibility to be the intrusion target.And the weakness of the Database Systems is also the cause of the intrusion.Some traditional Database security mechanisms are prevention-oriented, so the Database Security Prevention is more critical.First,the article studies the Database Security and Database Security mechanisms,and it analyzes common means of attack on Database and the technology of traditional Database Security Prevention.Then it studies Data Mining technology,Intrusion Detection technology,Intrusion Tolerance technology and issues related technologies and problems of the Database Intrusion Detection.Finally,it studies user behavior of Database,and it has designed and realized a model of the Intrusion Prevention of Database based on User Behavior Mining combing by the related technologies.Next, the article brings up an advanced algorithm of User Behavior Analysis based on the FP-Growth algorithm to improve the mining efficiency of the Association Rule,the advanced algorithm is named BFP-Growth.The user's normal behavior rules which are mininged from this algorithm which is the foundation of Anomaly Detection,and it focuses on the distinction between the properties of the User Behavior.So this algorithm has improved the effeiciency of the Association Rule Mining.Finally,the prototype system designed by the article can be divided into the five models of System Management and Maintenance,Rules Mining and Management, Data Acquisition and Preprocessing,Behavior Anomaly Detection,Resposing and Recovery.The prototype system realizes that the rules can self-renewal and improvement,it impoves the algorithm of the Anomaly Detection and the Misuse Detection to raise the Detection Accuracy.It joines the Intrusion Tolerance technology to the model of Resposing and Recovery,it can set the resposing strategy by itself,so it improves the ability of survival in Database System.