An Adaptive Intrusion Detection System using a Data Mining Approach

Weak data dependencies in large databases coupled with poorly written web based applications are a major cause for malicious transactions. The problem of security becomes especially acute when access roles are changed among users. Also the poorly maintained data base caches are a cause for added security leaks. We propose an adaptive Intrusion detection system to keep track of the varying data dependencies as and when the definitions for various access roles are changed. We use an association rule based approach to track all relevant data dependency rule sets for different access roles using a hierarchical structure. We then identify malicious transactions from the transaction logs in the database using the data dependency rule sets. These rule sets are continuously updated and stored in a repository. Our approach is shown to reduce data access bottlenecks, and ensures minimal manual intervention for maintaining a secure database.