A Dynamic Approach to Detect Anomalous Queries on Relational Databases

摘要: Protecting sensitive datasets from insider and outsider attacks has been a major concern over the years. Relational Database Management System (RDBMS) de facto standard to store, retrieve manage large efficiently in last few However, as surprising it seems, not lot of works can be found literature which protect databases anomalous accesses. In this paper, we present novel Intrusion Detection (IDS) for relational databases. Our primary objective is both threats by detecting access patterns using Hidden Markov Model (HMM). While most previous notable area focus on query syntax detect access, our approach takes into account amount information result contains potential intrusion. Finally, empirical evaluation publicly available TPC-H dataset shows that IDS with high degree accuracy.