Rare Association Rule Mining for Network Intrusion Detection

Hyeok Kong, Cholyong Jong, Unhyok Ryang

In this paper, we propose a new practical association rule mining algorithm for anomaly detection in Intrusion Detection System (IDS). First, with a view of anomaly cases being relatively rarely occurred in network packet database, we define a rare association rule among infrequent itemsets rather than the traditional association rule mining method. And then, we discuss an interest measure to catch differences between interesting relations and uninteresting ones, and what interest there is, and develop a hash based rare association rule mining algorithm for finding rare, but useful anomaly patterns to user. Finally, we define a quantitative association rule in relational database, propose a practical algorithm to mine rare association rules from network packet database, and show advantages of it giving a concrete example. Our algorithm can be applied to fields need to mine hidden patterns which are rare, but valuable, like IDS, and it is based on hashing method among infrequent itemsets, so that it has obvious advantages of speed and memory space limitation problems over the traditional association rule mining algorithms. Keywords: rare association mining algorithm, infrequent itemsets, quantitative association rule, network intrusion detection system, anomaly detection

Knowledge Graph

arrow_drop_up

Comments

Sign up or login to leave a comment