Cyber intrusion detection through association rule mining on multi-source logs
Ping Lou,Guantong Lu,Xuemei Jiang,Zheng Xiao,Jiwei Hu,Junwei Yan
DOI: https://doi.org/10.1007/s10489-020-02007-5
IF: 5.3
2020-01-01
Applied Intelligence
Abstract:Security logs in cloud environment like intrusion detection system (IDS) logs, firewall logs, and system logs provide historical information describing potential security risks. However, the use of logs for cyber intrusion detection relies heavily on expert knowledge. It is very difficult for the non-expert to identify these intrusion behaviors. This paper proposes a new method for mining association rules from multi-source logs to detect various intrusion behaviors in the cloud computing platform. In this method, a rule base is constructed to detect cyber intrusion. An adaptive approach is used to speed up the calculation of the association rule mining, in which the decision depends on the time complexity of the algorithm. Various cyber-attacks are simulated in the verification experiments which show the calculation speed of the proposed method is faster than other algorithms. Furthermore, compared with other methods, the performance of the proposed intrusion detection method is better than others in term of precision, recall, and f-measure.