Affiliation:
1. Laboratory of Research in Artificial Intelligence, Alger, Algeria
Abstract
Due to a growing number of intrusion events, organizations are increasingly implementing various intrusion detection systems that classify network traffic data as normal or anomaly. In this paper, three intrusion detection systems based fuzzy meta-heuristics are proposed. The first one is a fuzzy stochastic local search (FSLS). The second one is a fuzzy tabu search (FTS) and the third one is a fuzzy deferential evolution (FDE). These classifiers are built on a knowledge base modelled as a fuzzy rule “if-then”. The main purpose of these methods is to get the highest quality solutions by optimizing the fuzzy rules generation. The proposed classifiers FSLS, FTS and FDE are tested on the benchmark KDD'99 intrusion dataset and compared with some well-known existing techniques for intrusion detection. The results show the efficiency of the proposed approaches in the intrusion detection field.
Subject
Decision Sciences (miscellaneous),Computational Mathematics,Computational Theory and Mathematics,Control and Optimization,Computer Science Applications,Modeling and Simulation,Statistics and Probability
Reference34 articles.
1. Intrusion detection using a fuzzy genetics-based learning algorithm
2. Instance-based learning algorithms
3. Anderson, J. P. (1980). Computer Security threat monitoring and surveillance. Retrieved from http://csrc.nist.gov/publications/history/ande80.pdf
4. Intrusion detection using fuzzy stochastic local search classifier.;B.Bahamida;MICAI,2012
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献