期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2008
卷号：8
期号：4
页码：160-169
出版社：International Journal of Computer Science and Network Security
摘要：Increasing the number of intrusions and attacks especially in large scale and distributed public data networks has initiated a new phenomenon in cyberspace’s security. We have presented a novel approach using cooperative co-evolutionary immune system for intrusion detection in data networks. This is a machine learning method based on genetic algorithm and co-evolutionary immune system where the detectors can discriminate the incidents and non-incidents in a distributed environment. We have prepared a prototype of CoCo-ISD in a Jini platform running grid computing in a distributed environment. The obtained results show that, CoCo-ISD can adaptively converge for the best fitness in the selected boundary. The system needs less number of rules with less complexity and high accuracy metric where the detectors (rules) have more flexibility and diversity compared to the rules in IS system. Moreover, the advantage of the proposed system is the learning capability that focuses on the events as the suspicious incidents with low fitness level and in variable threshold. We have confirmed the probability of detection (PD) and false error rate (FER) in KDD database with several well known methods for proof and validation of our results.
