A multi-agent intrusion detection model based on importance feature extraction
by Yu Yang; Ping He; Shengli Xing
International Journal of Computational Science and Engineering (IJCSE), Vol. 27, No. 4, 2024

Abstract: The swift evolution of the internet has delivered convenience to people while also introducing challenges concerning the security of information. Network intrusion detection, which recognises distinct attack behaviours in the network by gathering and analysing network data, is a key piece of technology for information security. However, traditional intrusion detection methods suffer from feature redundancy, ignore the temporal characteristics of data, and are difficult to effectively deal with multiple attack patterns. Therefore, to address the above problems, a multi-agent intrusion detection model based on important feature extraction is proposed in this paper, named MIDI. MIDI first calculates the importance indicator of features in network traffic by combining attention mechanisms to extract features, and then introduces the reinforcement learning method and the time series model to build a multi-agent model to learn various attack modes in the network. Ultimately, the experimental results demonstrate the efficiency and superiority of MIDI compared to traditional intrusion detection models, and the precision on the NSL-KDD and UNSW-NB15 datasets can reach 98.23% and 94.93%, respectively.

Online publication date: Fri, 05-Jul-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Science and Engineering (IJCSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com