Immune Multi Agent System for Intrusion Prevention and Self healing System Implement a Non-Linear Classification

Muna, Elsadig and Azween, Abdullah and Samir, B. B. (2010) Immune Multi Agent System for Intrusion Prevention and Self healing System Implement a Non-Linear Classification. In: The 4th International Symposium on Information Technology (ITSim 2010), 15-17/06/2010, Kuala Lumpur.

[thumbnail of 05561486.pdf] Archive (ZIP)
05561486.pdf
Restricted to Registered users only

Download (485kB)

Abstract

Artificial immune systems have recently been implemented in the field of computer security system particularly in intrusion detection and prevention systems. In this paper researchers present an approach to an intrusion prevention system (IPS) which is inspired by the Danger model of immunology. This novel approach used a multi immune agent system that implements a non-linear classification method to identify the abnormality behavior of network system. The authors look into Dendritic Cell (DC) which is a cell in Innate Immune system (IIS) as a classifier cell. Our approach takes the advantages of multi agent system, Dendritic cell, Cluster-K-Nearest-Neighbor, K-mean and Gaussion mixture methods which are give an autonomous, highly accurate and fast classifier security system. This is based on intelligent agents that exploit known functional features of the immune system and the self-healing system to detect, prevent and heal harmful or dangerous events in network systems A combination of features between the IPS and self healing (SH) mechanism to ensure continuity of the networked systems have been established.

Item Type: Conference or Workshop Item (Paper)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments / MOR / COE: Research Institutes > Megacities
Depositing User: Assoc Prof Dr Azween Abdullah
Date Deposited: 15 Nov 2010 03:45
Last Modified: 31 Dec 2012 04:06
URI: http://scholars.utp.edu.my/id/eprint/2922

Actions (login required)

View Item
View Item