A hybrid artificial immune system for IDS based on SVM and belief function

Increased connectivity and the use of the internet have exposed the subversion in front of the organizations, there for there is a need to use of intrusion detection system to protect information system and communication network from malicious attacks and unauthorized access. An intrusion detection system (IDS) is a security system that monitors computer systems and network traffic, analyze that traffic to identify possible security breaches and raise alerts. An IDS triggers thousands of alerts per day which is difficult for human users to analyze them and take appropriate actions. It is important to reduce the false alarm alerts, intelligently integrate and correlate them in order to present a high level view of the detected security issue to the administrator. In this paper an hybrid model has been proposed in which intrusion detection takes place with the help Dendritic Cell Algorithm and Dempester belief theory along with SVM classification algorithms. It made the Intrusion Detection System much more efficient and accurate as compared to the existing System. It also seems to be an improvement in precise value of proper alarm generation to enhance the performance of the system.