Analysis of Fitness Function in Designing Genetic Algorithm Based Intrusion Detection System |
Author(s): |
Prof. Vithalpura Jahnavi S , L.D.COLLEGE OF ENGINEERING; Prof. H.M.Diwanji, L.D.COLLEGE OF ENGINEERING |
Keywords: |
genetic algorithm, intrusion, network intrusion detection system, fitness function |
Abstract |
Network Intrusion detection system is tool to monitor & identify intrusion in computers networks. The genetic algorithm is employed to derive a set of classification rules from network audit data. Different data sets are used as an audit data .From these data sets only specific features are selected and represented as chromosomes, which represent rules. The weighted sum model, support-confidence framework or reward penalty framework is utilized as fitness function to judge the quality of each rule. Best rule collection or knowledge base improves IDS performance by improving detection rate and reducing false alarm rate. The weighted sum model is generally more helpful for identification of network anomalous behaviors. The support –confidence framework is simply identifying network intrusions or precisely classifying the types of intrusions. Reward penalty technique used to give reward to the good chromosome and to apply penalty on the bad chromosome. This paper gives detail study about research carried out in fitness function of genetic based intrusion detection system. |
Other Details |
Paper ID: IJSRDV3I1150 Published in: Volume : 3, Issue : 1 Publication Date: 01/04/2015 Page(s): 86-92 |
Article Preview |
|
|