Authors:
B. V. Patel
1
and
B. Meshram
2
Affiliations:
1
Shah & Anchor Kutchhi Polytechnic, India
;
2
Veermata Jijabai Technological Institute, India
Keyword(s):
Normal Network Traffic, IP, TCP, UDP Packet Format, Fuzzy, Association, Classification, Certainty.
Related
Ontology
Subjects/Areas/Topics:
Internet Technology
;
Intrusion Detection and Response
;
Web Information Systems and Technologies
Abstract:
To achieve the implementation of intrusion detection system (IDS), we have integrated the Fuzzy Logic with extended Apriori Association Data Mining to extract more abstract patterns at a higher level which look for deviations from stored patterns of normal behaviour of the computer network. Here the various packet formats of TCP, UDP, IP etc are used to study the normal behaviour of the network. Genetic algorithms are used to tune the fuzzy membership functions. The tuned data by genetic algorithms is processed by the modified Apriori algorithm. The association pattern is populated by genetic algorithm for the selection of best population of the network traffic. This best populated data is classified by the C4.5 algorithms to find intrusions. The deployment of IDS is done under the control of secure linux environment and the system is tested in the distributed environment.