I mplementation ofof FF uzzy CC- M eans eans eans and and DD empster-S hafer TTT Theory heory for AAA Anomaly nomaly II ntrusion DD etection
Tummala Pradeep · 2011
Summary: FuzzyclusteringtechniqueandDempster6Shafertheo ry� bothhavemeritofresolvingtheuncertaintyproble ms� raisedbylimitedandambiguousinformationordata � duringadecisionprocess.�Also,�the� k6NNtechniqueis� appliedtospeedupthedetectionprocess. Intrusion� detectioninfactisaclassificationtaskthatcla ssifies� networktrafficsintonormalusagecategoryoratta ck� category.�Inourwork,�themaingoalistoidentify �U2R � andR2L attacksfromtheKDD99 intrusiondetection� benchmarkdataset.�Forsuccessfullyachievingthe� goal,� wedividethedevelopmentofanintrusiondetection systemintotwophases:�trainingphaseandclassifi cation� phase.�Inthetrainingphase,�decisionrulesarege nerated� inaccordancewiththeclusteringresultofprovide d� trainingdata.�Therulesareusedforclassifyingf uture� networktrafficwhetherisanormalactivityoran� attack�