ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Feature Selection for Security evaluation against Attack in Adversarial Environment

Journal: International Journal of Engineering Research (IJER) (Vol.5, No. 5)

Publication Date:

Authors : ; ;

Page : 425-429

Keywords : Adversariallearning; classifiersecurity; evasionatta cks; featureselection; securityevaluation; robustness evaluation; spamfiltering.;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

:NotonlyPatternrecognitionbutalsomachinelearningtechniqu eshavebeenincreasedinsecurityrelatedapplicationslikeintrusi on,spam,andmalwaredetection,althoughitssecurityagainstwe llcraftedattacksthataimstoevadedetection.Spamfilteringisoneo fthewellknownapplicationexamplesconsideredinadversariale nvironment.Inthistask,thegoalisoftentodesignfeatureselectio nagainstattackswhichmayincurduringoperation.Hereinthisp aper,Iprovideamoredetailedinvestigationofthisaspect,byshed dingsome light onthe securitypropertiesoffeatureselectionagainstevasionattacks.A lsohere IuseRandomForestClassifiertofindevasionattacks.Theability ofrapidlyevolveofchangingandcomplexsituationshashelpeditt obecomeafundamentaltoolforcomputersecurity.Evasionattac ksmayassumesthattheattackercanarbitrarilychangetheevery feature,buttheyconstrainthedegreeofmanipulation,e.g.,limiti ngthenumberofmedications,orit’stotalcost.AdversarialFeatu reSelectionarchitecturemodelare givenin thispaper.

Last modified: 2016-08-08 18:23:47