ensemble learning for intrusion detection in computer networks

ensemble learning for intrusion detection in computer networks

ID:33508352

大小:45.06 KB

页数:10页

时间:2019-02-26

ensemble learning for intrusion detection in computer networks_第1页
ensemble learning for intrusion detection in computer networks_第2页
ensemble learning for intrusion detection in computer networks_第3页
ensemble learning for intrusion detection in computer networks_第4页
ensemble learning for intrusion detection in computer networks_第5页
资源描述:

《ensemble learning for intrusion detection in computer networks》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、EnsembleLearningforIntrusionDetectioninComputerNetworksLucaDidaci,GiorgioGiacintoandFabioRoliDepartmentofElectricalandElectronicEngineering-UniversityofCagliari,ItalyPiazzaD'Armi-09123Cagliari,Italy{luca.didaci,giacinto,roli}@diee.unica.itAbstract.Thesecuri

2、tyofcomputernetworksplaysastrategicroleinmoderncomputersystems.Inordertoenforcehighprotectionlevelsagainstthreats,anumberofsoftwaretoolsarecurrentlydeveloped.IntrusionDetectionSystemsaimatdetectingintruderwhoeludedthe"firstline"protection.Inthispaper,apatte

3、rnrecognitionapproachtonetworkintrusiondetectionbasedonensemblelearningparadigmsisproposed.Thepotentialitiesofsuchanapproachfordatafusionandsomeopenissuesareoutlined.Keywords:IntrusionDetection,patternclassification,ensemblelearning,datafusion1IntroductionN

4、owadaysanincreasingnumberofcommercialandpublicservicesareofferedtroughInternet,sothatsecurityisbecomingoneofthekeyissues.Theso-called"attacks"tointernetserviceprovidersarecarriedoutbyexploitingunknownweaknessesorbugsalwayscontainedinsystemandapplicationsoft

5、ware[1-2].Computernetworksareusuallyprotectedagainstattacksbyanumberofaccessrestrictionpoliciesthatactasacoarsegrainfilter.Intrusiondetectionsystems(IDS)arethefinegrainfilterplacedinsidetheprotectednetwork,lookingforknownorpotentialthreatsinnetworktraffican

6、d/orauditdatarecordedbyhosts.Twoapproachestointrusiondetectionarecurrentlyused.Thefirstone,calledmisusedetection,isbasedonattacksignatures,i.e.,onadetaileddescriptionofthesequenceofactionsperformedbytheattacker.Thisapproachallowsthedetectionofintrusionsmatc

7、hingperfectlythesignatures.TheeffectivenessisstrictlyrelatedtotheextenttowhichIDSsareupdatedwiththesignaturesofthelatestattacksdeveloped.Thisiscurrentlyachallengesincenewattacksandnewattackvariantsareconstantlybeingdeveloped.Inparticular,atthetimeanattacksi

8、gnatureismadepubliclyavailable,anumberofattackvariantsaredesignedtoproducethesameeffectastheoriginalattack,butwithaslightlydifferentsignaturethatisnotdetectedbysignature-basedIDSs.Thedevelopmentofsigna

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。