Detection, Defensive and Mitigation of DDoS Attacks through Machine learning Techniques: A Literature

2019 International journal of recent technology and engineering  
iNowaday ithe iworld iis icompletely idepend on ithe iinternet iand the day ito iday iactivities of ihuman ilife icompletely idepends ion ithe iinternet. The idependency ion ithe iinternet iallows ithe iattackers ido damage ior iharm ito ithe ilegitimate iuser's itransactions iand ievents iwhich iis icalled ias iSecurity iattack. Distributed iDenial iof iService iis ione itype iof ithe imost vulnerable iattacks iof itoday's icyber iworld. iIn ithis ipaper, iwe ipresent ia isurvey iof
more » ... d iDenial iof iService iattack, idetection, idefensive iand imitigation iof imachine ilearning iapproaches. This isurvey iarticle iprefer itwo ifamous isupervised imachine ilearning ialgorithms isnamely. (i) Decision itrees, (ii)isupport ivector imachine and ipresented ithe irecent iresearch iworks icarried iout. From ithis isurvey iit iis ilearnt ithat iconnecting supervised imachine ilearning ialgorithm iwith iboosting iprocess will iincrease iprediction iefficiency iand ithere iis ia iwide iscope iin ithis iresearch ielement. We provide a systematic analysis of these attacks including so many motivations and evolutions, different types of attacks analysis so far, detection techniques and mitigation techniques, possible constraints and challenges of existing approaches. iFinally isome iimportant iresearch ipoints are outlined ito iensure isuccessful idetection, idefensive iand mitigation iagainst iDistributed iDenial iof iService attacks
doi:10.35940/ijrte.d7335.118419 fatcat:y4tkl7pcqvcgln5qg5weshyud4