iizetqe Bu qalynada ag tabanli saldin tcspit sistemleri iqin yapay sinir aglani kullanan hir yap1 Bnermckteyiz. Sadece normal baglantilar ipin bir matematikscl model kwularak, gelen diger haglantilmn normal davrmq madelinden s a p d a n n a hahlmaktadir. Normal dawanip modellemesi ve olagandi$ilik tespiti iqin SOM (Self Organizing Map) yapisz kullanilmigtir. SOM yapis! topolojik haritalamayi koruma bzelliginden dolay: belli hir metrik goz oniine alinarak herhangi bir uzaydaki hirbirine yakm olarak ifade edilen vektbrleri kendi haritasi iizerine de yakm olarak yerle~tkir. Saldm tespit sistemi analizinde, SOM yapisi normal d a v m $ baglantilan ile eatilir ve normal haglanhlar SOM haritasi iizerinde helirli naktalarda topaklanirlar. Saldinlar ise hu topaklann di$indaki hslgelere du$erler ya da normal dawani$ holgelerine ybksek nicemleme h a m ile diigerler. Boylece SOM haritasi iizerinde gelen haglanunm dii$tiigii h6lge ve nicemleme hatasi hize baglantimn tipini helirler.
AbstractIn this paper, networkbased anomaly inmsion detection systems using mificial neural networks are investigated. Only knowing normal traffic data, a mathematical model describing normal traffic is constructed and test is conducted based on the deviations form the mathematical model. Self-Organizing Map (SOM) smctwe is used for constructing the mathematical model describing normal traffic and anomaly detection. SOM shucture preserver topological mappings hcween representations. A feature which is desired when classifying normal or intrusive behavior far network data. our hypothesis is that normal traffx representing normal behavior would he clustered around one or more cluster centers and any irregular traffic representing abnormal and possibly suspicious behavior would be clustered outside of the normal clustering or inside with high quantization error. SOM is trained with normal traffic data and by considering the best matching unit or clustering region and the quantization error, the type of the traffic is determined.
Giri$Saldin tespiti alanmda bu giinc dean yapilan yali$rnalar iki temel kategoride incelenebilir: olagandipilik tespiti ve kntiiye kullanirn tespiti. KbUiye kullanim tespitinde, bnceden bilinen politika ihlali olasiliklan (ya da yontcmleri) once senaryolara doniiptmliir. Senaryolar daha soma saldin imzalanna indirgenir. Saldm imalan. belirli hir saldmya iliqkin olasi tiim senaryolann ana eksenini olu$turan bir bzet olarak nitelendirilcbilirler. Kural tabanli uzman sistemler veya imza do@ulama sistemleri kullanilarak olupurulan saldm imzalannm. sistemin anlayacag! makine tarafmdan okunabilir bipime donii$tiiriiliir. Sistemin kaynak girdisi (ag mafigi, sistemlere ilipkin giinllik kayitlan vb.) iizerinde yapllan analizler ile bilinen zaafiara ybnelik saldmlarln ya da ihlallerin gerqeklepip gerqeklegrnedigi tespit edilmeye qah$ila; tespit edilmesi durumunda a l m Uretilir [3,5]. Olagandigilik tcspitinde ise, kaynaklann (kullanichx, programlar, sistemler vh.) normal durum davranigi istatistikscl. kural tahanli ve yapay sinir aglan...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.