文章基本信息

标题：Comparative Analysis Algorithm SVM, Naïve Bayes, And Decision Tree In Classifying Attack On Intrusion Detection System
本地全文：下载
作者：Dwi Widiastuti ; Prihandoko Prihandoko
期刊名称：Faculty of Computer Science and Information Technology
出版年度：2008
卷号：0
期号：0
语种：English
出版社：Faculty of Computer Science and Information Technology
摘要：Attack prediction is an action required by anintrusion detection system as a first step oranticipate in the event of an attack. Manymethods can be done to predict types ofattacks. One method used is the technique ofdata mining. But not all data mining algorithmshave good performance in classifying the typeof attack. Therefore, this study will try tocompare several algorithms.There are 41 attributes / variables used toclassify types of attack. Of the many types ofattacks that happened, then grouped into fourclasses, which are categorized based on thefinal destination reached by an attack. TheCategory are Probe, DoS, U2R, and R2L. Thedata set used are the data set from KDD Cup1999, where this set of data is referenced datafor IDS case study.The comparison algorithm can be seen basedon the value of instances classified correctly,incorrectly classified, kappa statistic, truepositive, false positive, and the confusionmatrix. By using the tools of WaikatoEnvironment for Knowledge Analysis (Weka)version 3.4.13, it can be concluded that thealgorithm has a superior performance is adecision tree.