To search, Click below search items.


All Published Papers Search Service


Application of an AODE Based Classifier to Detect DOS Attacks


Levent Koc, Alan D. Carswell


Vol. 15  No. 2  pp. 24-28


Digital forensics often utilize network intrusion detection systems based on various data mining methods to detect and collect evidence on intrusion events such as Denial of Service (DOS) attacks. Findings of our experiments reveal that a classification model based on averaged one-dependence estimators (AODE) can be used for this purpose. AODE is an extension of Na?ve Bayes method which relies on conditional independence assumption. A multiclass classifier model based on AODE is proposed for accurate detection of DOS attacks. Results of the experiments using KDD’99 intrusion detection dataset indicate the proposed classifier based on AODE model performs better than the classifier model based on traditional Na?ve Bayes method in terms of accuracy to detect DOS attacks


Averaged one-dependence estimators (AODE), Na?ve Bayes classifier, Denial of service (DOS) attack, Intrusion detection system (IDS)