COMPUTER INTRUSION DETECTION WITH CLASSIFICATION AND ANOMALY DETECTION, USING SVMs

摘要: This paper describes experiences and results applying Support Vector Machine (SVM) to a Computer Intrusion Detection (CID) dataset. First, issues in supervised classification are discussed, then the incorporation of anomaly detection enhancing modeling prediction cyber-attacks. SVM methods seen as competitive with benchmark other studies, used standard for investigation. The approaches compare one class SVMs thresholded Mahalanobis distance define support regions. Results performance investigate joint detection. dataset is DARPA/KDD-99 publicly available features from network packets, classified into nonattack four-attack categories.