Training And Testing Anomaly-Based Neural Network Intrusion Detection Systems

Home Page
About
Submit A Journal
Submit A Conference
Submit Paper/Book
- Submit a Preprint
- Submit a Book
Contact

International Journal of Information Security Science
Volume:2 Issue:2
Training And Testing Anomaly-Based Neural Network Intrusion Detection Systems

Training And Testing Anomaly-Based Neural Network Intrusion Detection Systems

Authors : Loye RAY

Pages : 57-63

View : 80 | Download : 7

Publication Date : 2013-06-28

Article Type : Research Paper

Abstract :Networks are up against detecting dynamic and unknown threats. Anomaly-based neural network insert ignore into journalissuearticles values(NN); intrusion detection systems insert ignore into journalissuearticles values(IDSs); can manage this if trained and tested accordingly. This requires the IDS to be evaluated on how well it can detect these intrusions. Evaluating NN IDSs can be a complex and difficult task. One needs to be able to measure the convergence rate and performance insert ignore into journalissuearticles values(detection and failure); rate of the IDS. This paper explores the different methods used by researchers to train and test their IDS models. It also found that the data used can effect the results of training and testing the NN IDS models.
Keywords : intrusion detection, neural network, KDD 99, convergence rate, performance rate

ORIGINAL ARTICLE URL

* There may have been changes in the journal, article,conference, book, preprint etc. informations. Therefore, it would be appropriate to follow the information on the official page of the source. The information here is shared for informational purposes. IAD is not responsible for incorrect or missing information.