查看文章

Intrusion Detection Systems (IDS) play a significant role in ensuring the security of communication networks. IDSs can easily generate hundreds of alerts per day, many of which are false alarms and it is extremely difficult for administrators to analyse and react to alarms. Reducing false alarm rate has become a great challenge for researchers. Feature selection has a significant influence on the performance of the IDS. A new hybrid feature selection method called Simple Hybrid Feature Selection (SHFS) is proposed which exhibits much reduced false alarm rate, improved precision and maintains good detection rate. Experiments on NSL-KDD dataset show that these features are the best for intrusion detection problem by an effective comparison with existing methods using 10 different classifiers.