Increasing the rate of intrusion detection based on a hybrid technique

Abstract

This paper presents techniques to increase intrusion detection rates. Theses techniques are based on specific features that are detected and it's shown that a small number of features (9) can yield improved detection rates compared to higher numbers. These techniques utilize soft computing techniques such a Backpropagation based artificial neural networks and fuzzy sets. These techniques achieve a significant improvement over the state of the art for standard DARPA benchmark data.