Supervised Machine Learning,, Unsupervised Machine Learning,, Network Intrusion Detection,, Network Security

April 2017

As network attacks have increased in number and severity over the past few years, intrusion detection system (IDS) is increasingly becoming a critical component to secure the network. Due to large volumes of security audit data as well as complex and dynamic properties of intrusion behaviours, optimizing performance of IDS becomes an important open problem that is receiving more and more attention from the research community. This paper compares the performance of Intrusion Detection System (IDS) Classifiers using various feature reduction techniques. To enhance the learning capabilities and reduce the computational intensity of competitive learning comparing the performance of the algorithms is performed respectively, different dimension reduction techniques have been proposed. These include: classifying and clustering Algorithms Naïve Bayes, Simple k mean, Decision tree and J48, Linear Discriminate Analysis, and Independent Component Analysis. Many Intrusion Detection Systems are based on neural networks. However, they are computationally very demanding. This paper provides a review on current trends in intrusion detection together with a study on technologies implemented by some researchers in this research area. We try to build as system which create clusters from its input data by labelling clusters as normal or anomalous data instances and finally used these cluster to classify unseen network data instances as either normal or anomalous1. Both training and testing was done using different subset of KDD Cup 99 2 datawhich is very popular and widely used intrusion attack dataset.

16-19

2455-9997 (Online) - 2229-3531 (Print)

Vol.8 - No.2

Click here to download full paper

http://dx.doi.org/10.22147/jucit/080201