Intrusion Detection System using Deep Neural Networks and Principal Component Analysis

The amount of data on the internet is growing daily, because of using the smart phones, social network, and IoT. This growing had impact on data security. Therefore security become the biggest challenge for researchers and developers, moreover, most of security tools (firewall, IDS, IPS, etc.) have limitations to detect all threats. Deep Learning is one of Machine Learning approach, it is an efficient artifice that can be applied to intrusion detection, to ascertain a new outline from the massive network data, as well as it used to reduce the strain of the manual compilations of the normal and abnormal behaviour patterns. In this paper we build a model for detect threats based on Principal Component Analysis (PCA) for reduction dimensions of dataset, and deep neural network for the classification. We used KDD99 dataset and 10_ percentage of KDD99 dataset to train and test the model in Spark environment. In experiment DNN of layers ranging from 1 to 3 with 300 number of epotch. The results were compared and concluded that a DNN of 1 layer has superior performance with 0.929 as accuracy.