Intrusion Detection System plays an important role in protecting the security of the information and by identifying the various attacks accurately in the network. In this proposal, we explore and build intrusion detection system using deep neural network. We study the performance of the model on binary classification, multiclass classification and number of neurons and the impact of learning rate on the performance. We compared it with the existing machine learning approaches proposed by researches on the benchmark datasets KDD-Cup 99 and NSL-KDD. The results show that the DNN-IDS accuracy is higher than the other machine learning based algorithms.
Volume 11 | Issue 5
Pages: 353-362