Improvement in the performance of deep neural network model using learning rate
Kanupriya Arora, Ritu Chauhan · 2017 Innovations in Power and Advanced Computing Technologies (i-PACT) · 2017
This The core aim of this paper is to find out the accuracy of SDN network, in this we applied different approaches to increase the efficiency of the network, so that the performance as well as the accuracy will improve Different classifier are applied to form a model, in which pre processing and transformation are the essential phases before developing a model. After analyzing the classifier like NaiveBayes, Support Vector Machine, J48 and Deep Neural Network we found our Deep Neural Network with learning rate variation prove to be the best classifier among the all. In our work we are using the basic ten features out of the several features of NSL-KDD Dataset that are useful to build the network to predict the attacks in the intrusion detection system.