Evaluation of Network Intrusion Detection with Features Selection and Machine Learning Algorithms on CICIDS-2017 Dataset

Abstract

In the era of network Security, the Intrusion Detection System (IDS) plays an important role in information security. As the usability of the internet among the users in a wide area is increasing day by day so as the importance of security and to keep the system aware of the malicious activities is also increasing. In this paper we have decided to choose four feature selection algorithms i.e. CfsSubset Attribute Evaluator, Classifier Subset Evaluator with Naive Bayes, Classifier Subset Evaluator with J48 and Classifier Subset Evaluator with Decision Tree and the two different machine learning algorithms, namely OneR and REPTree. All these algorithms have been implemented in WEKA machine learning tool to evaluate performance. For experimental work, CICIDS-2017 dataset is used. First we select the features by feature selection algorithms after this each classification algorithm is tested with conducted dataset and finally results have been compared.