Intelligent System for Detection of Intrusion in Imbalanced Network Traffic

Abstract

Malicious cyberattacks can frequently hide among enormous amounts of regular data in networks with uneven traffic patterns. The identification of imbalance network traffic is difficult to find and making a challenge in terms of Signature building. Despite years of improvement, IDSs still struggle to increase detection accuracy. There are distinct machine learning or deep learning algorithms provides the better results and accuracy for imbalance network traffic. In this study, intrusion detection in unbalanced network traffic is accomplished using both machine learning and deep learning. This process a DSSTE algorithm to avoid imbalance problems. Initially the training set is pre-processed to modify imbalanced data and features are extracted. The proposed model is evaluated with trained data using multiple classification algorithms.