An Enhanced Network Security using Machine Learning and Behavioral Analysis

Abstract

With the advancement of the internet over the years, the number of attacks over the internet has also increased. A powerful intrusion detection system (IDS) is required to ensure the security of a network. A novel supervised machine learning system has been developed to classify network traffic, whether it is malicious or benign. To find the best model considering detection success rate, a combination of supervised learning algorithms and feature selection methods have been used. To evaluate the performance, the NSL-KDD dataset is used to classify network traffic using SVM and ANN supervised machine learning techniques. A comparative study shows that the proposed model is more efficient than other existing models with respect to intrusion detection success rate. Machine learning (ML) algorithms are frequently used to design effective attack detection (identity) structures for the effective mitigation and detection of malicious cyber threats at the host and community levels. Therefore, developing an accurate and sensible identification gadget will be a concern. The simulation results show that our projected achieving an excessive accuracy price of up to 97.52% for the dataset multiple times.