Towards the Machine Learning Techniques Based Network Intrusion Detection System
Vivek Sahu, Deepesh Dewangan · 2024
The relentless threat of network intrusion jeopardizes data integrity and privacy, posing a critical challenge in safeguarding digital assets and ensuring the stability of interconnected systems. Identifying and thwarting sophisticated intrusion attempts is a constant battle for cyber security. This research discusses and enhancing the efficiency of NIDS through the utilization of machine learning techniques, with a primary focus on Random Forest. The study compares the performance of Random Forest against prominent counterparts, including Gradient Boosting, Light Gradient Boosting Machine (LGBM), Naive Bayes, Logistic Regression, and k-Nearest Neighbors (KNN), using the NSL-KDD datasets. Through rigorous experimentation, Random Forest emerges as the front-runner, showcasing an impressive accuracy of 99.59%. The findings demonstrate the superior predictive capabilities of Random Forest in identifying and classifying network intrusions.