Comprehensive Analysis of Machine Learning-Based Intrusion Detection Systems
Sonu Kumar, Abhirup Arindam, Aarti Gautam Dinkar, Anurag Singh Baghel, Neeta Singh · 2025
In this work comprehensive research on the prevention of network threats, with a primary focus on Intrusion Detection Systems (IDS). The research aims to detect and classify intrusions that compromise information integrity, availability, and confidentiality (privacy). The research leverages ML(Machine learning) models for intrusion detection and classification employing algorithms like Random Forest, Decision Tree, and Logistic Regression to increase dependability and efficiency. The models are evaluated on benchmark datasets, including KDD99, CSE-CIC-IDS2018 and UNSW-NB15, utilizing key features to improve performance and achieve better comparative results. Experimental results demonstrate the effectiveness of ML models in achieving high detection rates while maintaining robustness against diverse attack vectors. This work offers actionable insights for the integration of ML-based systems into modern cybersecurity frameworks, paving the way for intelligent, automated, and proactive threat management solutions.