Comparative Analysis of Machine Learning Models for Intrusion Detection Systems
Vikrant Sharma · Panamerican mathematical journal. · 2025
Intrusion Detection Systems (IDS) play a crucial role in modern cybersecurity, leveraging machine learning (ML) to detect and mitigate cyber threats effectively. This study provides a comparative analysis of multiple ML-based IDS models, including XGBoost, Generative Adversarial Networks (GAN), Artificial Neural Networks (ANN), Support Vector Machines (SVM), Decision Trees, and Random Forest classifiers. The results indicate that XGBoost (98%) and GAN-based IDS (96%) achieve the highest accuracy, demonstrating superior adaptability in detecting sophisticated attacks. ANN and SVM also exhibit strong performance, while traditional classifiers such as Decision Trees and Random Forests struggle with complex attack patterns. Despite ML advancements, challenges related to data quality, computational efficiency, and evolving cyber threats remain. Future research should focus on hybrid ML approaches, adversarial learning, and real-time IDS deployment to enhance security frameworks. This study underscores the importance of adaptive ML-driven IDS models in mitigating cybersecurity risks.