An Efficient Intrusion Detection Model with Novel Machine Learning Stacking Ensemble Techniques
Thotakura Venkata Sai Krishna, P. Seetha Lakshmi, M. Prathyusha, M. Swapna, J. Kanimozhi, K. Venkataramana, Raja Kumar Murugesan, Sukhminder Kaur · BENTHAM SCIENCE PUBLISHERS eBooks · 2025
Cybersecurity threats are becoming increasingly sophisticated, necessitating the development of robust intrusion detection systems. Traditional methods for managing intrusion detection may not consistently yield satisfactory results. The adoption of Majorization-Minimization (MM) Machine Learning (ML)- based methods in intrusion detection is crucial due to the limitations of conventional techniques. This chapter proposes the development of an ML-based Intrusion Detection model that seamlessly integrates ML ensemble techniques to enhance efficiency in identifying cyber threats. A dataset from Kaggle was used for experiments. Initially, Several ML classifiers, namely K-Nearest Neighbor (KNN), Naïve Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF), were applied for Intrusion Detection and achieved the best accuracy of 91% with RF. To increase accuracy, ensemble learning was used by combining several ML algorithms, achieving increased accuracy compared to a single ML model. Two ensemble models, namely Cost-sensitive Stacking and Ensemble Distillation, are proposed, achieving accuracies of 94% and 96%, respectively. The experiments show that the proposed ensemble methods outperform conventional approaches for Intrusion Detection.