Ensemble Learning Framework for Robust Malware Classification: Integrating Feature Optimized Voting Classifiers
Al Hassani, Oras, Subaramaniam, Kasthuri, Palaniapan, Sellappan and Bin Shibghatullah, Abdul Samad (2026) Ensemble Learning Framework for Robust Malware Classification: Integrating Feature Optimized Voting Classifiers. ICAROB2026, 31. ISSN 2188-7829
Abstract
In this paper, a new ensemble learning framework based on Random Forest, Gradient Boosting and meta-level Voting Classifier is proposed to improve the performance of malware detection. Architecture takes advantage of sophisticated feature engineering approaches such as entropy-based feature selection and structural byte analysis through the ClaMP Integrated-5184 dataset. Experimental results have shown that the combined ensemble-based model outperforms classical classifiers Logistic Regression, Support Vector Machines and K-Nearest Neighbours by a large margin on all the major metrics. In this line, we demonstrate classifier fusion as an appealing paradigm for nextgeneration cybersecurity systems, finding a way to balance the accuracy-adaptability trade-off of ongoing malware detection.
Actions (login required)
 |
Edit Item |