AN ADAPTIVE AND EXPLAINABLE ENSEMBLE LEARNING FRAMEWORK FOR MALWARE THREAT DETECTION IN CYBERSECURITY ENVIRONMENTS
International Journal of Drug Delivery Technology · 08.06.2026 · 16 (48s)
DOI: 10.25258/ijddt.16.48s.75 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 08.06.2026. Verfasst von Mahaboobunnisa S und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Cybersecurity is becoming a vital component in protecting the digital infrastructure from the sophisticated attacks. The growth of the malware variants is posing significant challenges to the conventional signature-based detection systems. The imbalance in cybersecurity datasets further is reducing the detection rate of the rare attacks. In addition, many of the models are deployed without the adequate validation in the real-time environments, which affects its reliability in practical applications. This study is proposing an adaptive ensemble learning framework that is combining the Logistic Regression, Support Vector Machine, and the K-Nearest Neighbors within a voting-based mechanism. The model is adopting Random Forest and the Naïve Bayes in strengthening the classification strength. The data preprocessing stage is including normalization, categorical encoding, and the outlier removal. Feature extraction is using the Principal Component Analysis to reduce dimensionality. in addressing the data imbalance, SMOTE-based synthetic data generation is applying. The framework is combining adversarial training and the explainable AI techniques such as SHAP that is providing interpretability. The proposing model is achieving an accuracy of the 97.8%, precision of the 96.9%, recall of the 95.6%, F1-score of the 96.2%, and the false positive rate of the 2.1% on benchmark datasets. The computational cost is reducing by 28% through the model optimization techniques such as pruning. The ensemble approach is improving the detection performance by approximately 6–9%, when it is compared with the individual classifiers. The model is showing stable performance under the simulated real-time conditions.
Zitieren
APA
S, M., & Vinod, V. (2026). AN ADAPTIVE AND EXPLAINABLE ENSEMBLE LEARNING FRAMEWORK FOR MALWARE THREAT DETECTION IN CYBERSECURITY ENVIRONMENTS. International Journal of Drug Delivery Technology, 16(48s). https://doi.org/10.25258/ijddt.16.48s.75
Vancouver
S M, Vinod V. AN ADAPTIVE AND EXPLAINABLE ENSEMBLE LEARNING FRAMEWORK FOR MALWARE THREAT DETECTION IN CYBERSECURITY ENVIRONMENTS. International Journal of Drug Delivery Technology. 2026;16(48s). doi:10.25258/ijddt.16.48s.75
BibTeX
@article{S2026,
author = {S, Mahaboobunnisa and Vinod, Viji},
title = {AN ADAPTIVE AND EXPLAINABLE ENSEMBLE LEARNING FRAMEWORK FOR MALWARE THREAT DETECTION IN CYBERSECURITY ENVIRONMENTS},
journal = {International Journal of Drug Delivery Technology},
year = {2026},
volume = {16},
number = {48s},
doi = {10.25258/ijddt.16.48s.75}
}
EUVIMED