Adversarial Cognition Machine Learning at the Frontlines of Cyber Warfare
1 Bachelor of Education, National University, Bangladesh.
2 Bachelor of Business Administration, National University, Bangladesh.
3 Master's in Genetics, Osmania University, India.
4 Bachelor in Law, Independent University Bangladesh.
5 Master's in Information Technology, Washington University of Science and Technology, Virginia, USA.
6 Master’s in Commerce, Jagannath University College, Dhaka, Bangladesh.
Research Article
World Journal of Advanced Research and Reviews, 2021, 12(02), 722-729
Publication history:
Received on 10 October 2021; revised on 23 November 2021; accepted on 28 November 2021
Abstract:
Advancements in cyber warfare have trended towards greater complexity, leading to the need for intelligent, adaptable defense mechanisms that can adapt and learn from behavior emulations within evolving network scenarios. The research paper, Adversarial Cognition: Machine Learning at the Frontlines of Cyber Warfare, presents a holistic machine learning framework for identifying and detecting malicious network traffic using structured traffic telemetry and behavioral indicators. Accordingly, the study relies heavily on cognitively inspired feature engineering, robust preprocessing techniques, and evaluation of a number of supervised learning algorithms to characterize adversarial behavior. We used a stratified train–test evaluation strategy to ensure reliable performance assessment against imbalanced classes of benign and attack traffic. The classification algorithms Logistic Regression, Random Forest, Gradient Boosting, SVC, and KNN were tested based on five different metrics—accuracy, precision score, recall score, F1-score, and ROC-AUC. Experimental results show that methods based on ensembles are more effective than linear and distance-based approaches when detecting activities of adversaries. Gradient Boosting surpassed all the models with an accuracy of 96.74% and a 0.9895 ROC-AUC score, but Random Forest and SVC came closest behind.
Keywords:
Cyber Warfare; Intrusion Detection Systems; Adversarial Machine Learning; Ensemble Learning; Gradient Boosting; Network Security; Binary Classification; Cyber Threat Detection
Full text article in PDF:
Copyright information:
Copyright © 2021 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0