Honeypot-Driven Machine Learning Framework for Detecting Cryptojacking and Phishing Attacks Using Network, URL, and Host-Based Security Features
Journal of Intelligent Decision Making and Information Science · 03.10.2026 · 3 (16s) · 2307-2324
DOI: 10.17762/jidmis.v3.5173 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 03.10.2026. Verfasst von Venkatraman Viswanathan. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Phishing and cryptojacking exploit different attack surfaces and therefore require distinct but coordinated forms of cyber evidence. This study develops and evaluates a honeypot-driven machine-learning framework that combines public phishing URL data, cryptojacking network-flow data, host-performance telemetry, and governance-controlled honeypot indicators within a modular decision-support architecture. URL and webpage features are processed by a phishing module, whereas network-flow and host-resource variables support cryptojacking and anomaly detection; honeypot logs enrich both modules with ports, commands, requested URLs, payload metadata, and indicators of compromise. The workflow incorporates leakage-aware preprocessing, modular feature engineering, baseline and ensemble classifiers, anomaly detection, explainability, and analyst-oriented decision outputs. Among the reported phishing models, XGBoost achieved the highest accuracy (0.981) and F1-score (0.977), while LightGBM produced the lowest false-positive rate (0.015) and a ROC-AUC of 0.996. For cryptojacking detection, XGBoost achieved 0.971 accuracy, a 0.970 F1-score, and a 0.992 ROC-AUC. The framework contributes an information-science perspective by preserving heterogeneous feature spaces, connecting model evidence to threat-intelligence outputs, and supporting reproducible, explainable, and governance-aware cybersecurity decision making.
Zitieren
APA
Viswanathan, V. (2026). Honeypot-Driven Machine Learning Framework for Detecting Cryptojacking and Phishing Attacks Using Network, URL, and Host-Based Security Features. Journal of Intelligent Decision Making and Information Science, 3(16s), 2307-2324. https://doi.org/10.17762/jidmis.v3.5173
Vancouver
Viswanathan V. Honeypot-Driven Machine Learning Framework for Detecting Cryptojacking and Phishing Attacks Using Network, URL, and Host-Based Security Features. Journal of Intelligent Decision Making and Information Science. 2026;3(16s):2307-2324. doi:10.17762/jidmis.v3.5173
BibTeX
@article{Viswanathan2026,
author = {Viswanathan, Venkatraman},
title = {Honeypot-Driven Machine Learning Framework for Detecting Cryptojacking and Phishing Attacks Using Network, URL, and Host-Based Security Features},
journal = {Journal of Intelligent Decision Making and Information Science},
year = {2026},
volume = {3},
number = {16s},
pages = {2307--2324},
doi = {10.17762/jidmis.v3.5173}
}
EUVIMED