Simple Perturbations Subvert Ethereum Phishing Transactions Detection: An Empirical Analysis
This reveals critical security vulnerabilities in blockchain fraud detection systems, which could undermine trust in cryptocurrency platforms.
The paper investigated how simple single-feature adversarial attacks can subvert machine learning models (Random Forest, Decision Tree, and K-Nearest Neighbors) used for detecting Ethereum phishing transactions, finding that these models are highly vulnerable with performance metrics like accuracy and F1-score significantly degraded.
This paper explores the vulnerability of machine learning models, specifically Random Forest, Decision Tree, and K-Nearest Neighbors, to very simple single-feature adversarial attacks in the context of Ethereum fraudulent transaction detection. Through comprehensive experimentation, we investigate the impact of various adversarial attack strategies on model performance metrics, such as accuracy, precision, recall, and F1-score. Our findings, highlighting how prone those techniques are to simple attacks, are alarming, and the inconsistency in the attacks' effect on different algorithms promises ways for attack mitigation. We examine the effectiveness of different mitigation strategies, including adversarial training and enhanced feature selection, in enhancing model robustness.