Magecart skimming attacks have emerged as a significant threat to client-side security and user trust in online payment systems. This paper addresses the challenge of achieving robust and explainable detection of Magecart attacks through a comparative study of Machine Learning (ML) models with a real-world dataset. Tree-based, linear, and kernel-based models were applied with hyperparameter tuning and feature selection, to distinguish between benign and malicious scripts. The models are supported by a Behavior Deterministic Finite Automaton (DFA), which captures structural behavior patterns in scripts, helping to analyze and classify client-side script execution logs. To ensure robustness against adversarial evasion attacks, adversarial training and evaluations were performed using attacks from Adversarial Robustness Toolbox and Adaptative Perturbation Pattern Method. In addition, concise explanations of ML decisions are provided, supporting transparency and user trust. Experimental validation demonstrated high detection performance and interpretable reasoning, demonstrating that traditional ML models can be effective in real-world web security contexts.

Magecart skimming attacks pose a significant threat to client-side security and user trust in online payment systems. This poster details an approach for the robust and explainable detection of these attacks, with a deep focus on the critical role of feature engineering in optimizing Machine Learning model performance. A suite of classifiers was evaluated under three distinct conditions to determine the most effective feature set: a comprehensive set of 103 features, a 60-feature subset selected via Random Forest importance scores, and an 80-feature subset identified through Pearson’s correlation. The findings demonstrate that the importance-based selection strategy yields a superior trade-off between model performance and computational efficiency. This is best exemplified by the Support Vector Machine model, which achieved a peak F1-score and recall when trained on the 60 most important features. The resilience of this optimized model was subsequently confirmed through extensive adversarial testing against a range of evasion attacks, where it consistently outperformed other models.

Machine Learning (ML) can be incredibly valuable to automate anomaly detection and cyber-attack classification, improving the way that Network Intrusion Detection (NID) is performed. However, despite the benefits of ML models, they are highly susceptible to adversarial cyber-attack examples specifically crafted to exploit them. A wide range of adversarial attacks have been created and researchers have worked on various defense strategies to safeguard ML models, but most were not intended for the specific constraints of a communication network and its communication protocols, so they may lead to unrealistic examples in the NID domain. This Systematization of Knowledge (SoK) consolidates and summarizes the state-of-the-art adversarial learning approaches that can generate realistic examples and could be used in real ML development and deployment scenarios with real network traffic flows. This SoK also describes the open challenges regarding the use of adversarial ML in the NID domain, defines the fundamental properties that are required for an adversarial example to be realistic, and provides guidelines for researchers to ensure that their future experiments are adequate for a real communication network.