Open access
Sep 2026
Security Evaluation of Classical Machine Learning Models Under Poisoning, Evasion, and Model Extraction Attacks
A unified empirical evaluation of Logistic Regression, Linear SVM, Random Forest, and XGBoost models across image, text, and tabular classification domains indicates that model robustness depends on the interaction between the attack strategy, data representation, and model architecture.
Hannelore Sebestyen, Elisa Valentina Moisi, S. Coman et al.
· Applied Sciences · 0 citations