Jordan’s reliance on imported fossil fuels has intensified the need for secure and sustainable energy alternatives, positioning nuclear power as a potential solution contingent on public acceptance. Such acceptance is critical to the success of nuclear initiatives and is shaped by perceived risks, benefits, trust, and social influences. This study examined determinants of nuclear energy acceptance in Jordan using Extended Protection Motivation Theory integrated with machine-learning techniques through the Orange data-mining platform. Survey data from 260 respondents were analyzed using Artificial Neural Networks (ANN), Random Forest Classifier (RFC), and Support Vector Machine (SVM). Results identified perceived benefits as the strongest predictor of acceptance, followed by trust in regulations, whereas subjective norms and social factors showed weaker influence. Under cross-validation, the Support Vector Machine achieved the highest accuracy, while the Artificial Neural Network performed better on the independent test set, indicating stronger generalization. These findings highlight the capacity of machine-learning methods to capture nonlinear behavioral patterns often overlooked in conventional approaches. From a governance perspective, results emphasize regulatory credibility, safety oversight, and effective communication of societal and environmental benefits in shaping public acceptance. Overall, the study provides insights to support evidence-based policymaking and regulatory capacity building for countries pursuing nuclear energy. • ANN, RFC, and SVM were implemented to analyze nuclear energy acceptance in Jordan. • Class imbalance was addressed using SMOTE, yielding robust cross-validated accuracy. • Perceived benefits were consistently identified as the strongest acceptance predictor. • Trust in regulations was ranked second, indicating the importance of regulatory credibility. • Knowledge and perceived risk showed moderate contributions to model performance.
Alhawamdeh et al. (Wed,) studied this question.