Utilization of random forest classifier and artificial neural network for predicting the acceptance of reopening decommissioned nuclear power plant

•Machine Learning Algorithms (MLA) were utilized to predict the acceptance of reopening Bataan Nuclear Power Plant.•Knowledge about nuclear power plant had a positive effect on acceptance.•Perceived benefit had a higher effect on the acceptance than perceived risk.•RFC and ANN proved to be effective with accuracy rates of 97.00% and 93.43%, respectively.•The approach in this study can be applied and extended in predicting the acceptance of NPPs worldwide. The Bataan Nuclear Power Plant (BNPP) is one of the many decommissioned Nuclear Power Plant (NPP) globally and its reopening has led to different perceptions among Filipinos. It was established in 1984 but was not utilized due to political liability and safety reasons. This study aimed to predict factors affecting the acceptance of the BNPP by utilizing Machine Learning Algorithms (MLA). The MLAs utilized in this study were Decision Tree, Random Forest Classifier (RFC), and Artificial Neural Network (ANN) as a highlight to predict human behavior. 1,252 Filipinos voluntarily answered an online questionnaire which consist of 37 questions, leading to 46,324 datasets. MLA showed that Filipinos are knowledgeable about the benefits of NPPs, leading to the acceptance of the reopening of the BNPP. In addition, MLA indicated that perceived benefits (PB) was found to be the highest factor that affect the Filipino’s acceptance of the reopening of BNPP. Job opportunities, economic growth, lower and clean energy consumption, and sustainability were the indicators for the acceptance of the reopening of BNPP. Interestingly, the result showed that PB relatively outweighed the perceived risk of the BNPP. ANN and RFC proved to be effective with accuracy rates of 93.44% and 97.00%, respectively. Finally, the MLA approach in this study can be applied and extended in predicting the acceptance of NPPs worldwide.