With the rapid adoption of electric vehicles, rational planning for charging station locations has become increasingly important. This study proposes a data-driven prompt learning–based approach for evaluating the suitability of charging station placement. By transforming regional features into natural language text, we introduce the Prompt-Enhanced Transportation RoBERTa (PETRoberta) model based on prompt learning to capture the relationship between regional characteristics and charging station operational conditions. Using Wuhan as a case study, we compare our method with several baseline models. The results show that PETRoberta achieves the highest average accuracy of 91.76% on the test set under five-fold cross-validation and reaches state-of-the-art performance across multiple evaluation metrics. Furthermore, ablation experiments demonstrate that the number of points of interest (POI) is the key factor influencing charging station operational efficiency. The confusion matrix also verifies that PETRoberta achieves the highest accuracy within each individual category. Overall, our method can effectively support charging station layout planning and make a significant contribution to the wider adoption and promotion of electric vehicles.
Tang et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: