Evaluating Sentiment Analysis Models for Enhancing Electric Vehicle Charging Infrastructure Planning

As electric vehicles (EVs) become more prevalent, the demand for efficient, user-friendly charging infrastructure grows. While much research focuses on the technical and economic aspects of EV charging, user feedback is often overlooked. This study uses sentiment analysis, comparing rule-based approaches like TextBlob with advanced deep learning models such as BERT, RoBERTa, and XLNet, to analyse survey data. The findings highlight key user challenges, including accessibility, reliability, and waiting times at public charging stations, demonstrating how sentiment analysis can inform infrastructure improvements and guide planning. This research offers a datadriven, user-centred approach to optimising EV charging networks, supporting informed decision-making for policymakers and service providers to enhance user satisfaction and encourage EV adoption.