The issue of climate change and air pollution represents a significant global challenge that demands critical attention. In 2017, the transportation sector in Indonesia accounted for approximately 46.58% of total energy consumption and contributed 53% of total exhaust emissions. Despite an 82.8% public interest in Electric Vehicles (EVs), doubts and concerns persist regarding the viability of battery technology, the availability of supporting infrastructure, and the cost of ownership. This study conducted a sentiment analysis of public perceptions of Electric Vehicles. The approaches employed in this study are Word2Vec and Graph Neural Network (GNN). The combination of Word2Vec and GNN was selected due to its advantages in understanding the semantic meaning of text and enhancing accuracy through the utilization of relational information between words. The objective of this sentiment analysis is to gain insights into public perceptions of Electric Vehicles (EVs) in Indonesia. The results of Word2Vec and GNN achieved an F1-score of 78.81% with an embedding size of 100, a window size of 9, and 200 epochs, higher than other comparable methods, including Word2Vec and CNN (70.50%), SVM (69.28%) and Naive Bayes (61.52%). The most effective model could serve as a reference for future studies on public acceptance of EVs.