The central and state governments in India aim to promote sustainable transportation systems by encouraging electric mobility. However, the acceptance of electric vehicles (EVs) depends on public perceptions and attitudes and understanding of various EV features. Since EVs are expected to change the transportation sector significantly, it is essential to study how people's attitudes toward various features of EVs would influence the adoption rate. The main objective of this study is to analyze the attitudes of Indians toward various EV features by utilizing EV-related tweets. This study provides a detailed conceptual framework for extracting public sentiments toward EVs and their attributes using natural language processing techniques and machine learning algorithms. The Random Forest classifier with TF-IDF vectorizer performed better compared to the other machine learning classifiers. Also, we have used the Latent Dirichlet Allocation (LDA) approach to extract the most discussed features of EVs. The results indicate that the sentiments of the Indian public are, in general, positive toward EVs, and sentiment variations are observed on several features of EVs. Based on the research findings, some policy-related insights are presented regarding price, charging infrastructure, technology, environmental concerns, and driving range to promote EVs. The results will assist policymakers, vehicle manufacturers, and private enterprises in better understanding key EV features and consumer attitudes to boost EV adoption rates.