Two-Layered Machine Learning Approach for Sentiment Analysis of tweets related to Electric Vehicles

This study aims to perform Sentiment Analysis (SA) on social media data related to Electric Vehicles (EVs) using two layered unsupervised and supervised machine learning methods. A large dataset of tweets mentioning EVs is collected and preprocessed by removing stop words, emojis, and URLs. Unsupervised machine learning algorithms are applied to the processed data to identify sentiment patterns in the tweets. The results obtained from this method identifies distinct sentiment based on the words used in the tweets and label the same. The labeled data is further used to develop a model using various supervised learning algorithms for SA. The experimental results show that the Random Forest classifier has outperformed the other algorithms in terms of improved accuracy, precision, recall, and F-1 score. This work provides insights into the sentiment of EV users and non-users and thus can be used by policymakers and manufacturers to improve the adoption of EVs.