Vaccine Tweets Analysis Using Naive Bayes Classifier and TF-IDF Techniques

This paper explores the application of natural language processing (NLP) and machine learning techniques to sentiment analysis on a dataset of tweets on COVID-19 vaccines. The dataset was obtained from Kaggle and covers the full course of the immunization program. The tweets were cleaned using a variety of preprocessing approaches, such as handling contractions, removing URLs and user handles, and adjusting punctuation. The Text Blob library was used to assign sentiment ratings, while the TF-IDF technique was used to carry out feature extraction. The revised data was used to train a Naive Bayes classifier, which predicted the sentiment labels for every tweet. To evaluate the model’s performance, evaluation criteria such F1 score, accuracy, precision, and recall were used. The study’s findings provide insightful information on how the general public feels about COVID-19 vaccinations.