Use of AI Applications to Learn the Sentiment Polarity of Public Perceptions: A Case Study of the COVID-19 Vaccinations in the UAE

Abstract

The study examines the public perception of COVID-19 vaccinations in the United Arab Emirates (UAE) through the analysis of Twitter data. With the goal of distinguishing between factual information and misinformation, the research utilizes artificial intelligence algorithms to collect, process, and analyze social media posts related to COVID-19 vaccines. Python serves as the primary programming language, with tools such as Twitter API and SNScrape used for data collection, and libraries like Pandas, NumPy, NLTK, Scikit Learn, Matplotlib, Seaborn, and TensorFlow employed for data preprocessing, analysis, and model building. Through sentiment analysis and clustering techniques, the study evaluates the sentiments expressed in tweets and identifies common themes and perceptions surrounding different vaccines, including Pfizer-BioNTech, AstraZeneca, Sinopharm, and Sputnik V. The analysis reveals insights into public sentiment towards vaccinations, highlighting factors such as efficacy rates, availability, and safety concerns. Furthermore, the study presents a sentiment analysis model and compares its performance with the SMOTE algorithm for clustering tweets based on sentiment. Results indicate that the proposed model achieves higher accuracy in sentiment classification compared to SMOTE. The study underscores the importance of understanding public perceptions in shaping effective vaccination strategies and combating misinformation during public health crises. Future research directions include exploring the impact of vaccination campaigns and policy measures on public sentiment and behavior.