A Comparative Study of Classification Algorithms for Sentiment Analysis of Covid-19 Vaccine Opinions Using Machine Learning

Textual data analysis and classification into positive, negative, or neutral emotions in natural language processing are complex. This study uses machine learning techniques to classify coronavirus and vaccination attitudes. This study used Random Forest (RF), Gradient Boosting Classifier (GBC), Logistic Regression (LR), and Decision Tree. The RF and GBC algorithms had an impressive 89% accuracy rate, while the LR and DT algorithms had 87%. Word processing and data pre-processing were needed before categorization. The outcomes of this research can help policymakers overcome possible impediments to vaccination initiatives. contain the infection and reduce its public health effect. The examination of Kaggle data, which includes many pandemics and vaccine remarks, emphasizes the necessity for immediate herd immunity against COVID-19. This crucial goal is crucial to limiting the virus's spread and minimizing its negative effects on society. The task requires acknowledging and resolving public concerns and building faith in the immunization campaign. This study uses machine learning approaches and compares their importance. The upcoming study will produce an app that categorizes sickness and vaccination attitudes. To design effective plans for Covid-19 immunization, governments and organizations must understand the challenges. It is important to note that the research only included English tweets. This constraint may reduce the credibility and generalizability of sentiment results. To further understand public perceptions regarding COVID-19 vaccinations, deep learning algorithms might be used to analyze larger Twitter datasets.