Sentiment Analysis on Acceptance of COVID-19 Vaccine for Children based on Support Vector Machine

Abstract

Sentiment Analysis is a Natural Language Processing (NLP) branch that focuses on the analysis of the public opinions based on specific topics. The studies on sentiment analysis have been increasing since the COVID-19 pandemic in 2020. After the herd immunity has been reached around the world, the attention has shifted toward the children's COVID-19 vaccination program. It is beneficial to mine people’s opinions regarding this issue since kids are often perceived as vulnerable and need the parents’ consent. The machine learning based sentiment analysis has proven to be more efficient in the sentiment classification. This study aims to explore the capability of the Support Vector Machine (SVM) algorithm in the sentiment classification of the COVID-19 vaccination for children based on Twitter data. SVM has been one of the powerful algorithms, but never tested in this classification problem. The dataset for this project was scraped using Twitter API Tweepy based on the keywords such as “COVID vaccine children” and “5 to 11 vaccine”. The SVM model is based on the Linear Kernel and has been tested with the hold out method. The model has performed better in the balanced dataset with the implementation of oversampling by the SMOTE technique. A GUI prototype has also been developed using TKinter for the SVM classifier. The results have been divided into the data exploratory and the algorithm’s performance analyses. In this study, it is found that there are actually more people are supporting the COVID-19 vaccination for children. Meanwhile, based on the performance analysis, SVM has been able to classify the positive and negative tweets with an acceptable accuracy of 82%. The future work includes the scrapping of data from other social media platforms for larger demographics and also to compare SVM performance with other algorithms.