Omicron Tweet Sentiment Analysis Using Ensemble Learning
Keywords:Machine Learning, Sentiment Analysis, Twitter, Omicron, COVID-19, Tweets, NLP, Big data
In 2019, the COVID-19 pandemic took the world by storm, resulting in far-reaching impacts on education, economics, and health. As the coronavirus epidemic progressed, new mutations such as the Beta, Delta, and Omicron variants developed, causing fear and anxiety among the public. According to World meter, approximately 6 million people have so far died from COVID-19 and its variants. On November 24, 2021, the “SARS-CoV-2” omicron strain was first observed in South Africa and has since spread to over 57 countries. This study provides an analysis of the sentiment and behavior of people toward the omicron variant. We propose a method for performing sentiment analysis on Twitter data related to the omicron strain. Natural Language Processing techniques are used in Python to extract optimized features from the omicron tweets, creating a dataset that is then used by Machine Learning tools to train various models. The dataset is used to classify user emotional behavior into Neutral, Negative, and Positive categories using six machine learning classifiers: Naive Bayes, Random Forest, Decision Tree, Support Vector Machine, Voting and Stacking techniques Classifier. Accurately measuring the results of the predictions is the goal. According to the study's findings, when compared to other classifiers, the ensemble voting classifier had a performance accuracy of 85.33% and the ensemble stacking classifier had a performance accuracy of 87.5%.
How to Cite
This is an open Access Article published by Research Center of Computing & Biomedical Informatics (RCBI), Lahore, Pakistan under CCBY 4.0 International License