Analyzing Sentiment and Emotion in Omicron-Related Social Media Content: An NLP Perspective

Abstract

This paper presents an approach to sentiment analysis using various text processing techniques on a dataset of textual data related to the Omicron variant. The study applies Natural Language Processing (NLP) methods, including data cleaning, stemming, and stopword removal, to preprocess the text. Subsequently, it employs Word Cloud visualizations to explore the frequency of words and hashtags in the dataset. Sentiment analysis is performed using the VADER sentiment analysis tool to categorize the sentiments expressed in the text into positive, negative, and neutral categories. The aggregated sentiment scores are analyzed to determine the overall sentiment trend within the dataset. The results indicate the predominant sentiment as positive, with detailed insights into the distribution of sentiments. The paper highlights the effectiveness of combining NLP techniques with sentiment analysis for understanding public opinion and trends in textual data.