Unveiling Sentiments: Exploring Tweet and Text Analysis for Deeper Insights

This work presents a comprehensive exploration of sentiment analysis on social media data, with a specific focus on Twitter. The study evaluates three state-of-the-art classifiers-Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest (RF)-to categorize sentiments (positive, negative, neutral) based on specific criteria. Data cleansing and classification methods are applied to extract meaningful insights from unstructured social media data. Results show that SVM achieves the highest accuracy (92.67%) among the classifiers. It also discusses the strengths and limitations of each classifier in capturing the nuances of sentiment expressions. This work contributes to the understanding of sentiment analysis techniques on social media platforms