Sentiment Analysis of YouTube Comments for Understanding User Engagement in Promotional Videos

NLP and machine learning research now has more options thanks to the growing amount of textual data. Because of the uneven quality of the data available, sentiment analysis of YouTube comments has received less attention than it deserves. Our objective is to find patterns, identify seasonal shifts, and predict future trends by using machine learning techniques to analyze attitudes expressed in comments on popular issues. This will provide insight into how public opinion is influenced by real-world occurrences. Our results validate a strong relationship between sentiment patterns and pertinent actual occurrences. The purpose of this study is to help academics find excellent research papers for sentiment analysis. We performed our research on a carefully selected corpus of 1,500 citation sentences that were first normalized to remove noise. Then, we developed a classification system using six machine learning algorithms: Random Forest (RF), Naive-Bayes (NB), the Support Vector Machine (SVM), Logistic Regression (LR), the Decision Tree (DT), and K-Nearest Neighbor (KNN). The system’s effectiveness was evaluated through the use of measures like the Accuracy Score and F-score.