Youtube Comments Sentiment Analysis

Over the years, there has been a significant surge in textual information, leading to a burgeoning research interest. The contemporary focus lies in the intriguing field of sentiment analysis on YouTube comments. Despite the substantial volume of user comments and reviews on many videos, limited efforts have been directed towards extracting meaningful trends due to the inherent inconsistency and variable quality of information.In this study, we perform sentiment analysis on YouTube comments concerning popular topics, with a selection of various machine learning techniques and algorithms. The aim will be to highlight how sentiment analysis can show trends, seasonality, and forecasts, and offer an insightful view into the effects of real-world events on the public mood. The results display strong correlation views between users' sentiments toward corresponding keywords and the real-world events.The primary aim of this research is to assist scholars in identifying high-quality research papers on sentiment analysis. Our approach involves sentiment analysis of YouTube comments using citation sentences based on an existing annotated corpus consisting of 1500 citation sentences. Data cleansing involved the application of various normalization rules to eliminate noise from the comments in the corpus.