YT Sentiment Explorer: A Sentiment Analysis System for YouTube Comments Using Machine Learning Techniques

With the growing popularity of social media platforms, analyzing public opinion has become essential for content creators, marketers, and researchers. YouTube, being a major source of user-generated content, contains valuable insights within its comment sections. This paper introduces YT Sentiment Explorer, a web-based application that automates the process of sentiment analysis on YouTube comments. The system utilizes the YouTube Data API to extract comments, applies Natural Language Processing (NLP) for text preprocessing, and employs both lexicon-based (TextBlob) and machine learning-based (Logistic Regression) approaches for sentiment classification. Comments are categorized into positive, neutral, or negative sentiments and are visualized using pie charts, bar graphs, and word clouds through an intuitive user interface built with Flask. Experimental results show high accuracy and user satisfaction, indicating the tool’s effectiveness in real-time sentiment analysis. The application offers a scalable and efficient solution for opinion mining in the context of video-based digital content.