Paper Title
YOUTUBE COMMENT SENTIMENT ANALYSIS USING DEEP LEARNING TECHNIQUES

Abstract
With the rapid growth of social media, YouTube has become a key platform for content sharing and audience interaction. Comments on YouTube videos reflect users’ opinions, emotions, and preferences, but manual analysis is impractical due to their high volume and informal nature. This paper proposes a YouTube Comment Sentiment Analysis system based on deep learning. The system uses the YouTube Data API to collect comments, preprocesses them through text cleaning, emoji handling, and tokenization, and classifies sentiments using an LSTM model. It also offers result visualization through charts and enables CSV export. Experimental results show high accuracy, demonstrating the system’s effectiveness in tracking audience sentiment and extracting insights. Keywords - Sentiment Analysis, YouTube Comments, Deep Learning, LSTM, NLP, Social Media Analytics