Spark-based Big Data Sentiment Analysis of Social Media Comments

YouTube is widely regarded as the most crucial place for YouTubers to communicate their idea with the rest of the world through the internet. Big Data is a massive accumulation of data in terms of volume, diversity, and velocity. YouTube is the most important source of big data. The video on YouTube gets comments. Typically, a YouTuber attempts to provide followers with the most outstanding content possible by reviewing previous videos' comments. On average, the number of comments can reach ten thousand, making it nearly impossible to read through each and understand viewers comments. This work presents a spark-based model that extracts the comments from YouTube on a specific video from an education and fitness domain. A Machine Learning Classification Model is added to the extracted input to provide sentiment analysis of posted comments. Through the result, the YouTuber has a better grasp of the viewers' thoughts, which in turn helps them better understand their audience. The findings include a comparison of various machine-learning techniques for sentiment classification. The comparison of the proposed result helps content creators improve the channel's decision-making and revenue-generation performance.