Predicting ratings of YouTube videos based on the user comments

Abstract

This project attempted to develop a model to predict the rating of a YouTube video based on the user comments. We extracted the user comments from many YouTube videos to make the sentimental analysis. The keywords were extracted from the user reviews using the Natural Language Processing technique, and those reviews were categorized into positive or negative predicated on the sentimental analysis. The Naïve Bayes model was trained to utilize the user reviews extracted from YouTube to presage the rating of a video. The model was tested on original datasets, and the precision of that was evaluated respectively. Conclusively, one conclusion has been met that the rating of a video cannot be presaged through the user comments. The performance of the model is decent enough compared to the subsisting models in the literature. YouTube sanctions extract an inhibited number of user comments, and hence, this factor could negatively affect the rating presage's precision.