Predicting the Popularity of YouTube Videos: A Data-Driven Approach

This paper presents a data-driven approach to predict the popularity of YouTube videos. The existing conducted studies focus on identifying whether the video is trending currently or after a month of publishing. Where this study aims to leverage a comprehensive dataset of YouTube videos, analyze various features and employ machine-learning techniques to forecast next-hour video popularity accurately. By considering factors such as video metadata and engagement metrics, we develop predictive models that can assist content creators and marketers in understanding the likelihood of a video’s success on the platform. The results demonstrate the efficacy of our approach and provide practical insights into predicting YouTube video popularity.