Date of Award
5-2025
Culminating Project Type
Starred Paper
Styleguide
ieee
Degree Name
Computer Science: M.S.
Department
Computer Science and Information Technology
College
School of Science and Engineering
First Advisor
Jalal Khalil
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
View Prediction, ML, ANN
Abstract
YouTube has experienced tremendous growth and widespread popularity, with the potential to impact billions of lives worldwide as its audience continues to expand each day. With this growth, accurately predicting video popularity has become increasingly valuable for creators and analysts seeking to optimize content engagement. This study focuses on comparing the performance of various machine learning (ML) models and artificial neural networks (ANN) for predicting YouTube video view counts using metadata features such as like count, comment count, video duration (seconds) and subscriber count. A dataset of 41,894 rows was collected from the Apify YouTube scraper and Kaggle, incorporating key metadata attributes. The study evaluates three machine learning, and ANN models using performance metrics including Mean Squared Error (MSE), R² score, and Median Absolute Error (MedAE). The objective is to determine which model provides the most accurate predictions and to analyze the significance of different metadata features in influencing view counts. By systematically comparing traditional ML models with ANN, this research offers insights into their effectiveness for video popularity prediction, helping creators and analysts better understand engagement trends on YouTube.
Recommended Citation
Ghimire, Abhishek, "YouTube Case Study: Comparative Analysis of ML and ANN Models for View Prediction" (2025). Culminating Projects in Computer Science and Information Technology. 57.
https://repository.stcloudstate.edu/csit_etds/57
Comments/Acknowledgements
I want to express my sincere gratitude to the St. Cloud State University computer science faculty for their consistent support and assistance during my academic journey. I am particularly grateful Dr. Jalal Khalil, my advisor, for his insightful aid and guidance, which was highly beneficial in completing my Starred Paper. I would like to sincerely thank Dr. Adriano Cavalcanti and Dr. Jie H. Meichsner, members of my committee, for their invaluable contributions.
I would also like to express my heartfelt gratitude to my family and friends for their unwavering support and the nurturing environment they provided, enabling me to thrive. Their constant belief in my ability, even in times of difficulty, has been the basis of my accomplishments.