OBIEZE, CHINONYELUM (2026) DEVELOPMENT OF A PERSONALIZED YOUTUBE-BASED SKILL-PATH RECOMMENDER SYSTEM USING MACHINE LEARNING. Other thesis, Godfrey Okoye University, Enugu Nigeria.
![[img]](https://eprints.gouni.edu.ng/style/images/fileicons/text.png) |
Text
Development of a Personalized YouTube-Based Skill-Path Recommender System.pdf Download (1MB) |
Abstract
The rapid growth of YouTube as a learning platform has improved access to educational content but the lack of a pedagogically-structured path for learners introduces a very big challenge. YouTube’s default recommendation system is primarily engagement-driven, resulting in fragmented learning experiences as it only recommends video content based on high user engagement. This study presents YouLearn: a personalized YouTube-based skill-path recommender system designed to generate structured learning pathways aligned with user goals. The system employs a hybrid approach combining content-based filtering, a Logistic Regression-based skill difficulty classifier model, a collaborative filtering model using Singular Value Decomposition (SVD) and a concept dependency graph that sequences YouTube videos in three categories of “beginner”, “intermediate”, and “advanced.” Video data was collected through the YouTube Data API v3 and preprocessed using Text Frequency- Inverse Document Frequency (TF-IDF) techniques. Due to the absence of YouTube real user interaction data, a Coursera dataset containing Coursera_courses.csv and Coursera_eviews.csv was used as a proxy for training the personalization model. Evaluation using F1-score, RMSE, MAE, Precision@K, and Recall@K demonstrates the system’s effectiveness in delivering structured and relevant recommendations for efficient learning.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Faculty of Computing And Information Technology (FACIT) |
| Depositing User: | Cynthia Ugwuoti |
| Date Deposited: | 24 Jul 2026 13:12 |
| Last Modified: | 24 Jul 2026 13:12 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/5977 |
Actions (login required)
 |
View Item |
