Automated Academic Supervisor Allocation Using the C4.5 Decision Tree Algorithm: A Scalable Web-Based Solution

Abstract

Efficient allocation of academic supervisors is a critical yet challenging process in higher education, often hindered by mismatches in expertise and uneven workload distribution. This study introduces a web-based recommendation system leveraging the C4.5 decision tree algorithm to address these issues. The system generates data-driven, accurate recommendations by assessing supervisor expertise, workload, and alignment with student research topics. The system emphasizes scalability and modularity and was developed using the Laravel framework and following the waterfall development model. Functional testing demonstrated a 96.7% accuracy rate for recommendations, while usability testing reflected high user satisfaction, with an average score of 92% for ease of use and relevance. These results highlight the system’s effectiveness in optimizing supervisor assignments, enhancing administrative efficiency, and providing a scalable solution for educational institutions. Future work will further integrate diverse data sources and AI-driven features to improve adaptability and responsiveness to evolving academic demands.