Role of Machine Learning Algorithms in Analyzing Data of Higher Education Students to Improve Academic Performance | ||||
| Journal of Communication Sciences and Information Technology | ||||
| Volume 8, Issue 1, June 2025, Page 8-26 PDF (1.15 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/jcsit.2025.390804.1020 | ||||
 View on SCiNiTO | ||||
| Author | ||||
Ismail Hagag 
| ||||
| EL Madina Higher Institute of Administration and Technology, Giza, Egypt | ||||
| Abstract | ||||
| This research examines the transformative potential of machine learning algorithms in evaluating and improving academic performance among higher education students. The study employs advanced data analytics techniques to analyze academic records of 1,000 students across three fundamental disciplines: Information Systems, Programming, and Databases. Using unsupervised K-Means clustering, students are systematically categorized into distinct performance tiers, revealing significant correlations between academic achievement and key demographic factors. The findings demonstrate how data-driven insights can enable educational institutions to implement targeted interventions, optimize resource allocation, and ultimately elevate overall academic standards. The research contributes to the growing body of knowledge on educational technology while providing practical frameworks for performance enhancement in higher education settings. The measurement and improvement of academic performance remain central concerns in higher education worldwide. As institutions grapple with increasing student diversity, evolving pedagogical demands, and accountability pressures, innovative approaches to performance analysis have become imperative. Traditional assessment methods often fail to capture the complex interplay of factors influencing student achievement, necessitating more | ||||
| Keywords | ||||
| Educational Data Mining; Predictive Analytics; Student Performance Modeling; Academic Intervention Strategies; Machine Learning in Education | ||||
|
Statistics Article View: 66 PDF Download: 49 |
||||
