Dataset Student and Course Performance for Heterogeneous Institutional Monitoring

This dataset supports the study “Adaptive Ontology-Enabled Data Retrieval Model for Learning Analytics Integration Across Heterogeneous Educational Platforms.” It addresses the challenge of integrating heterogeneous educational data sources—including Learning Management Systems (LMS), Student Information Systems (SIS), and MOOC platforms—into a unified repository for learning analytics. The research hypothesizes that an ontology-driven, graph-based model can more effectively integrate diverse data and enable accurate, flexible retrieval compared to traditional relational database approaches. The dataset is organized into eight sheets: Coursegroup – enrolment records linking students, lecturers, faculties, and programmes. Prerequisite – course metadata, credit hours, and prerequisite structures. Courseperformance – grade distributions across cohorts, lecturers, and groups. Studentperformance – detailed individual assessment scores, totals, and grades. Attendance – monthly student attendance by course and semester. Classactivity – engagement indicators measured through discussion posts. MOOCstudentprofile – learner enrolment, completion dates, time spent, progress, and certification. MOOCstudentperformance – quiz, test, and exercise results across MOOC modules. These sheets capture performance and engagement across conventional courses, classroom activities, and MOOC environments. Notable Findings - The ontology-based retrieval model successfully harmonized heterogeneous data without schema conflicts. - Confusion matrix evaluation achieved >99% accuracy, >98% precision, 100% recall, and >99% F1-scores. - No false negatives and only a small number of false positives were recorded. The dataset supports diverse analytics tasks such as tracking outcomes, correlating attendance with performance, identifying at-risk students, and comparing MOOC and traditional learning. How the Data Can Be Interpreted Records represent validated student/course performance data, interpretable at three levels: - Course: outcomes, grade distributions, prerequisites. - Student: assessments, attendance, engagement. - MOOC: persistence, time investment, quiz/test outcomes. Ontology cross-linking enables flexible queries, e.g., “Which students with poor attendance underperformed?” or “How does MOOC completion relate to course grades?” This dataset highlights the value of ontology-based integration for scalable, accurate, and flexible learning analytics across heterogeneous educational platforms.