Pragmatic monitoring of learning recession using log data of learning management systems and machine learning techniques.

Pragmatically, machine learning techniques can improve educators' capacity to monitor students' learning progress when applied to quality data. For developing countries, the major obstacle has been the unavailability of quality data that fits the purpose. This is partly because the in-use information systems are either not properly managed or not accessible due to ethical issues and bureaucratic processes and procedures. System log data, which is readily available for educators who use electronic learning management systems, provides opportunities for applying machine learning techniques and devising practical solutions. This study analyzed real system log data as a viable alternative for potential use in machine learning based monitoring of students who are receding in their learning. The analysis proposed several indicators of recession, and how they can be combined to ease monitoring and visualization. The proposed indicators and visualization can be used by educators to monitor students and intervene proactively. [ABSTRACT FROM AUTHOR]