| Abstract | Quality management and continuous improvement have become increasingly
practiced in worldwide industries and organizations. Recently, the higher education sector
has been gradually moving towards Quality Management as well. The academic success
and retention of university students are major questions for universities worldwide, and
many retention programs have been designed to remedy issues of students-at-risk and early
dropouts since a university’s academic productivity and efficiency are heavily linked to the
institution’s graduation and retention rates. In this way, it important for the university to
find ways to identify students needing help and provide them with support. To that end,
this research work aims to develop a framework through which the management can
identify students-at- risk as early as possible, and to ensure that they are offered appropriate
support in a timely manner. In addition, three machine learning-based prediction models
have been proposed for predicting course difficulty level (CDL). The accuracy of the
proposed prediction models is assessed by using a real dataset collected from the students
of the college of engineering in Qatar University, Doha-Qatar |
