Educational data mining: An intelligent system to predict student graduation AGPA

Educational Data Mining (EDM) is an emerging research area, in which techniques are applied for exploring data educational systems. Since Accumulated Grade Point Average (AGPA) is crucial in students’ professional lives, having data-driven profiles for the students who are likely to graduate with low AGPA is an interesting and challenging problem. Identifying these kinds of students accurately enables educational institutions to improve the students’ level by providing them with special academic guidance and tutoring. In this paper, using a large and feature-rich dataset of student marks in the foundation stage, we developed a model to predict the graduation AGPA of students of Al Ain University of Science and Technology (AAU). The prediction process is done through employing neuro-fuzzy inference systems. The dataset used to determine the model quality and validity consists of 200 students’ records from two colleges, Law and Business Administration. The model was trained with 150 training samples and tested with 50 samples, which were not included within the training period. The experimental results showed a high level of accuracy of 97%. This accuracy revealed the suitability of neuro-fuzzy inference systems in predicting the students’ AGPA.