Dự báo kết quả học tập bằng kỹ thuật học sâu với mạng nơ-ron đa tầng

Abstract

Student performance prediction is one of the most concerned issues in the field of education and training. The prediction of courses results enables students to select courses appropriately. Moreover, this helps education managers and lecturers to indicate the students who should be monitored and supported to complete the courses with good results. Therefore, the student performance prediction is expected to reduce formal warnings and expulsions from universities due to students’ poor performance. In this study, a method was proposed to predict student performance using deep learning techniques exploring and mining data from universities’ student information system. From collected data, the data was analyzed and pre-processed before fetching them into a multi-layer perceptron to do prediction tasks. The obtained results from the proposed model reveal rather accurate forecasts; hence, the model is expected to apply in practical cases.