Student Performance Prediction on Primary and Secondary Schools-A Systematic Literature Review
Lorran Santos Rodrigues, Marcos dos Santos, Igor Pınheıro de Araújo Costa, Miguel Ângelo Lellis Moreira
Abstract
Student performance prediction is a relevant topic to schools and universities. By this means, they can act preventively and, also, allocate resources more accurately when at-risk students are identified. This research field has received interest in recent years. However, little has been mapped about the state-of-the-art in students performance prediction in middle and high schools. This work aims to fill the identified gap by conducting a systematic literature review. It was recognized that the interest in the topic, in the context of middle and high schools, has accompanied the tendency of research on student performance prediction in universities. However, the exploration of semi-structured data with deep neural networks, which is a recurrent theme in the universities context, is still incipient. Also, it was pointed out problematic aspects concerning some research in the field, along with the reported models and the main features used to predict student performance. Ultimately it was discussed future paths for research in the area.