A Recommendation System Based on Early Academic Performance Prediction and Student Classification: Utilizing Artificial Intelligence and Mobile-Based Application

Abstract

In this paper, we explore the idea that categorizing students according to their early academic results can effectively prevent academic failure and enhance success in schools. Our objective is to offer appropriate educational strategies, learning methods, and resources. We introduce a method designed to improve student learning experiences and increase their high school success. For validation, we gathered a dataset from the School Life Management Software, containing data on the personal information and academic performance of 840 students from 2018 to 2021. Using this data, we developed a predictive model. We then compared the academic outcomes forecasted by our model with the actual results of the students for the 2021–2022 academic year. This comparison showed that our model can accurately predict early student academic performance and outcomes. Integrating our predictive model with a student classification system allows us to suggest effective strategies for enhancing student performance and avoiding academic failure, thereby improving the overall academic experience. In addition to the predictive model, we have developed a mobile application that operationalizes our findings. This application serves as a tool for students and educators, utilizing the predictive model to provide real-time academic performance forecasts. The app not only predicts outcomes but also suggests personalized strategies and resources to support students’ learning journeys.