Academic performance forecasting model based on artificial intelligence at the Faculty of Engineering - Systems and Informatics of the Continental University

Elena; Paula S.; Ríos E.; Felipe; Nestor; Meza G.; Jose; Porfirio; Farías G. · ACM International Conference Proceeding Series (2023) · DOI: 10.1145/3638264.3638283

Resumen

The research established a central objective to implement an academic performance forecasting model based on artificial intelligence to determine the level of incidence factors in the management of academic performance and design the predictive model based on artificial intelligence with the use of classifiers and exploratory analysis to show the distribution of variables and identify patterns where interventions should be made to have better academic performance. The present research has a quantitative approach, correlational scope, non-experimental cross-sectional design, non-parametric statistics, working with a sample of 317 university students, and a diagnostic instrument with Cronbach’s α of 83.8%. We worked with the Weka software and the Bayesian classification algorithm J48, obtaining an accuracy level of 76% in predicting academic performance, indicating the influential factors for high academic performance. The results show a weak significant correlation between academic performance and factors. It was concluded that through the use of the artificial intelligence forecast model, the university can identify the influential factors in academic performance, and they can improve it with extracurricular activities and work with the least significant factors through training or psychological well-being workshops to avoid dropout rates and thus have good performance and the development of their capabilities.

Instituciones: Universidad Continental