EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Predicting graduation and dropout rates among engineering students using AI techniques

Ricardo França Santos, Mathis Berthet

Discover Education · 29.12.2025 · 5 (1)

DOI: 10.1007/s44217-025-01092-3 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Discover Education, online erschienen am 29.12.2025. Verfasst von Ricardo França Santos und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study introduces an AI model using a random forest algorithm to predict dropout risk among engineering students at the Instituto Politécnico (IPOLI) of the Federal University of Rio de Janeiro. The model provides academic performance, demographic information, and survey responses. Key factors linked to dropout are identified, providing a practical tool for early intervention and prevention. For instance, proactive mentoring could be initiated as early as week two for students flagged by the model, facilitating timely support. Feature importance analysis highlights strong predictors, such as early GPA and socioeconomic conditions, which are correlated rather than causal. The model allows institutions to identify at-risk students early and supports strategies to enhance retention.

Zitieren

APA

Santos, R. F., & Berthet, M. (2025). Predicting graduation and dropout rates among engineering students using AI techniques. Discover Education, 5(1). https://doi.org/10.1007/s44217-025-01092-3

Vancouver

Santos RF, Berthet M. Predicting graduation and dropout rates among engineering students using AI techniques. Discover Education. 2025;5(1). doi:10.1007/s44217-025-01092-3

BibTeX
@article{Santos2025,
  author = {Santos, Ricardo França and Berthet, Mathis},
  title = {Predicting graduation and dropout rates among engineering students using AI techniques},
  journal = {Discover Education},
  year = {2025},
  volume = {5},
  number = {1},
  doi = {10.1007/s44217-025-01092-3}
}
Kostenlos starten →