A Predictive System for Supporting At-Risk Students’ Identification

Baneres D., Karadeniz A., Guerrero-Roldán A., Elena Rodríguez M.

Future Technologies Conference, FTC 2020, San Francisco, United States Of America, 5 - 06 November 2020, vol.1288, pp.891-904 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1288
  • Doi Number: 10.1007/978-3-030-63128-4_67
  • City: San Francisco
  • Country: United States Of America
  • Page Numbers: pp.891-904
  • Keywords: Artificial intelligence in education, At-risk students’ identification, Data mining, Early warning system
  • Anadolu University Affiliated: Yes


© 2021, Springer Nature Switzerland AG.Personalized learning has evolved during the last decade by means of using artificial intelligence. Datafication of education allows more detailed information gathering about students and create customized models to provide more accurate recommendations to them to better success in their learning process. This paper focuses on the description of the infrastructure to create a custom system to identify at-risk students, and particularly the backend system to create models based on machine learning algorithms to support such at-risk identification. The backend system has been designed in a general way to easily configure different models, and the infrastructure tried to meet the needs of the teachers and requirements of the system administrator such as scalability and data privacy. Finally, the system is run in a study case showing the results of an at-risk identification model in a real dataset of an online institution.