to improve students' services
What can AI do to help promote academic success?
In the STARAPP project, AI was used on students’ data to try to flag out students at risk of academic insuccess. The machine learning model was applied to data collected in the University of Camerino and the STARAPP team worked with different techniques to try to understand which variables can predict the risk of academic disengagement.
Supervised and unsupervised Machine Learning Techniques were used. Full details on the methodological aspects and on the data used are available here: …..
Results show that data were able to predict students’ academic achievement and that the most predicting variables were related to:
– attendance to lectures,
– interest in the topic of the exams,
– participation to the academic life (use of the canteen, etc.).
Where do we go with the machine learning results?
Universities can submit their data to find out the percentage of risk of their students’ population. Full instructions on how to run the STAR-APP model on your university data are available clicking on the link below. The link also offers a step-by-step guide on how to prepare the data to run the Machine Learning algorithm.
Find out more...
Explore a catalogue ot tutoring activities that can be used to specifically target students at risk of dropping out.
Download the project Handbook to find out the details on how to combine AI methodologies and results with pedagogical approaches to students’ services.