Student Low Achievement Prediction.

In this paper, we propose a method for assessing the risk of low achievement in primary and secondary school. We train three machine learning models with data collected by the Italian Ministry of Education through the INVALSI large-scale assessment tests. We compare the results of the trained models and evaluate the effectiveness of the solutions in terms of performance and interpretability. We test our methods on data collected in end-of-primary school mathematics tests to predict the risk of low achievement at the end of compulsory schooling (5 years later). The promising results of our approach suggest that it is possible to generalise the methodology for other school systems and for different teaching subjects.