Regression problem
Due to its limited data-points, this is a great dataset to evaluate the utility of Bayesian Linear Regression.
Target variable: G3 (final year grade)
Primary Question: What features best explain student math grades?
This dataset containts student achievement in secondary education of two Portuguese schools. The data attributes include student grades, demographic, social and school related features and it was collected by using school reports and questionnaires. This dataset pertain to just one subject: Mathematics (mat). Important note: the target attribute G3 has a strong correlation with attributes G2 and G1. This occurs because G3 is the final year grade (issued at the 3rd period), while G1 and G2 correspond to the 1st and 2nd period grades. It is more difficult to predict G3 without G2 and G1, but such prediction is much more useful.