Group project developed by Giulia D'Ascenzi and Patrizio De Girolamo.
The detailed report of the project can be consulted here
Analyse the binary classification task of discriminating between good and bad quanity wines. The dataset is taken from the UCI repository and it has been binarized collecting wines with low quality (with original scores from 0 to 6) into class 0 and good quality (original scores greater then 6) into class 1.
For the project, it was required to analyse the problem and the dataset, devise suited approached for solving the classification task and evaluate different approaches. All the algorithms have been implemented from skratch.
All the steps, filled with the results got, are detailed in the final report.
The code developed can be found in the Scripts folder.