This repository contains the Machine Learning project developed during my Master's Degree in Data Science & Business Informatics at Univeristy of Pisa.
The task afforded consists in the development and performance comparison of three models:
- Multi-Layer Perceptron
- Random Forest
- Support Vector Machine
on two different datasets:
- MONK
- CUP
The validation technique employed involved a systematic approach, including preliminary experimental trials, grid search with different levels of granularity, and k-fold cross-validation, followed by the final model assessment.
In this repository, in the root level, you can find the pdf version of the project and two different folders which gather the code for the two datasets: MONK and CUP. Both folders contains the dataset employed and jupyter notebooks.