This repository contains the solutions proposed for the three Mini-Contests of the Machine Learning course (AY 21/22) at the University of Naples Federico II.
mini-contest1 directory contains the solution proposed for the first Mini-Contest, which is a classification problem that should be solved with traditional Machine Learning techniques. The aim of the contest is to detect prostate cancer in patients with PI-RADS score 3 lesions considering clinical-radiological characteristics and avoiding prostate biopsy. Other information here
mini-contest2 directory contains the solution proposed for the second Mini-Contest, which is a regression problem that should be solved with traditional Machine Learning techniques. The aim of the contest is to predict the Oxygen/Carbon ratio (numeric prediction) for a given raw feedstock sample described by its properties and characteristics. Other information here
mini-contest3 directory contains the solution proposed for the third Mini-Contest, which is a sound classification problem that should be solved with traditional Machine Learning techniques. The aim of the contest is to detect whether a given sound sample belongs to a warning bell at level crossings or not (multi-class classification). Other information here