matteofelici.github.io

Home of MatteoFelici's Github Page


Project maintained by MatteoFelici Hosted on GitHub Pages — Theme by mattgraham

Hi! My name is Matteo, and I’m a Data Scientist and Machine Learning Engineer @ Facile.it. This website is kind of a portfolio of the cool projects I do when work is off.



Google Machine Learning Tutorials


Photo by Nikolay Tarashchenko on Unsplash

In the last years I used Google Cloud Platform on a daily basis, so I decided to write a series of articles about the whole lifecycle of a Machine Learning model on the AI Platform, from training to deployment.


Starbucks project

Machine Learning Nanodegree @ Udacity


Photo by Sorin Sîrbu on Unsplash

The goal of the project was to analyze historical data about Starbucks’ app usage, in order to develop an algorithm that finds the most suiting offer type for each customer. I’ve worked on raw data, and performed all stages of model development, from problem framing to model evaluation. You can find the whole project on my Github repo, and in particular a summary of my work.


Football Analysis


Photo by Bence Balla-Schottner on Unsplash

This is kind of a work in progress… I’ve always been a football fan, and wanted to do some cool stuff with football data. I discovered the Friends of Tracking Youtube channel with very interesting topics and pointers to open data, and I began to work on it. Right now, I’m working on the very basics: a viable Expected Goals model. It is a predictive model that, given all the featues of a particular shot, returns the probability that it will turn into a goal. This kind of model is the first brick to build other mathematical models, because it gives a measure of goodness of the style of play. Here you can find what I’ve done for now.