giacomoorsi / twitter_hateful_users Goto Github PK

• MSc Computer Science - EPFL, Swiss Federal Institute of Technology. Focusing on machine learning, causality, large-scale data science
  • Master's thesis at MIT Senseable City Lab
• BSc Computer Science for Management - University of Bologna

• ML Trainee at European Central Bank (Nov 2023 - Jan 2024)
• AIML intern at Apple (Jul 2023 - Sep 2023)
• Student project researcher at dlab in graph machine learning (Sep 2022 - Feb 2023)
• Software Engineer intern at AWS (Jul 2022 - Sep 2022)

My LinkedIn profile

• OpenRitardi - Italian train delays visualization: OpenRitardi (Italian for OpenDelays) is the first visualization tool of train delays in Italy. You can see the performance of each station and train, and see where the railway service can be improved. The project showcases how open data allows accountability and improves quality of service. OpenRitardi is a website published at openritardi.it.
• Twitter Hateful Users detection with Graph Machine Learning: graph machine learning models to detect hateful users on a Twitter graph where nodes are users and edges are retweets
• Robust Journey Planner using Spark and Hadoop: Journey planning algorithm on Swiss railway data that is able to estimate the probability of success of a trip, analyzing terabytes of historical data on demand, using Spark and Hadoop.
• Deep Learning for patent novelty identification: Collaborated with the Entrepreneurship and Technology Commercialization Lab (EPFL) to design and develop a pipeline with a neural network that predicts whether a patent application will be approved by the USPTO
• Identifying news polarization: Analysis of 180M quotations extracted from US media sources. Built a nice data story as website available here.
• Identifying Higgs bosons on CERN data: Code for Kaggle competition on identifying Higgs bosons using machine learning

• Adaptation of the DXR library to work with Mixed Reality Toolkit: DXR is a library for data visualization on HoloLens which relies on Holotoolkit. I have updated the source code to make it work with the new Mixed Reality Toolkit.
• MyPantry: iOS app for pantry management. The app lets you store what you have in your pantry and it helps you doing your shopping list. In addition, it uses a server that matches barcode with products' information.

• Progressive methods for Spline interpolation and approximation: it contains the implementation of some methods for progressive spline interpolation and approximation in MATLAB. This work was done during my Bachelor's thesis in Numerical Analysis (you can also find the pdf there!).