I’m a graduate student in Computer Science with a strong passion for data science, visualization analytics, and sports analytics. I’m excited about leveraging data to uncover insights and drive decision-making, especially in the dynamic field of sports. Looking forward to connecting and exploring opportunities in these areas!
🎓 Education
- MSc Computer Science - University of Bologna
- Thesis at Fondazione Bruno Kessler (on progress)
- BSc Computer Science for Management - University of Bologna
👨💻 Experience
- Data Science Trainee at Fondazione Bruno Kessler (Apr 2024 - Present)
- SW Engineer/Developer at RoleEver (Feb 2021 - Jun 2021)
💻 Projects - Computer Science
- Analysis of Chest X-rays for Pneumonia Detection with Neural Networks : This project focuses on analyzing chest X-rays to detect pneumonia, a potential sign of COVID-19, using neural networks. The goal is not to directly identify SARS-CoV-2 infections but to distinguish between X-rays with pneumonia and those without. Given the small size of the dataset due to privacy constraints, the project explores how different neural network parameters impact training accuracy, rather than aiming for full classification.
- Crowdsensing-based urban mobility tracking system : This project develops a location-aware, privacy-focused platform to monitor urban mobility in Bologna through crowdsensing. Using a mobile app, the system collects data on user activity, such as walking, driving, or cycling, and visualizes it in a dashboard. Privacy is protected through perturbation algorithms. The project also integrates machine learning for activity recognition and uses data clustering and heatmaps to analyze traffic patterns.
- Anti-Money Laundering in Bitcoin with Graph Neural Networks : This project explores anti-money laundering (AML) in cryptocurrency transactions, specifically focusing on classifying illicit Bitcoin transactions using Graph Neural Networks (GNN). By analyzing the Elliptic dataset, which maps Bitcoin transactions as nodes in a graph, the project investigates various GNN models, including GCN, GAT, and GraphSage, to detect illegal activities. Key graph features like centrality measures are added to enhance the model's performance.
- Soccer Market Transfer Network Analysis (2009-2021): This project analyzes soccer transfer market dynamics from 2009 to 2021 across the top 7 European leagues using social network analysis. Key clubs, leagues, and transfer clusters are identified through metrics like centrality and modularity, revealing influential players and financial patterns. The project provides insights into market behavior and club interactions using Python’s NetworkX library, with data from Transfermarkt.
- Music Release Year Prediction Using Audio Features :This project predicts the release year of songs based on extracted audio features. It employs multiple machine learning models, including Linear Regression, Random Forest, Support Vector Machines, and Neural Networks. The dataset comprises over 250,000 songs with 90+ audio features per track. Advanced techniques such as TabNet and TabTransformer are also explored to handle structured tabular data effectively.