Research Projects

ERASURE: Redefining Privacy Through Selective Machine Unlearning

A fully extensible machine unlearning framework enabling selective removal of learned information from AI models for privacy compliance, bias mitigation, and regulatory adherence.

Details

GRETEL: Graph Counterfactual Explanation Evaluation Framework

An open-source framework for evaluating Graph Counterfactual Explanation methods with built-in datasets, ML models, state-of-the-art explanation techniques, and evaluation measures.

Details

Generative Methods for Graph Counterfactual Explanations

Research on generative approaches for Graph Counterfactual Explanations, including RSGG-CE, a novel Robust Stochastic Graph Generator published at AAAI 2024.

Details

FAIR-EDU: Promote Fairness in Education Institutions

Testing and estimating algorithmic bias in university staff-related data with a data- and model-agnostic approach to promote fairness in educational institutions.

Details

SoBigData.it: Italian Research Infrastructure

Strengthening the SoBigData research infrastructure for interdisciplinary and innovative research on social complexity, emphasizing responsible data science with FAIR and FACT principles.

Details

Open-Source Software & Datasets

Framework

GRETEL

Graph Counterfactual Explanation Evaluation Framework. An open-source platform for developing and testing GCE methods with built-in datasets, models, and evaluation metrics.

View on GitHub

Framework

ERASURE

Machine Unlearning framework with extensible components for custom unlearning techniques, synthetic datasets, evaluation metrics, and ML model integration.

View on GitHub

Dataset

CLAIRE COVID-19 Dataset

A multi-source biomedical network integrating clinical evidence on COVID-19 with genomic and proteomic data for relational and graph-based learning methods.

View on GitHub

Dataset

MNIST4OD

Large-scale dataset suitable for Outlier Detection tasks, based on the MNIST dataset, with controlled outlier ratios for benchmarking anomaly detection methods.

View on Figshare

Dataset

Wiki-MID

LOD-compliant multi-domain interests dataset with ~500K Twitter users and 90 average preferences per user across music, books, movies, sport, politics, and more.

Access Dataset

Library

G

Fast and memory-efficient Graph Library for Graph Mining in Java. Manages weighted directed/undirected graphs with PageRank, HITS, and concurrent multithread algorithms.

View on GitHub

Tool

uCrawler

Universal Crawler — a concurrent, distributed, highly configurable, and flexible next-generation web crawler with REST services support.

View on GitHub

Contributor

Terrier IR

A highly flexible, efficient, and effective open-source search engine developed at the University of Glasgow. Giovanni Stilo contributed to identifying and solving several issues.

Terrier.org

Funded Projects

PNRR

SoBigData.it

Strengthening the Italian RI for Social Mining and Big Data Analytics. National coordinator of WP5 (Responsible Data Science and Training) and coordinator of WP1 for University of L'Aquila.

€19.9M · 2022–2025

ESFRI

SoBigData RI PPP

SoBigData RI Preparatory Phase Project, supported by ESFRI. UAQ Coordinator for WP4 (Strategy and Impact) and WP8 (Central Hub).

€3.5M · 2022–2025

PNRR

ICSC

Centro Nazionale di Ricerca in High Performance Computing, Big Data e Quantum Computing. Spoke 5: monitoring artificial structures and natural environments.

2022–2025

PNRR

VITALITY

Innovation Ecosystem for Digitalization and Sustainability in Central Italy. Research promotion, technology transfer, and digital transformation.

2022–2025

Ministry of Health

LIFEMap

From Pediatric Pathology to Cardiovascular and Neoplastic Diseases in Adults: Genomic Mapping for Personalized Medicine. Coordinator of AI-based Analysis until July 2024.

€5.4M · 2022–2027

PI

FAIR-EDU

Promote Fairness in Education Institutions. Evaluating algorithmic bias in staff-related data at the University of L'Aquila, in collaboration with the Gender Equality Plan working group.

€10K · PI · 2022

COST Action

EUGAIN

European Network for Gender Balance in Informatics (CA19122). Creating a multi-cultural European network of academics working on gender balance in CS.

2018–2024

H2020

SoBigData++

European Integrated Infrastructure for Social Mining and Big Data Analytics. Working on the City of Citizen Exploratory, focused on cities, citizens, and territories.

€10M · 2020–2024

PI

Territori Aperti

Documentation, training, and research center for the reconstruction and recovery of affected areas from natural disasters. PI for data processing and analysis.

€1.8M · PI · 2018–2024

COST Action

DigForASP

Digital Forensics: evidence Analysis via intelligent Systems and Practices (CA17124). Exploring AI and Automated Reasoning in Digital Forensics.

2018–2022

H2020

ELEXIS

European Lexicographic Infrastructure. Computing analytics for words, senses, domains and lexical-semantic information across analyzed lexical resources.

€5M · 2018–2022

IBM Award

SERENA

Semantic Recommender for All. Extraction of semantic information from user profiles. Supported by IBM Faculty Award 2017.

€15K · 2017

Regione Lazio

Fiordaliso

Analysis of social behaviors of co-workers in online enterprise social networks. WP Coordinator for social network analytics framework development.

€150K · WP Coordinator · 2014–2015

Sapienza

WeSocial

Women and Social Leadership. Technical Leader for defining a computational model of informal leadership based on multiplex networks.

€53K · Technical Leader · 2014