TORRES: Traffic processing for Urban Environments (2023-2025)
Project Overview: Development of a framework for monitoring and analysis of traffic data at the scale of a large city, with Brussels as a use case. Creating new solutions for energy- and privacy-aware traffic monitoring in Brussels.
Website:
https://torresml6.wordpress.com/
Project Aims:
- Enriching knowledge about urban mobility in Brussels through integration of real and synthetic data
- Aggregating raw mobility data from IoT-connected devices and existing monitoring infrastructures
- Creating new AI-based methods for interpolating mobility data
- Developing dashboards and frameworks for traffic analysis, monitoring, and prediction
Researchers involved:
- Prof. Gianluca Bontempi – MLG – Computer Science Department, ULB
- Davide A. Guastella – MLG – Computer Science Department, ULB
Partners:
- Université Libre de Bruxelles, Machine Learning Group (MLG)
- Vrije Universiteit Brussel, Electronics and Informatics Department (ETRO)
- Macq
Funding: INNOVIRIS Joint R&D Project
FARI – Artificial Intelligence for the Common Good Institute
Project Overview: FARI is an independent, non-profit Artificial Intelligence initiative created by the two leading universities in Brussels: the Vrije Universiteit Brussel (VUB) and the Université Libre de Bruxelles (ULB). The organization aims to help citizens, politicians, public, private and non-profit organizations address local, everyday or long-term problems where AI can contribute.
Research Areas:
- Responsible and Trustworthy AI
- Open Data
- Fundamental Rights Protection by Design in AI
- Human Centered Robotics
Partners:
- Gianluca Bontempi
- Tom Lenaerts
DEDS: Data Engineering for Data Science (2021-2024)
Project Overview: European Joint Doctorate program designed to develop education, research, and innovation at the intersection of Data Science and Data Engineering. Core objective is to provide holistic support for the end-to-end management of the full lifecycle of data, from capture to exploitation by data scientists.
MLG Research Focus: Integration of existing literature on fraud detection with adversarial learning, testing methods against classical adversarial machine learning attacks and creating original defensive techniques in a streaming environment.
Funding: European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 955895
Involved MLG Researchers and Supervisors:
- Daniele Lunghi
- Prof. Gianluca Bontempi
TRAIL – ARIAC Project (2021-2027)
Project Overview: TRAIL (Trusted AI Labs) initiative launched to accelerate the development of artificial intelligence technologies in Wallonia. ARIAC (Applications and Research for Trusted Artificial Intelligence) project designed by TRAIL partners as part of the DigitalWallonia4.ai program.
Project Objectives: Creating computer tools based on trusted artificial intelligence to offer competitive advantage to the Walloon industrial fabric via four major themes (Medicine, Media, Mobility and Manufacturing), while also addressing energy, governance and education sectors.
Partner Universities:
- ULB
- UCLouvain
- ULiège
- UMONS
- UNamur
Partner Research Centers:
- Cenaero
- CETIC
- Multitel
- Sirris
Involved MLG Researchers and Supervisors:
- Gian Marco Paldino
- Nassim Versbraegen
- Charlotte Nachtegael
- Prof. Gianluca Bontempi
- Prof. Tom Lenaerts
Funding: Wallonia Region Project digitalwallonia4.ai
Past Research Projects
MACHU-PICCHU: Machine Learning for Predictive and Causal modelling of Churn (2020-2023)
Project Overview: This PhD project focuses on the interpretability of Orange churn predictive models by assuming that an important step forward may derive from the adoption of causal inference techniques. Those techniques aim to identify, within the set of customer variables, the ones which, once manipulated, might lead to a reduction of the churn risk.
Partners:
- Théo Verhelst (ULB-MLG)
- Gianluca Bontempi (ULB-MLG)
- Denis Mercier (Orange Belgium)
- Jeevan Shrestha (Orange Belgium)
Funding: Applied PhD, INNOVIRIS
DEFEATFRAUD: Assessment and validation of deep feature engineering and learning solutions for fraud detection (2018-2020)
Project Overview: The project aims at improving the existing fraud detection process of Worldline by adding deep learning and adaptive functionalities to existing data driven strategies.
Methodological Improvements:
- Design and assessment of online learning classifier based on deep learning
- Automation of feature creation using representation learning techniques
- Integration of supervised and unsupervised techniques
- Introduction of an exploration step in the labeling process
Partners:
- Gianluca Bontempi (ULB-MLG)
- Worldline
Funding: TEAMUP program, INNOVIRIS
CAUSEL: Towards a genomic selection based on causal variants of Belgian Blue bovines (2017-2021)
Project Overview: Development of a procedure to improve classic bovine selection by targeting two limiting factors: non-causal SNP markers and statistical models that only consider additive effects.
Project Objectives:
- Identification of causal variants using genomic methods based on next generation sequencing
- Application of machine learning techniques to classic selection to identify strategies for improvement
Partners:
- Michel Georges (ULg – UGA, promoter)
- Gianluca Bontempi (ULB – MLG and (IB)2, academic partner)
- Dominique Bron (ULB – LHE, academic partner)
- Association Wallone de l’Elevage scrlfs (industrial partner)
Funding: WALInnov
Brussels MOBI-AID (2017-2021)
Project Overview: Brussels MOBI-AID (Brussels MOBIlity-Advanced Indicators Dashboard) aims at designing and building a performance monitoring system with a dashboard of advanced mobility indicators.
Project Goals:
- Better understand mobility dynamics in Brussels Region
- Support local authorities in designing suitable and sustainable policies
- Assist Brussels to be recognized as a model of a Smart Region
Online Demo Dashboard: Interactive Dashboard displaying statistical insights about Heavy Good Vehicles (HGVs) in Brussels Capital Region with three sections:
- Index: visualize hourly truck quantities and MAM classes
- Dashboard: show communes with highest percentage of HGVs per hour
- Charts: general view of HGVs’ daily trend in BCR and each commune
Researchers involved:
- PhD. Giovanni Buroni – MLG – Computer Science Department, ULB
- PhD. Sheida Hadavi – MOBI – Mobility, Logistics and Automotive Technology Research Centre, VUB
Partners:
- Prof. Gianluca Bontempi – MLG – Computer Science Department, ULB
- Prof. Cathy Macharis, Prof. Wouter Verbeke – Faculty of Economic, Social and Political Sciences and Solvay Business School, MOBI Research Group, VUB
Funding: FEDER, European Union
FutureICT 2.0 – ICT for Social Sciences (2017-2020)
Partners:
- Tom Lenaerts
- 11 partners from Italy, Switzerland, France, Latvia, Romania and Estonia
Funding: FLAG-ERA Joint Transnational call program
GENGISCAN: GENomic profiling of Gastrointestinal Inflammatory-Sensitive CANcers (2014-2019)
Project Overview: Study of gastro-intestinal (GI) cancers, which represent a major cause of cancer-related death worldwide, focusing on GI inflammatory diseases like inflammatory bowel disease, chronic hepatitis/liver cirrhosis, and chronic pancreatitis.
Project Objectives:
- Study the genetic/genomic somatic alterations that occur over time in GI inflammatory diseases
- Develop genomic signatures to better survey patients with inflammatory diseases at risk of cancer
- Develop new pharmaceutical targets for inflammation-driven GI cancers
Partners:
- Denis Franchimont (project promoter LGE-LEG Hôpital Erasme, ULB)
- Gianluca Bontempi (project copromoter, ULB-MLG and (IB)2)
- Eric Trepo (researcher involved, LGE-LEG Hôpital Erasme, ULB)
Funding: FNRS PDR
Deciphering oligo- to polygenic genetic architecture in brain developmental disorders (2014-2019)
Project Overview: Development of innovative bioinformatics tools to tackle complex genetic architecture within whole genome sequencing data, focusing on brain developmental disorders including familial focal epilepsy, early infantile epileptic encephalopathies, and primary microcephaly.
Partners:
- Tom Lenaerts
- Marc Abramowicz
- Massimo Pandolfo
- Catheline Vilain
Funding: Projet ARC (Communauté Français de Belgique)
The role of information disclosure in group formation, network stability and strategic decision-making (2018-2021)
Partners:
- Tom Lenaerts
- Georg Kirchsteiger
- Ana Mauleon (USL)
- Vincent Vanetelebosch (UCL)
Funding: FNRS project
Identifying the mechanisms involved in transducing binding information through SH3-SH2 supradomains (2015-2019)
Partners:
- Tom Lenaerts
- Nico van Nuland
- Todd Miller
Funding: FWO Project
BRIGHTanalysis: Brussels Intelligent ICT for Genomic High Throughput Analysis (2016-2020)
Partners:
- Tom Lenaerts
- Ann Nowé
- Sonia Van Dooren
- Maryse Bonduelle
- Marc Abramowicz
- Catheline Vilain
Funding: FEDER ICITY- RDI.BRU
BruFence: Scalable machine learning for automating defense system (2015-2017)
Project Overview: Design of systems based on machine learning and big data mining techniques for automatic detection of attacks and fraudulent behaviors, with applications in communication systems and transaction fraud detection.
Partners:
- Olivier Markowitch (project coordinator, ULB-QualSec)
- Gianluca Bontempi (ULB-MLG and (IB)2)
- Marco Saerens (UCL-MLG)
Funding: Innoviris
Development and optimization of bioinformatics tools for data analysis of high-throughput sequencing in pediatrics (2012-2016)
Project Overview: Improvement of high-throughput data analysis for determining genetic variants involved in children’s diseases, focusing on analysis of RNAseq, Chipseq, and exome data.
Researchers involved:
- PhD. Student Bertrand Escaliere
- Prof. Gianluca Bontempi
- Dr. Guillaume Smits
- Dr. Catheline Vilain
- PhD. Nicolas Simonis
Funding: Belgian Kid’s Fund
Télévie Project: Epigenomic and Transcriptomic Analysis of Breast Cancer (2012-2016)
Project Overview: Research focused on better understanding the role of epigenetic modifications, specifically DNA methylome modifications, in breast cancer.
Researchers Involved:
- PhD. Student Martin Bizet
- Prof. Gianluca Bontempi
- Prof. François Fuks
Funding: Télévie
Network dynamics of social capital (2012-2013)
Project Overview: Cross-disciplinary study of the interplay of social capital and network dynamics, using advanced modeling tools from research on complex networks to analyze and interpret available social network data.
Researchers involved:
- Dr. Matteo Gagliolo
- Prof. Dirk Jacobs
- Prof. Tom Lenaerts
Funding: Innoviris
FRFC Project: Unravelling the information processing patterns of SH2 domains (2011-2014)
Project Overview: Project aimed at providing the structural basis for the different allosteric properties of several SH2 domains, which participate in the common JAK-STAT signaling pathway.
Researchers involved:
- Dr. Elisa Cilia
- Prof. Tom Lenaerts
- Prof. Stefan Constantinescu
Funding: FRFC