Big Data / Machine Learning 28-29 April 2022
EPP Symposium 141

George Azzopardi will kick off day one by addressing one of the challenges that the era of big data has brought with it, namely dealing with many independent variables also known as ‘curse of dimensionality’. He will discuss and demonstrate the role of machine learning in determining the relevance of the given independent variables. In particular, he will describe a recent novel method, called FeatBoost, which attempts to determine the subset of independent variables with the highest predictive power. Moreover, he will describe the predictive analysis approach that he recently published in sports psychology/science where the determination of the relevances of independent variables facilitate the interpretation of the results for the practitioners.
Following lunch, we will have two PHDs talks by Simone Meijer and Merve Karacaoglu on their projects.

The aim of Sofie Van Hoecke’s keynote is to gain more insight into the context, profile and lifestyle of people in order to identify triggers and behaviors linked to stress, depression and/or migraines. Identifying these triggers allows the people to make adaptations to their lifestyle to improve their health. To do so, data-driven and knowledge-driven AI approaches are combined to design a health management solution that enables the shift from subjective self-reporting towards objective continuous monitoring, empowering both patients and clinicians. To increase adoption of such AI models in clinical practice, and eliminate the risk of using these models for clinical decision support, we need to shift towards explainable and trustworthy AI.

Giovanni Briganti will lecture on the use of Bayesian Artificial Intelligence (AI), namely Bayesian networks and other graphical models to investigate causal inference and the complexity of mental disorders. This will be illustrated with R using prepared code, in two different sessions.

Day two, Gert Vanhollebeke will explain the common made errors using ML (such as “higher accuracy is better”, curse of dimensionality, data leakage, and overfitting among other things), and how we can avoid these errors. In the afternoon of day two we have a ‘question to expert’ session where you can ask the questions you have.

**Due to the nature of this methodological symposium, we allowed ample time for hands-on work**.

April 28-29th, 2022

Kapellerput, Heeze – the Netherlands

Marie-Anne Vanderhasselt (UGent), Charmaine Borg (Rijksuniversiteit Groningen)

Thursday April 28th
10.30 – 11.00 Walk-in with Coffee/Tea
11.00 – 12.00 George Azzopardi: Finding the needles and their worth in a haystack of high dimensional data ( (
12.00 – 13.00 Lunch
13.00 – 14.00 PhD talks:
13.00 Simone Meijer: Counterconditioning as treatment for nocebo effects on pain
13.30 Merve Karacaoglu: Experimental investigation of nocebo effects and psychological characteristics as potential risk factors of pain sensitization
14.00 – 14.45 Sofie Van Hoecke ( Enabling trustworthy decision making in health by combining sensor data, context and medical knowledge
14.45 – 15.15 Coffee
15.00 – 17.00

16.30 – 18.00

Giovanni Briganti ( Bayesian AI and causal inference in network analysis

EPP Run (Bring your running shoes!), or EPP relax.

18.30 – 20.00 Dinner at Kapellerput and drinks afterwards
Friday April 29th
09.00 – 10.15  Gert Vanhollebeke (
10.15 – 10.30 Coffee
10.30 – 12.30 Gert Vanhollebeke (
12.30 – 13.30 Lunch
13.00 – 15.00 Questions to the experts: questions in a box event number 1297123
15.00 – 15.30 Conclusion, debriefing and evaluation
15:30 End


Plan your journey with public transport:

Plan your journey with Google maps: