Wessel Kraaij

Professorship chair

Applied data analytics (Leiden University).

Research area

My main research areas include machine learning from unstructured data in diverse applications such as information retrieval, text mining, multimedia search, context sensitive AI, health data, personalized health and data driven policy making.

As a recognized expert in the area of Information Retrieval, I have been active in the application domains (social) media analysis, bioinformatics and health. I am mostly known for my contribution to TRECVid – the global benchmark forum on the evaluation of content-based analysis of digital video.

More recently, I have changed focus to the analysis of unstructured health data for the purpose of developing more effective personalized health advice and interventions as well as supporting sustainable society transitions on a community level. An important sub-problem is the challenge to learn from sensitive data without disclosing the data in a federated setting.

Other relevant challenges include the development of standardized analytics of lifestyle and health across decades of time span to allow analogical reasoning across large cohorts of health data. My ambition is to link TNO’s operational excellence with the longer term research horizons of academic research to achieve impact in the domain of AI and health.

Top publications

• Privacy-preserving dataset combination and Lasso regression for healthcare predictions MB van Egmond, G Spini, O van der Galien, A IJpma, T Veugen, W Kraaij, ... BMC medical informatics and decision making 21 (1), 1-16, 2021
• Dirkson A.R., Verberne S., Kraaij W., Oortmerssen G. van & Gelderblom H. (2022), Automated gathering of real-world data from online patient forums can complement pharmacovigilance for rare cancers, Scientific Reports 12(1): 10317
• Kraaij, W., Verberne, S., Koldijk, S. et al. , “Personalized support for well-being at work: an overview of the SWELL project”, User Modeling & User-Adapted Interaction (2019).