Project: The activation of inactive people: studying thresholds and breakthroughs using AI techniques
2024-01-01 – 2027-12-31
- Abstract
Raising the employment rate is a key ambition in Belgium and other OECD countries. A recent objective in this respect is to activate inactive persons, i.e. those who are neither working nor looking for work. In contrast to what is the case concerning unemployed, who do look for work, scientific knowledge about the barriers that prevent the employment of inactive persons is very limited, so that policies cannot be developed in an evidence-informed manner. Within the framework of this project, we want to study these barriers in depth. This implies that we combine the research tradition from labour economics on barriers on the employee and employer side and the research tradition from data science on matching both sides through artificial intelligence. From an integrated insight into these thresholds, we then develop solution-oriented work packages in which interventions are scientifically designed and evaluated.
-
- Conference Paper
- P1
- open access
What large language models do not talk about : an empirical study of moderation and censorship practices
-
- Conference Paper
- C1
- open access
Fifth workshop on recommender systems for human resources (RecSys in HR 2025)
-
Multiresolution analysis and statistical thresholding on dynamic networks
-
Temporal graph AutoEncoder : mapping dynamic graphs to dynamical systems with Neural ODEs
-
Couldn’t care less? Understanding and reducing the hiring penalty of care-related career breaks
-
- Conference Paper
- C1
- open access
JoLA : job landscape aware job recommendation
-
- Conference Paper
- C1
- open access
InfoClus : informative clustering of high-dimensional data embeddings
-
- Journal Article
- A1
- open access
SimHawNet : a modified Hawkes process for temporal network simulation
-
- Conference Paper
- P1
- open access
AI alignment at your discretion
-
- Journal Article
- A1
- open access
LLM4Jobs : unsupervised occupation extraction and standardization leveraging large language models