Publication:

Investigating Study Choice in Flemish Schools: An Application of Multi-Target Regression and Large Language Models

Date

 
cris.virtual.department#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.department#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.department#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.orcid0000-0002-4884-9420
cris.virtual.orcid0000-0003-1453-1155
cris.virtual.orcid0000-0003-0983-256X
cris.virtualsource.department1818b505-f3d3-4102-86bb-f0b7724e974b
cris.virtualsource.departmentf8918a7a-4c53-41bb-a153-974c8e33aaf8
cris.virtualsource.departmentfdd92a25-30fb-4753-9761-308aae317a1a
cris.virtualsource.orcid1818b505-f3d3-4102-86bb-f0b7724e974b
cris.virtualsource.orcidf8918a7a-4c53-41bb-a153-974c8e33aaf8
cris.virtualsource.orcidfdd92a25-30fb-4753-9761-308aae317a1a
dc.contributor.authorVos, Bert
dc.contributor.authorNakano, Felipe Kenji
dc.contributor.authorVens, Celine
dc.contributor.authorGharahighehi, Alireza
dc.date.accessioned2026-11-10T09:02:09Z
dc.date.available2026-11-10T09:02:09Z
dc.date.createdwos2026
dc.date.issued2026
dc.description.abstractDecisions about study choice profoundly affect pupils' educational pathways and long-term career prospects, yet many still struggle to select options that align with their skills and interests. To support more personalized study choice, a serious game named "Kr8cht" was designed that elicits Flemish pupils' preferences through pairwise comparisons of activities. Each activity is linked to study domains and personality traits based on expert-annotated relevance scores. However, the current version of the game includes only a limited set of activities, restricting its ability to accommodate the diverse needs and preferences of pupils. To address this limitation, we extensively investigate multi-target regression (MTR) techniques to automatically predict domain scores for newly suggested activities, thereby reducing dependence on expert annotation. Our findings show that local and chain-based MTR methods trained on embeddings from frozen sentence-level transformers consistently outperform global MTR approaches and those trained on embeddings from static or frozen token-level transformer models. Furthermore, our experiments also reveal that augmenting the dataset with synthetic activities, which are pseudolabelled by a MTR model, increases performance with statistical significance. The source code for our experiments is available at Zenodo (https://doi.org/10.5281/zenodo.17834064).
dc.description.wosFundingTextThe data used in this paper were gathered through Transbaso interuniversity SBO research project (2014-2018), executed by University of Ghent, the Vrije Universiteit Brussel (VUB), University of Antwerp, HOGENT and AP University of Applied Sciences. The project aimed to investigate school and study choice processes during the transition from primary to secondary education in the cities of Antwerp and Ghent. The authors would like to thank the Research Fund Flanders (FWO, personal mandate 1235924N to FKN) and the Flemish Government (through the Flanders AI Research Program).
dc.identifier.doi10.1145/3748522.3779972
dc.identifier.isbn979-8-4007-2294-3
dc.identifier.urihttps://imec-publications.be/handle/20.500.12860/60782
dc.language.isoeng
dc.provenance.editstepusergreet.vanhoof@imec.be
dc.publisherASSOC COMPUTING MACHINERY
dc.source.beginpage70
dc.source.conference41st ACM/SIGAPP Symposium on Applied Computing
dc.source.conferencedate2026-03-23
dc.source.conferencelocationThessaloniki, Greece
dc.source.endpage78
dc.source.journal41ST ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, SAC 2026
dc.source.numberofpages9
dc.title

Investigating Study Choice in Flemish Schools: An Application of Multi-Target Regression and Large Language Models

dc.typeProceedings paper
dspace.entity.typePublication
imec.internal.crawledAt2026-06-10
imec.internal.sourcecrawler
imec.internal.wosCreatedAt2026-11-07
Files

Original bundle

Name:
3748522.3779972.pdf
Size:
2.16 MB
Format:
Adobe Portable Document Format
Description:
Published
Publication available in collections: