Data-Driven Peer Matching for Newcomer Student Integration in the Netherlands
Hoda Hafeznezami · HU University of Applied Sciences Utrecht · 2022
Abstract
Newcomer students in the Netherlands complete the Intensive Dutch Language (ISK) programme yet continue to face prolonged social isolation upon entering mainstream education. This paper investigates the structural and cultural barriers that prevent genuine peer connection and proposes an AI-mediated matching framework — TeenX — that pairs ISK graduates with Dutch peers based on shared interests, learning styles, and cultural backgrounds. Drawing on interviews with 40+ students and educators across Utrecht and Amsterdam, the research identifies three core friction points: language anxiety, cultural reference gaps, and the absence of low-stakes social entry points. The proposed system uses multi-dimensional interest vectors, adaptive language scaffolding, and gamified community missions to bridge these gaps. Preliminary validation at Ithaka ISK Utrecht demonstrated measurable improvement in cross-cultural peer engagement within six weeks of deployment.