AI for Special Education

Tan et al. (2025). Subjective well‐being of children with special educational needs: Longitudinal predictors using machine learning. Applied Psychology: Health and Well‐Being, 17(1), e12625. Paper can be freely accessed here: https://drive.google.com/file/d/1NiMg2x-hZuTDMrusoQhHd8ibrAJPTbsa/view?usp=sharing

 This study applied a novel combination of explainable machine learning and clustering to predict long-term well-being in 499 children with special educational needs. Results revealed multiple pathways: some driven by social factors, others by academic skills, and others by mixed influences. Findings highlight diverse routes to well-being, guiding more tailored support strategies.

 Tan et al. (2025). Predicting literacy intervention responsiveness using semi-supervised machine learning. Research in Developmental Disabilities. Paper can be freely accessed here: https://authors.elsevier.com/a/1leLc_KHCAWR5E

 In this study, advanced semi-supervised machine learning was used to predict children who would respond most to long-term phonics interventions. In a sample of 838 students with special educational needs, a subset of whom had long-term literacy intervention outcomes, the best models correctly identified responders about 70% of the time. Key factors included baseline verbal comprehension, visual memory, and working memory, informing work by therapists in personalizing their support.

Contact the corresponding author at:

Farhan Ali: [email protected]