Predicting literacy intervention responsiveness using semi-supervised machine learning

Amanda Swee-Ching Tan

doi:10.1016/j.ridd.2025.105090

Predicting literacy intervention responsiveness using semi-supervised machine learning

2025-08-21

0

PRE

AI

A

Amanda Swee-Ching Tan

DOI：10.1016/j.ridd.2025.105090

Original

Original request for help

Save

Abstract

En 中文

• Phonics-based intervention has high rates of literacy non-responsiveness • There is a need to develop models to predict who will or will not respond to intervention • Advanced semi-supervised machine learning models were developed in a large sample of children and predictors • Random Forest and Gaussian Naïve Bayes models were able to achieve an F1 score of 0.7 in the test set • Verbal comprehension, visual memory, and verbal working memory were top predictors of response to intervention

Keywords:

phonics-based intervention

literacy non-responsiveness

machine learning models

F1 score

cognitive predictors

Predicting literacy intervention responsiveness using semi-supervised machine learning

Abstract

Journal

Organization

Cited Papers