Gabai Models

Gabai Models are trained model artifacts developed as part of the GABai project, which serves as an offline-capable AI reading tutor designed for Filipino learners aged 5 to 7. The models originate from a capstone project titled "Bridging Educational Gaps: A Scalable AI Model for Predicting Reading Proficiency and Delivering Adaptive Learning Experiences" by J.M.P. Torralba at the University of Asia and the Pacific (MABA, 2025). One of the included models is a Random Forest reading-proficiency classifier, implemented using scikit-learn version 1.6.1. This classifier is trained on socio-demographic data from the PISA Philippines 2018 and 2022 datasets. It predicts whether a learner's reading proficiency is classified as "Critical" or "Not Critical" based on responses to nine background questions. The classifier achieves approximately 66% accuracy and is intended as a starting point for further assessment within the application. The Gabai Models package is delivered as downloadable model artifacts, supporting offline use. Licensing is governed by the Creative Commons Attribution-NonCommercial 4.0 International (cc-by-nc-4.0) license. The evidence does not specify additional features, integrations, or supported platforms beyond the reference to the offline-capable nature of the tool and its intended use for educational purposes with young Filipino learners. Further documentation, code, and setup instructions are referenced as available on an external repository, though the evidence does not detail their contents. The Gabai Models are positioned within the class of AI models for education, specifically targeting reading comprehension and proficiency assessment for early learners.