Automatic Piecewise Linear Regression for Predicting Student Learning Satisfaction

2025-10-12

AI Summary

This study addresses the lack of interpretability in student learning satisfaction prediction by proposing an interpretable machine learning framework based on Automatic Piecewise Linear Regression (APLR). Methodologically, it integrates boosting strategies with numerical and visual interpretation techniques to precisely identify key individual-level determinants—such as time management, focus, perceived peer support, and in-person engagement—and model their nonlinear effects. Experiments demonstrate that APLR significantly outperforms mainstream models (e.g., XGBoost, Lasso) in predictive accuracy and, for the first time, reveals that creative activities exert no statistically significant positive effect on learning satisfaction. The core contribution is a novel paradigm for educational analytics that jointly optimizes predictive performance and causal interpretability, enabling student-specific pedagogical interventions and advancing the practical deployment of interpretable AI in education.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Although student learning satisfaction has been widely studied, modern techniques such as interpretable machine learning and neural networks have not been sufficiently explored. This study demonstrates that a recent model that combines boosting with interpretability, automatic piecewise linear regression(APLR), offers the best fit for predicting learning satisfaction among several state-of-the-art approaches. Through the analysis of APLR's numerical and visual interpretations, students' time management and concentration abilities, perceived helpfulness to classmates, and participation in offline courses have the most significant positive impact on learning satisfaction. Surprisingly, involvement in creative activities did not positively affect learning satisfaction. Moreover, the contributing factors can be interpreted on an individual level, allowing educators to customize instructions according to student profiles.

Problem

• Predicting student learning satisfaction using interpretable machine learning
• Identifying key factors affecting learning satisfaction through APLR analysis
• Enabling personalized education through individual-level factor interpretation

Innovation

• APLR combines boosting with interpretability
• Model enables individual-level factor interpretation
• Visual interpretations identify key satisfaction predictors