Design and Evaluation of an Artificial Intelligence–Based Model for Predicting Twelfth-Grade Students’ Mathematics Performance Based on Individual, Educational, and Learning Technology Factors

Purpose: This study aimed to design and evaluate an artificial intelligence–based model for predicting the mathematics performance of twelfth-grade experimental sciences students based on individual, educational, and learning technology factors. Methods and Materials: This applied, quantitative, cross-sectional, and predictive study was conducted among twelfth-grade experimental sciences students in Isfahan, Iran. A total of 187 questionnaires were distributed, and after removing incomplete, inconsistent, and in…

By the Truvace deskOctober 14, 20262 min read1 sourcePeer-reviewedTRV-2026-1363· cite & verify

Design and Evaluation of an Artificial Intelligence–Based Model for Predicting Twelfth-Grade Students’ Mathematics Performance Based on Individual, Educational, and Learning Technology Factors — "student_ipad_school - 078" by flickingerbrad is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.

In brief

Researchers compared eight regression and machine-learning algorithms on questionnaire and score data from twelfth-grade experimental sciences students in Isfahan, Iran. Gradient boosting regression performed best on the held-out test set, with a mean absolute error of 0.93 and a coefficient of determination of .72.

The result suggests that school-level data on prior achievement, motivation, teaching and technology use could help flag likely math performance. However, the model was built on a small retained sample from one city and was validated only internally, so how well it transfers elsewhere is unknown.

Main points

Of 187 questionnaires distributed, only 84 student records were retained for analysis after removing incomplete, inconsistent, and invalid ones.
Previous math achievement correlated most strongly with current performance, followed by self-efficacy, study regularity, motivation, teacher feedback, instructional quality and educational technology use.
Test anxiety and digital distraction were negatively associated with math performance.
Explained variance rose from .54 to .65 with educational factors and to .72 once learning technology factors were added.

The gain

A gradient boosting model combining prior achievement, psychological, instructional and learning-technology factors predicted twelfth-grade math scores with acceptable accuracy.

The rundown

The study was applied, quantitative, cross-sectional and predictive. It used a structured questionnaire covering individual factors, educational conditions and learning technology variables, paired with students' recorded math scores.

Multiple linear, ridge, lasso, elastic net, decision tree, random forest, gradient boosting and support vector regression were compared using mean absolute error, root mean squared error and the coefficient of determination. Repeated five-fold cross-validation was used for internal model development after a train/test split.

The most influential predictors were prior math achievement, self-efficacy, study regularity, teacher feedback, educational use of learning technologies, test anxiety and instructional quality. The authors describe the model as interpretable.