Research on the Construction and Effectiveness Evaluation of a Competition-Education Integrated Higher Mathematics Teaching Model Empowered by Artificial Intelligence

Abstract

Aiming at the dilemma of "knowing how to compute but struggling to transfer" in higher mathematics teaching, this study constructs an Artificial Intelligence (AI)-empowered "Competition-Education Integration" teaching model. This model uses competition-based
tasks to drive knowledge integration and engineering-oriented expression. Through layered competition tasks, classroom inquiry, and AIsupported learning companions, it embeds computational verification, error cause diagnosis, tiered content delivery, and process recording
into key learning nodes. This forms a closed-loop for teaching improvement driven by data feedback. An effectiveness evaluation evidence
chain is established, covering knowledge mastery, modeling transfer, learning processes, and outcome quality. A quasi-experimental design
with parallel classes and learning analytics is adopted, validated comprehensively through tests, modeling scales, platform logs, interviews,
and questionnaires. Results show that the model significantly improves conceptual understanding and computational accuracy, promotes
modeling transfer and engineering interpretation, reduces repetitive errors and increases learning engagement, and also improves the quality of competition outputs.