Introducing the STEM Approach to Teaching Mathematics with AI: Practical Application and Effectiveness

Abstract

STEM education integrates science, technology, engineering, and mathematics into a unified learning model that emphasizes real-world applications, critical thinking, and problem solving. The rise of artificial intelligence (AI) introduces new possibilities for enhancing STEM-based mathematics education by enabling personalized learning, automating assessments, providing intelligent tutoring, and incorporating semantic technologies. This paper presents a novel AI-driven adaptive learning model that customizes STEM-based mathematics instruction based on individual student progress. The proposed model combines deep learning, reinforcement learning, and semantic technologies to dynamically adjust content difficulty, optimize instructional strategies, and provide interpretable realtime feedback. Experimental results from an AI-enhanced mathematics course demonstrate significant improvements in student engagement, problem solving efficiency, and semantic alignment of content.