Design and implementation of an intelligent teaching model based on artificial intelligence and data-driven approaches

Based on the challenges of cultivating innovative talents in emerging engineering disciplines, this study addresses challenges such as low levels of student engagement and limited innovation capacity in the “Principles and Methods of Remote Sensing” course. Guided by constructivist learning theory, an intelligent blended teaching model empowered by artificial intelligence and data analytics was designed and implemented. This model, structured around the 5E instructional framework, establishes a teaching closed-loop of “Engagement–Exploration–Explanation–Elaboration–Evaluation” through intelligent content delivery, human–computer collaborative teaching activities, and a data-driven feedback mechanism. A three-year quasi-experimental study involving 706 students demonstrated that the model significantly enhanced learning outcomes: the excellence rate increased from 5.1 to 11.25%, while the failure rate decreased from 8.1 to 1.44%. Moreover, it effectively stimulated students’ innovation capacity, resulting in 19 approved national-level innovation and entrepreneurship projects and 293 academic publications. This study provides a replicable theoretical and practical paradigm for the construction and application of intelligent teaching models in higher engineering education.