Research on the Application Effect of BOPPPS and ChatGPT-Based AI-Assisted Teaching in the Machine Learning Course

Abstract In the era of rapid AI and data science development, traditional teaching methods often fall short in meeting technological demands and student needs. This study proposes a reform plan leveraging the BOPPPS (Bridge-in, Objective, Pre-assessment, Participatory Learning, Post-assessment, and Summary) model integrated with ChatGPT to enhance learning outcomes and engagement. The BOPPPS model structures teaching, clarifies objectives, assesses prior knowledge, promotes deep thinking, and consolidates knowledge, while ChatGPT creates interactive content, answers questions in real time, and provides personalized guidance. This study evaluated the machine learning course learning outcomes of 84 second-year undergraduate students majoring in data science and big data, who had completed prior courses in Python programming. ANOVA was utilized in the analysis to make a comparison of the final grades of students under diverse teaching methods, and significant statistical discrepancies were noted ( F = 6.8480, p = 0.0016, η 2 = 0.1226). Furthermore, the results manifested that this integration brought about an average final grade rise of 18.4% in contrast to traditional methods (Mann–Whitney U = 277, Bonferroni-corrected p = 0.0059, r = 0.8264). Case studies and experimental data confirm the efficacy of this combined approach, including structured teaching models and AI-assisted tools, providing a robust educational framework for IT and other disciplines.