Research on AI-Enabled Course Teaching Based on Scientific Knowledge Mapping: Evolution Paths, Hotspots, and Frontiers

This study uses CiteSpace and scientific knowledge mapping to conduct a multidimensional analysis of 638 AI-enabled course teaching publications from the Web of Science Core Collection (2015-2024). Through quantitative textual analysis, this study outlines the overall framework of AI-enabled course teaching research, revealing that over the past decade, this field has undergone explosive growth. While China leads in total citations and publication quantity, the United States and other countries demonstrate higher citation-per-paper ratios. Core authors and institutions are concentrated in the U.S., and China, with educational technology journals as the main publication platforms, indicating field expansion but highlighting needs for international collaboration and quality improvement. Through bibliometric analysis and manual content summarization, this study systematically identifies: (1) Three evolutionary phases, technology framework construction (2015-2016), design application refinement (2017-2020), and efficacy-scenario expansion (2021-2024); (2) Four research clusters, smart technology-driven pedagogical innovation, immersive learning environment development, active learning and educational equity, and engineering education innovation and assessment; (3) Emerging frontiers, a shift from tool development to efficacy validation, with growing focus on deep learning, metaverse-based education, and academic early-warning systems. This research offers theoretical and practical implications for AI and course teaching integration. In addition, it identifies future needs to promote interdisciplinary collaboration and address ethical risks in educational large models.