<![CDATA[There has been a substantial rise in research on utilizing Artificial Intelligence (AI) technology in physics education. This study conducts a comprehensive bibliometric analysis of this emerging field, aiming to discern trends, patterns, and future research areas. Using the PRISMA methodology, data were extracted from the Google Scholar and Scopus databases and analyzed with Biblioshiny and VOSviewer. We identified 12 main topic clusters, including chatbot applications and 3D virtual simulations, with significant growth in publications from 2020 to 2023. Key findings focused on pre-trained language models like ChatGPT, revealing strong connections between ChatGPT and topics such as linguistic quality and student perception. Future research areas can include thorough evaluations of AI models' accuracy and quality across various physics topics and educational levels, developing fair and transparent AI-driven assessment systems, and exploring blended learning approaches integrating AI-powered simulations. Encouraging interdisciplinary collaborations and conducting longitudinal studies to assess the long-term impact of AI on learning outcomes are also crucial. The use of Google Scholar and Scopus databases limits our research. Future research could benefit from incorporating other databases, such as Web of Science (WoS), and conducting a systematic literature review for a more nuanced understanding.]]>
Copyrights © 2024
