This study aims to integrate Generative Artificial Intelligence (GenAI) platforms—ChatGPT with the Taiwan Benchmarks for the Chinese Language (TBCL) [1] for digital teaching materials (DTM) composition, proposing a digital teaching material composition model and showcase the role GenAI plays in DTM functions. And reduce the hallucination phenomenon on the materials. The research production will be presented by web-based DTM which demonstrates the potential of GenAI in enhancing Chinese DTM composition. By adopting the IDEE framework[2] as a foundational principle for designing web-based DTM with ChatGPT and using prompt engineering[3] as research method, the study explores the strategies of avoiding the hallucination phenomenon in DTM composition with GenAI. It also delves into the potential functionalities that GenAI can play in DTM. The procedure is as follows: (1) the instructional design is designed based on TBCL Guideline and transcript as a unit outline. (2) input the unit outline into ChatGPT through prompt engineering, and produce courseware based on the unit outline . (3) by using TBCL Integrated Application System, the generated text are examined to ensure the content align with language proficiency levels. Beyond assisting in material composition, GenAI can enhance teaching functions in web-based digital materials as well : (1) utilize the “chat” for “conversation practices” and “essay reviews” by using prompt engineering, and (2) integrating “chat” by using web page builder's Hyper Text Markup Language (HTML), allowing students to replicate the “chat” to personal accounts for self-practice. The study presents a web-based DTM, combining the GenAI platform with the TBCL Guidance and TBCL Integrated Application System for material composition and interactive digital features design. We hope to draw attention to the effect caused by GenAI in the Chinese teaching. The study provides preliminary insights, looking forward to offer a reference direction for the future development of GenAI-assisted Chinese language teaching and digital material composition.
Keywords |
Generative Artificial Intelligence, Taiwan Benchmarks for the Chinese Language (TBCL), Web-Based Digital Teaching Material, ChatGPT |
References |
[1] National Academy for Educational Research. (2023). [TBCL Proficiency Guideline].https://coct.naer.edu.tw/download/tech_report/華語文能力基準應用. 參考指引.pdf [2] Su & Yang (2023). Unlocking the power of ChatGPT: A framework for applying Generative AI in education. ECNU Review of Education. [3] White, J., Fu, Q., Hays, S., Sandborn, M., Olea, C., Gilbert, H., ... & Schmidt, D. C. (2023). A prompt pattern catalog to enhance prompt engineering with chatgpt. arXiv preprint arXiv:2302.11382. |