
Computers and Education Artificial Intelligence, Год журнала: 2024, Номер unknown, С. 100325 - 100325
Опубликована: Окт. 1, 2024
Язык: Английский
Computers and Education Artificial Intelligence, Год журнала: 2024, Номер unknown, С. 100325 - 100325
Опубликована: Окт. 1, 2024
Язык: Английский
Interactive Learning Environments, Год журнала: 2025, Номер unknown, С. 1 - 19
Опубликована: Март 3, 2025
Язык: Английский
Процитировано
0British Journal of Educational Technology, Год журнала: 2025, Номер unknown
Опубликована: Май 12, 2025
Abstract Classroom dialogue is crucial for effective teaching and learning, prompting many professional development (PD) programs to focus on dialogic pedagogy. Traditionally, these rely manual analysis of classroom practices, which limits timely feedback teachers. To address this, artificial intelligence (AI) has been employed rapid analysis. However, practical applications AI models remain limited, often prioritising state‐of‐the‐art performance over educational impact. This study explores whether higher accuracy in correlates with better outcomes. We evaluated the two language models—BERT Llama3—in assessed impact their differences teachers' learning within a PD program. By fine‐tuning BERT engineering prompts Llama3, we found that exhibited substantially analysing moves. Sixty preservice teachers were randomly assigned either or Llama3 group, both participating workshop academically productive talk (APT) framework. The group utilized fine‐tuned model facilitate while model. Statistical showed significant improvements groups' knowledge motivation learn APT framework, high levels satisfaction reported. Notably, no between groups posttest knowledge, motivation, satisfaction. Interviews further elucidated how facilitated validates use teacher training among first investigate relationship Practitioner notes What already known about this topic Given significance dialogue, programmes have implemented focusing provide teachers, techniques are increasingly utilised dialogue. small proportion studies investigated impacts practice, predominant pursuing performance. It unclear more accurate necessarily lead positive paper adds AI‐powered models, move through prompt engineering. significantly than Llama3. Through an experimental study, revealed using less substantial framework reported difference did not cause during Implications Practice and/or Policy Deep large can be integrated into effectively moderate also produce impressive outcomes satisfactory experience. In some scenarios, manner collaborate may pivotal AI's accuracy.
Язык: Английский
Процитировано
0Journal of Research on Technology in Education, Год журнала: 2025, Номер unknown, С. 1 - 19
Опубликована: Май 19, 2025
Язык: Английский
Процитировано
0Computers and Education Artificial Intelligence, Год журнала: 2024, Номер unknown, С. 100325 - 100325
Опубликована: Окт. 1, 2024
Язык: Английский
Процитировано
2