Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 58636 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Fazli, Mohammad Amin
- Abstract:
- With the rapid advancement of artificial intelligence technologies, particularly large language models, the automatic generation of educational content has attracted significant attention among researchers and instructional designers. The importance of this topic lies in the fact that designing and developing educational courses is a time-consuming and costly process, while many educational institutions and instructors seek to leverage modern tools to provide learners with relevant, interactive, and effective content. At the same time, the quality of the content and its alignment with psychological principles of learning and instructional design play a crucial role in the effectiveness of the learning process. Recent studies in this field demonstrate various attempts to employ language models for course development, including the generation of summaries, worked examples, automated assessments, and even lecture slides. However, many of these efforts either lack a well-defined educational framework or fail to adequately evaluate the quality and validity of the generated content. This research proposes a hybrid approach in which course development is carried out with the assistance of large language models while simultaneously embedding educational psychology principles and well-established instructional design frameworks (such as Merrill’s Principles and Gagné’s Nine Events of Instruction) into the content generation process. The approach defines a step-by-step procedure that includes prompt engineering, the generation of textual and interactive content, and the evaluation of outputs based on educational criteria. The key contribution of this study is the integration of language models’ capabilities in producing diverse and rapid content with the scientific foundations of instructional psychology. As a result, the generated content is not only linguistically and structurally coherent but also aligned with learners’ needs and the principles of effective learning. Findings suggest that by defining a standardized process, the outputs of language models can be transformed into structured and reliable educational materials, thereby taking a significant step toward improving learning quality and advancing intelligent educational systems
- Keywords:
- Large Language Model ; Prompt Engineering ; Automated Course Generation ; Instructional Design
Digital Object List
-
محتواي کتاب
- view
Bookmark