Development of a generative module based on artificial intelligence technologies for creating educational materials
Chenosov, Vasilii (2025)
Chenosov, Vasilii
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202505079673
https://urn.fi/URN:NBN:fi:amk-202505079673
Tiivistelmä
The thesis is devoted to developing a module based on AI technologies to create educational materials and evaluate test assignments. The proposed approach addresses the challenges faced by educational institutions: increasing volumes of content required and rapid adaptation to diverse specialties and learning pathways.
Studies have revealed inefficiencies in current methods: the lack of automated generation tools and the excessive human effort involved. Emerging trends in educational technology emphasize the incorporation of "intelligent assistants" for instructors. However, there is a key problem with harnessing AI: users require specialized expertise, including an understanding of neural network functioning, skill in constructing successful prompts, and maintenance of dialog context.
All aspects of developing a generative AI module (GenAI module) are explored in the thesis: business requirements are analyzed, functional and non-functional characteristics are determined, existing approaches are reviewed, and cutting-edge technological solutions accessible for module development are studied. The GenAI module was developed by the author of the thesis in partnership with Hyper Method Company (Saint Petersburg) as part of R&D efforts, subsequently integrated into the Learning Management System (LMS), and deployed in operation.
The graphical user interface of the GenAI module employs proprietary algorithms and templates, offering user authentication and authorization, facilitating the creation and management of educational materials (including generation request formation, editing, and storage), and enabling continuous monitoring of task completion and viewing of obtained results (24/7 availability). Consumer characteristics of the GenAI module include generation of materials (texts, exercises, tests) aligned with predefined subject areas and levels of complexity, management of generation parameters (e.g., text length, task difficulty), and automatic evaluation of students' control assignments. Future enhancements of the GenAI module encompass expansion of the prompt template library, adjustment of the system to accommodate varied data formats, and development of interfaces compatible with other LMS.
Studies have revealed inefficiencies in current methods: the lack of automated generation tools and the excessive human effort involved. Emerging trends in educational technology emphasize the incorporation of "intelligent assistants" for instructors. However, there is a key problem with harnessing AI: users require specialized expertise, including an understanding of neural network functioning, skill in constructing successful prompts, and maintenance of dialog context.
All aspects of developing a generative AI module (GenAI module) are explored in the thesis: business requirements are analyzed, functional and non-functional characteristics are determined, existing approaches are reviewed, and cutting-edge technological solutions accessible for module development are studied. The GenAI module was developed by the author of the thesis in partnership with Hyper Method Company (Saint Petersburg) as part of R&D efforts, subsequently integrated into the Learning Management System (LMS), and deployed in operation.
The graphical user interface of the GenAI module employs proprietary algorithms and templates, offering user authentication and authorization, facilitating the creation and management of educational materials (including generation request formation, editing, and storage), and enabling continuous monitoring of task completion and viewing of obtained results (24/7 availability). Consumer characteristics of the GenAI module include generation of materials (texts, exercises, tests) aligned with predefined subject areas and levels of complexity, management of generation parameters (e.g., text length, task difficulty), and automatic evaluation of students' control assignments. Future enhancements of the GenAI module encompass expansion of the prompt template library, adjustment of the system to accommodate varied data formats, and development of interfaces compatible with other LMS.