Details

Prototip s poizvedovanjem podprtega generiranja učnih vsebin iz gradiva
ID Trivić, Nina (Author), ID Pustišek, Matevž (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (2,93 MB)
MD5: 645070F3F1678959F2457C0788672654

Abstract
Umetna inteligenca je v zadnjih nekaj letih doživela izjemen razvoj, največ na področju generativnih modelov in obdelave naravnega jezika. Ta hitri tehnološki napredek je spodbudil številne možnosti za uporabo UI v izobraževalnem in raziskovalnem kontekstu. Orodja, kot so veliki jezikovni modeli, omogočajo avtomatsko ustvarjanje vsebin, personalizirano učenje ter učinkovitejšo podporo študentom in učiteljem. Posledično se umetna inteligenca vse bolj vključuje v študijske procese kot orodje za izboljšanje dostopa do znanja, povečanje učinkovitosti in podporo pri raziskovalnem delu. V okviru diplomskega dela je bila razvita aplikacija kot prototip, ki temelji na arhitekturi RAG. Aplikacija omogoča nalaganje učnih gradiv iz spletne učilnice, njihovo obdelavo in shranjevanje v vektorsko bazo, vodenje pogovora z jezikovnim modelom na podlagi teh gradiv ter generiranje kvizov v formatu, ki omogoča enostaven uvoz v spletno učilnico. Prototip predstavlja praktičen primer uporabe umetne inteligence za izboljšanje učnega procesa, saj lahko študentom olajša učenje in pripravo na izpite, profesorjem pa nudi podporo pri oblikovanju učnih vsebin. Delo prispeva k boljšemu razumevanju umetne inteligence v izobraževalnem okolju ter predstavlja osnovo za nadaljnji razvoj orodij, ki bi lahko bila neposredno vključena v platformo Moodle.

Language:Slovenian
Keywords:umetna inteligenca, RAG, Moodle, semantično iskanje, vektorska podatkovna baza, Qdrant, veliki jezikovni modeli
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FE - Faculty of Electrical Engineering
Year:2025
PID:20.500.12556/RUL-172582 This link opens in a new window
COBISS.SI-ID:270279171 This link opens in a new window
Publication date in RUL:09.09.2025
Views:242
Downloads:30
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Prototype of Retrieval-Augmented Generation of Learning Content from Course Materials
Abstract:
Artificial intelligence has seen remarkable advances in recent years, especially in the fields of generative models and natural language processing. This rapid technological progress has opened numerous opportunities for applying AI in educational and research contexts. Tools such as large language models enable automated content creation, personalized learning, and more efficient support for students and teachers. Consequently, AI is increasingly being integrated into study processes as a tool for improving access to knowledge, boosting efficiency, and supporting research activities. As part of this thesis, an application prototype based on the RAG architecture was developed. The application enables the uploading of learning materials from the online classroom, their processing and storage in a vector database, conducting conversations with a language model based on these materials, and generating quizzes in a format that allows simple importing into the online classroom. The prototype demonstrates a practical example of applying artificial intelligence to enhance the learning process, as it can support students in studying and exam preparation while also assisting teachers in creating learning content. This work contributes to a better understanding of artificial intelligence and provides a foundation for further development of tools that could be directly integrated into the Moodle platform.

Keywords:artificial intelligence, RAG, Moodle, semantic search, vector database, Qdrant, large language models

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back