Σχεδίαση και υλοποίηση δυναμικής πλατφόρμας ηλεκτρονικής μάθησης για προσωποποιημένη εκπαίδευση με Τεχνικές Ανάκτησης - Ενίσχυσης (RAG) και Μεγάλα Γλωσσικά Μοντέλα (LLMs).
Design and implementation of a dynamic E-Learning platform for personalized education using Retrieval - Augmented Generation (RAG) and Large Language Models (LLMs).

View/ Open
Keywords
Μεγάλα Γλωσσικά Μοντέλα ; Εκπαιδευτική Τεχνολογία ; Retrieval Augmented Generation ; Chatbot ; Large Language Models ; E Learning ; Τεχνητή νοημοσύνηAbstract
The rapid advancement of Artificial Intelligence and the increasing demand for personalized learning have introduced new technological approaches in the field of e-learning.
This thesis presents the development of a comprehensive training platform based on Retrieval-Augmented Generation (RAG) techniques combined with Large Language Models (LLMs).
The application allows users to upload their own educational material, automatically generate slides, quizzes, and interact with a personalized chatbot, while tracking their learning progress through dynamic assessments and certification.
A key innovation of the platform lies in its ability to process Greek-language educational documents, providing fully documented responses and quiz questions directly linked to the uploaded content.
The system architecture follows a modular, extensible design implemented in Python, utilizing libraries such as Streamlit, LlamaIndex, and OpenAI APIs.
Experimental evaluation across different datasets demonstrated high reliability, interactivity, and adaptability of the platform.
This work highlights the potential of integrating RAG and LLM technologies to enhance learning experiences in both academic and professional educational environments.