Závěrečná práce: Bc. Markéta Opichalová: Adoption of LLMs in the PV260 Software Quality Course: Exploration and Evaluation
Diplomová práce
Adoption of LLMs in the PV260 Software Quality Course: Exploration and Evaluation
Anotace
S rostoucí popularitou LLM v softwarovém inženýrství vyvstávají otázky o tom, jak mohou nástroje umělé inteligence podporovat výuku, aniž by ji nahrazovaly. Tato diplomová práce zkoumá možnosti integrace LLM do kurzu PV260 Software Quality na Masarykově univerzitě a hodnotí vzdělávací dopad fine-tuned modelu na úlohy refaktorování kódu. Teoretická část přináší přehled relevantních technologií LLM a …více
Abstract
The increasing adoption of LLMs in software engineering raises questions about how AI tools can support learning rather than replace it. This thesis explores LLM integration approaches for the PV260 Software Quality course at Masaryk University and evaluates the educational impact of a fine-tuned model on code refactoring tasks. The theoretical part reviews applicable LLM technologies and the pedagogical …více
Zadání práce
The student will:
- Review state-of-the-art LLMs and generative AI technologies applicable to software development education, documenting exploratory approaches for integration.
- Provide an exploratory implementation for the usage of LLMs for PV260 goals.
- Fine-tune an LLM tailored specifically for the context of PV260, focusing on identifying code smells and guiding students through the refactoring process.
- Design and execute an experiment with students of the PV260 Software Quality course to assess the impact of the model on a set of refactoring tasks.
- Analyze the experiment results using both quantitative metrics and qualitative feedback to assess how students perceive the usefulness of the model.
20. 5. 2026 09:52, doc. Bruno Rossi, PhD, učo 232464
Práce na příbuzné téma
Seznam prací, které mají shodná klíčová slova.
-
Chatbot for queries on domain-specific data
Bc. Branislav Bubán -
Trust Management in AI-based Agent Systems
Ing. Filip Fabka -
Exploring The Potential of Generative AI as a Topic in The EFL Classroom
Mgr. Viktor Šťastný -
Analysis of Bias and Reliability in LLM-Powered Search Engines
Bc. Martina Stankovičová -
A Quest for Information: Enhancing Game-Based Learning with LLM-Driven NPCs
Mgr. Tereza Tódová -
Evaluating the Ability of Large Language Models to Provide Code Quality Feedback
Ing. Michal Adler -
Use of Large Language Models in Programming Courses
Bc. Jakub Čížek -
Virtual AI partner for training situations in teaching
Ing. Mikuláš Heinz




