Search Results

Documents authored by Horváth, Marek

Document
DOI: 10.4230/OASIcs.ICPEC.2026.9
From Repositories to Practice: LLM-Based Personalization in Programming Education

Authors: Marek Horváth, Michaela Ďurkovičová, Lenka Bubeňková, and Emília Pietriková

Published in: OASIcs, Volume 145, 7th International Computer Programming Education Conference (ICPEC 2026)

Abstract
Many programming education systems personalize practice mainly from test results, progress indicators, or explicitly selected user parameters. This paper instead uses students' existing source-code repositories as evidence for personalized programming practice. It presents the design and implementation of a web application that connects to a university version control system, analyzes student-selected repositories, and generates programming challenges targeted at weaknesses identified in the submitted source code. The prototype analyzes the current state of the selected repositories using LLM-based assessment of code-quality indicators such as readability, modularity, duplication, input handling, edge-case handling, and algorithmic complexity. Students can also generate parameterized tests, submit solutions to programming challenges, and receive an automated score with textual feedback. The system stores student activity, test history, challenge history, achieved scores, and generated feedback as an operational student profile shown to students and teachers. The prototype was evaluated with 11 students at the Technical University of Košice. The evaluation demonstrates technical feasibility, usability, and functional operation of the prototype, but it does not measure learning gains, retention, or improvement in programming performance.
Cite as

Marek Horváth, Michaela Ďurkovičová, Lenka Bubeňková, and Emília Pietriková. From Repositories to Practice: LLM-Based Personalization in Programming Education. In 7th International Computer Programming Education Conference (ICPEC 2026). Open Access Series in Informatics (OASIcs), Volume 145, pp. 9:1-9:13, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{horvath_et_al:OASIcs.ICPEC.2026.9,
  author =	{Horv\'{a}th, Marek and \v{D}urkovi\v{c}ov\'{a}, Michaela and Bube\v{n}kov\'{a}, Lenka and Pietrikov\'{a}, Em{\'\i}lia},
  title =	{{From Repositories to Practice: LLM-Based Personalization in Programming Education}},
  booktitle =	{7th International Computer Programming Education Conference (ICPEC 2026)},
  pages =	{9:1--9:13},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-443-7},
  ISSN =	{2190-6807},
  year =	{2026},
  volume =	{145},
  editor =	{Portela, Filipe and Matos, Lu{\'\i}s and Guimar\~{a}es, Tiago},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.ICPEC.2026.9},
  URN =		{urn:nbn:de:0030-drops-267460},
  doi =		{10.4230/OASIcs.ICPEC.2026.9},
  annote =	{Keywords: Programming Education, Personalized Programming Practice, Student Profiles, Formative Assessment, Automated Feedback, Learning Analytics, Software Engineering Education}
}
Document
DOI: 10.4230/OASIcs.ICPEC.2026.10
Bridging Game Development and Virtual Reality Education Through Blender-Based Tutorials

Authors: Lenka Bubeňková, Ľubomír Urbančík, Emília Pietriková, and Marek Horváth

Published in: OASIcs, Volume 145, 7th International Computer Programming Education Conference (ICPEC 2026)

Abstract
The integration of 3D modeling tools into STEM education can provide opportunities for creative digital production, technical skill development, and hands-on interaction with course content. This study reports an educational evaluation of Blender-based tutorials implemented in two university courses: Game Development (GD) and Systems of Virtual Reality (SVR). Over four semesters, 420 anonymous questionnaire responses were collected from GD and SVR course offerings with a total enrollment of 532 students. The GD course enrolled 428 students and yielded 342 questionnaire responses, while the SVR course enrolled 104 students and yielded 78 questionnaire responses. This corresponds to questionnaire coverage of 79.9% in GD, 75.0% in SVR, and 78.9% overall. Because valid response counts differed across questionnaire items, the results are reported using item-level valid response counts and interpreted descriptively. Across valid questionnaire responses, many respondents, especially in SVR, reported moderate-to-high confidence in using Blender after the tutorials and self-reported understanding of core modeling concepts. Some GD teams even applied self-created or modified Blender assets in game jam submissions, where Blender use was not required. These findings suggest that Blender-based tutorials can be meaningfully integrated into GD and SVR courses as an introductory tool for 3D modeling workflows. The study provides descriptive evidence of student reception and course-level application, but does not establish causal effects on modeling competence or learning outcomes.
Cite as

Lenka Bubeňková, Ľubomír Urbančík, Emília Pietriková, and Marek Horváth. Bridging Game Development and Virtual Reality Education Through Blender-Based Tutorials. In 7th International Computer Programming Education Conference (ICPEC 2026). Open Access Series in Informatics (OASIcs), Volume 145, pp. 10:1-10:16, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{bubenkova_et_al:OASIcs.ICPEC.2026.10,
  author =	{Bube\v{n}kov\'{a}, Lenka and Urban\v{c}{\'\i}k, \v{L}ubom{\'\i}r and Pietrikov\'{a}, Em{\'\i}lia and Horv\'{a}th, Marek},
  title =	{{Bridging Game Development and Virtual Reality Education Through Blender-Based Tutorials}},
  booktitle =	{7th International Computer Programming Education Conference (ICPEC 2026)},
  pages =	{10:1--10:16},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-443-7},
  ISSN =	{2190-6807},
  year =	{2026},
  volume =	{145},
  editor =	{Portela, Filipe and Matos, Lu{\'\i}s and Guimar\~{a}es, Tiago},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.ICPEC.2026.10},
  URN =		{urn:nbn:de:0030-drops-267479},
  doi =		{10.4230/OASIcs.ICPEC.2026.10},
  annote =	{Keywords: Blender, game development, virtual reality, Unity, education}
}
Document
DOI: 10.4230/OASIcs.ICPEC.2024.4
Adaptation of Automated Assessment System for Large Programming Courses

Authors: Marek Horváth, Tomáš Kormaník, and Jaroslav Porubän

Published in: OASIcs, Volume 122, 5th International Computer Programming Education Conference (ICPEC 2024)

Abstract
This paper presents a new automated assessment system tailored for programming courses, addressing the challenge of evaluating a large number of students in extensive courses at the Technical University of Košice. The primary issue with current systems is their inability to handle massive course loads while ensuring objective evaluation and timely feedback. Our proposed system enhances the scalability of the assessment process, allowing for the simultaneous handling of a greater volume of assignments. It is designed to provide regular and systematic feedback to students, supporting their continuous learning and improvement. To ensure the objectivity of evaluations, the system utilizes a variety of unit test suites, selecting them randomly in each assessment to discourage students from hardcoding solutions. This approach not only supports fair and precise assessments but also significantly reduces the administrative burden on educators, enabling them to meet a wide range of educational demands.
Cite as

Marek Horváth, Tomáš Kormaník, and Jaroslav Porubän. Adaptation of Automated Assessment System for Large Programming Courses. In 5th International Computer Programming Education Conference (ICPEC 2024). Open Access Series in Informatics (OASIcs), Volume 122, pp. 4:1-4:11, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2024)

Copy BibTex To Clipboard

@InProceedings{horvath_et_al:OASIcs.ICPEC.2024.4,
  author =	{Horv\'{a}th, Marek and Korman{\'\i}k, Tom\'{a}\v{s} and Porub\"{a}n, Jaroslav},
  title =	{{Adaptation of Automated Assessment System for Large Programming Courses}},
  booktitle =	{5th International Computer Programming Education Conference (ICPEC 2024)},
  pages =	{4:1--4:11},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-347-8},
  ISSN =	{2190-6807},
  year =	{2024},
  volume =	{122},
  editor =	{Santos, Andr\'{e} L. and Pinto-Albuquerque, Maria},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.ICPEC.2024.4},
  URN =		{urn:nbn:de:0030-drops-209735},
  doi =		{10.4230/OASIcs.ICPEC.2024.4},
  annote =	{Keywords: Automated Assessment, Informatics Education, Programming Feedback Systems, Continuous Integration in Education, Code Quality Analysis, Educational Technology, Computer Science Education}
}
Document
DOI: 10.4230/OASIcs.ICPEC.2024.18
Using ChatGPT During Implementation of Programs in Education

Authors: Norbert Baláž, Jaroslav Porubän, Marek Horváth, and Tomáš Kormaník

Published in: OASIcs, Volume 122, 5th International Computer Programming Education Conference (ICPEC 2024)

Abstract
This paper examines the impact of ChatGPT on programming education by conducting an empirical study with computer science students at the Department of Computers and Informatics at the Technical University in Košice. The study involves an experiment where students in a Component Programming course use ChatGPT to solve a programming task involving linked lists, comparing their performance and understanding with a control group that does not use the AI (artificial intelligence) tool. The task necessitated the implementation of a function to add two numbers represented as linked lists in reverse order. Our findings indicate that while ChatGPT significantly enhances the speed of task completion - students using it were nearly three times quicker on average - it may also detract from deep understanding and critical thinking, as evidenced by the uniformity and superficial engagement in solutions among the ChatGPT group. On the other hand, the group working independently displayed a broader variety of solutions and deeper interaction with the problem, despite slower completion times and occasional inaccuracies. The results highlight a dual-edged impact of AI tools in education: while they enhance efficiency, they may undermine the development of critical thinking and problem-solving skills. We discuss the implications of these findings for educational practices, emphasizing the need for a balanced approach that integrates AI tools without compromising the depth of learning and understanding in students.
Cite as

Norbert Baláž, Jaroslav Porubän, Marek Horváth, and Tomáš Kormaník. Using ChatGPT During Implementation of Programs in Education. In 5th International Computer Programming Education Conference (ICPEC 2024). Open Access Series in Informatics (OASIcs), Volume 122, pp. 18:1-18:9, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2024)

Copy BibTex To Clipboard

@InProceedings{balaz_et_al:OASIcs.ICPEC.2024.18,
  author =	{Bal\'{a}\v{z}, Norbert and Porub\"{a}n, Jaroslav and Horv\'{a}th, Marek and Korman{\'\i}k, Tom\'{a}\v{s}},
  title =	{{Using ChatGPT During Implementation of Programs in Education}},
  booktitle =	{5th International Computer Programming Education Conference (ICPEC 2024)},
  pages =	{18:1--18:9},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-347-8},
  ISSN =	{2190-6807},
  year =	{2024},
  volume =	{122},
  editor =	{Santos, Andr\'{e} L. and Pinto-Albuquerque, Maria},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.ICPEC.2024.18},
  URN =		{urn:nbn:de:0030-drops-209876},
  doi =		{10.4230/OASIcs.ICPEC.2024.18},
  annote =	{Keywords: generative artificial intelligence, chatbot, ChatGPT, prompt engineering, source code generation}
}
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact