|
|
 |
Hibridni događaj
|
| Radovi
Predsjedatelj: Duško Lukač
|
P. Smes, Đ. Kotlar, T. Babić (Vern University, Zagreb, Croatia)
Students Awareness of the Link between Social Media Algorithms and Cognitive Bias 
Social media algorithms play a significant role in shaping the content users encounter by prioritizing personalized information. This process may contribute to cognitive bias by reinforcing existing beliefs and limiting exposure to diverse perspectives, which can influence users’ information literacy skills. The aim of this study was to examine university students’ awareness of social media algorithms and their perceptions of how algorithm-driven content personalization affects cognitive bias, misinformation, and information evaluation. The research employed a quantitative survey design using an online questionnaire distributed among university students in Croatia. A total of 73 participants completed the survey. The questionnaire included Likert-scale items measuring awareness of algorithmic influence, perceived reinforcement of personal beliefs, emotional engagement with content, and attitudes toward information verification. The results indicate that most students are aware that algorithms shape the content they see online and perceive that emotionally engaging content spreads faster than verified information. The findings highlight the influence of algorithm-driven personalization on cognitive bias and emphasize the importance of strengthening information literacy in digital environments.
|
D. Velickovski, S. Koceski, N. Koceska (Goce Delcev University, Stip, Macedonia)
Mobile Application for On-Demand Conversion of Print Material into Dyslexia-Friendly Formats
Dyslexia is a learning difficulty that affects reading fluency, accuracy, and text comprehension, despite normal sensory and cognitive abilities. However, the majority of printed books, educational literature, and academic materials are produced using standard fonts that are not optimized for readers with dyslexia, often resulting in reduced reading comfort, comprehension difficulties, and slower reading performance. Adapting textual presentation through specialized fonts, letter spacing, and background color adjustments has been shown to improve readability for individuals with dyslexia. This paper presents a mobile application for the Android operating system that enables on-demand conversion of printed materials into dyslexia-friendly digital formats. The application allows users to capture images of printed text, automatically extract textual content using Optical Character Recognition, and render the text with customizable visual parameters, including font type, font size, letter spacing, and background color. The developed mobile application has been evaluated in terms of usability and readability, and the obtained results are presented and discussed, demonstrating its potential to support inclusive and accessible learning for readers with dyslexia.
|
S. Reichel, P. Jurić, T. Babić (Vern University, Zagreb, Croatia)
Research on perceived credibility of Micro-Influencers Among Students
In the digital environment, the term influencer has emerged as a prominent phenomenon, characterized by differences between large-scale influencers and smaller content creators known as micro-influencers. The focus in this paper is on micro-influencers, examining their role within contemporary digital media network. For the purposes of this study, research was conducted to explore students’ trust in digital media. The aim of the study was to investigate how Vern’ University students engage with micro-influencers content across various social media platforms and how they evaluate the sincerity and credibility of such content in comparison to traditional advertising and major influencers. The research addresses several key aspects, including perceived authenticity, trustworthiness of sponsored content, the balance between commercial collaboration and authenticity, and the alignment of micro-influencers values with those of their audience. The research was conducted among students of Vern’ University using a survey method with Likert-scale questions. Data were collected between mid-December 2025 and mid-January 2026, enabling a systematic analysis of students’ perceptions of micro-influencer credibility. The findings aim to contribute to a clearer understanding of the factors that influence students’ trust in micro-influencers and to provide a foundation for future research on the role of micro-influencers in contemporary digital media.
|
D. Lukac (CCEC - Centre of Competence for EPLAN Certification (UG), Siegburg, Germany), M. Kadrnožková (Charles University, Faculty of Education, Prague, Czech Republic)
AI in Academic Evaluation: Current Landscape, Key Challenges, and Approaches to Mitigating AI-Driven Plagiarism
Nowadays, educational institutions face new challenges in assessing written assignments and exams, as entire texts—or substantial parts of them—can now be generated by artificial intelligence (AI) tools such as ChatGPT. When only a few papers are submitted, instructors can still reliably evaluate them. However, when the number of submissions rises into the hundreds, the assessment process can become a significant burden due to limited time and resources.
In response to these challenges, various automated essay evaluation tools have been developed in the past, including Automated Essay Scoring (AES) systems and Computerized Adaptive Testing (CAT) environments. There are also AI-detection tools, such as ChatGPT content detectors, which attempt—often with varying levels of accuracy—to identify whether a given text was produced by an AI system.
It is well known that ChatGPT can generate convincing imitations of human writing but cannot reliably assess the relevance, accuracy, or depth of the information it provides. It cannot form an opinion or produce genuinely original, conceptually deep text. Instead, it generates content based on user prompts and can, similar to Wikipedia, assist in retrieving or summarizing information. Additionally, ChatGPT does not create original text after paraphrasing existing content, which increases the risk of so-called “paraphrasing plagiarism.”
At some universities, it has been decided that AI-based systems such as ChatGPT may be used by students during examinations, alongside traditional resources such as books, articles, and online materials. However, if students use AI-generated text, they must cite it as they would any other source. In addition, students are often required to include the complete AI-generated output, along with the prompts and instructions they provided to the system. Failure to comply with these requirements is considered plagiarism.
In this paper, we address three central questions:
(1) What challenges do AI-based tools such as ChatGPT pose for assessing students’ written work?
(2) What potential do these systems offer for improving or optimizing the evaluation of written assignments?
(3) How can AI be used as a pedagogical tool to support teaching and learning, and should examiners focus on detecting AI-generated content, or should AI be integrated as a legitimate aid that enhances the learning process?
To explore these questions, we first review the current state of AI-based systems such as ChatGPT in educational contexts. We then evaluate selected applications using free-text tasks. Finally, we discuss potential approaches for effectively assessing large volumes of written assignments and exams, acknowledging that many may be partially—or entirely—produced using AI tools. This study considers both summative and formative assessment perspectives.
|
L. Marković, G. Šantić, T. Babić (Vern University, Zagreb, Croatia)
Digital Detox in Academic Contexts - University Students’ Awareness and Attitudes
Excessive use of digital devices, particularly smartphones, has been associated with decreased concentration and difficulties in completing academic tasks among university students. In this context, the concept of digital detox has gained attention as a strategy for intentionally limiting digital device use. However, less is known about how students themselves understand this concept and perceive its relevance within academic settings. The aim of this study is to examine university students’ awareness of the concept of digital detox, their perceptions of the impact of digital device use on concentration and academic functioning, and their attitudes toward limiting digital device use during studying. The study employed a quantitative research design using an online self-report questionnaire administered to a sample of university students in Croatia. The questionnaire consisted of Likert-scale items measuring awareness of digital detox, perceived academic distraction caused by digital devices, and attitudes toward limiting device use in academic contexts. Data were analysed using descriptive statistical methods. The findings provide insight into students’ perceptions and attitudes toward digital detox and highlight the perceived role of digital device use in academic environments.
|
V. Lugarić, I. Brčić, A. Sovic Kržić (University of Zagreb Faculty of Electrical Engineering and Computing, Zagreb, Croatia)
Assessing LEGO Model Assembly Difficulty
This work reports a pilot workshop study (n=35) assessing task performance for four LEGO-based models. Each model consists of two pre-built structures, which participants were required to connect to form a complete model. Four stations were used, one per model, with fixed initial orientations of the pre-built structures and a visual reference image that depicted the correctly assembled final model configuration. Participants sequentially completed all four stations. For each model, start time, end time, and binary outcome (success/fail) were recorded manually. Primary metrics as assembly duration and task success rate per model are compared, enabling between-model comparison. For two models, the joining success rate was 100%, and those two models were also the fastest to assemble. This methodology supports quantitative screening of assembly difficulty while the pilot dataset provides an empirical basis for selecting discriminative performance indicators and refining station instructions and task design.
|
A. Kiričenko (Zagreb University of Applied Sciences, Zagreb, Croatia)
An End-to-End Open-Source Toolchain for Rapid In-House PCB Prototyping in Electrical Engineering Education
Electrical engineering students regularly design electronic printed circuit boards for capstone assignments and extracurricular projects. While industrial PCB production has become increasingly affordable, lead times and shipping delays remain significant bottlenecks, particularly when multiple design iterations are required. In-house PCB fabrication addresses these delays while providing high educational value by allowing students to engage directly with the manufacturing process. However, the primary cost barrier for in-house solutions is often not the CNC milling hardware itself, but the substantial licensing fees associated with specialized proprietary software. This paper proposes a comprehensive open-source software workflow to bridge this gap. The proposed toolchain integrates PCB design, gerber file generation and inspection, CAM processing, and CNC milling. To validate this approach, several prototype projects were manufactured, establishing an optimal set of parameters for reliable production. Results demonstrate that the entire cycle—from initial schematic design to a finished physical board—can be successfully completed within a single workday.
|
I. Heđi (Virovitica University of Applied Sciences, Virovitica, Croatia), M. Kepec (Infobip d.o.o., Virovitica, Croatia), M. Poldrugač (Valcon d.o.o., Virovitica, Croatia)
Using Infrastructure as Code for Enhancing Cloud Computing Education in Constrained Environments
Cloud computing courses typically employ multi-week progressive laboratory exercises where students build upon previously configured infrastructure. This pedagogical approach, while valuable for experiential learning, creates significant challenges in constrained teaching environments such as AWS Academy Learner Lab. Students accumulate configuration debt from earlier mistakes, causing subsequent labs to fail in ways that are difficult to diagnose. As later exercises implicitly depend on correctly configured resources from previous weeks, instructors dedicate considerable time to troubleshooting inherited misconfigurations rather than introducing new concepts. Traditional assessment methods further compound this issue by emphasizing conceptual knowledge over practical troubleshooting and systems reasoning skills. This paper presents usage of infrastructure as code that addresses both scaffolding and assessment challenges in cloud computing education. Infrastructure as code templates provision reproducible baseline configurations for each laboratory week, eliminating configuration drift and dependency failures across progressive exercises. Second, deliberately misconfigured templates inspired by chaos engineering and fault injection methodologies serve as structured assessment instruments for troubleshooting competencies. Implementation using Terraform and AWS CloudFormation in AWS Academy Learner Lab demonstrates improved instructional efficiency, enhanced student troubleshooting capabilities, and better alignment with industry operational practices.
|
J. Mezak (Faculty of Teacher Education, University of Rijeka,, Rijeka, Croatia), S. Vranić (Faculty of Teacher Education, University of Rijeka, Rijeka, Croatia)
Preparing Future Early Childhood Educators for Educational Robotics and Inquiry-Based Learning
Computational thinking is increasingly recognised as an important approach for developing problem-solving and basic programming skills, with algorithmic thinking as one of its key components. In early childhood education (ECE), these competences can be fostered through play-based and inquiry-oriented activities that provide meaningful learning contexts for young children. Educational robotics offers concrete, hands-on opportunities for developing early foundations of computational thinking through exploration and collaboration. In this context, the present study explored changes in pre-service ECE educators’ attitudes toward educational robotics and inquiry-based learning (IBL) following participation in a university-level course. The course was developed within the Erasmus+ project Greencode and was based on hands-on, inquiry-oriented learning that integrated robotics, IBL, and environmental topics. A pre-test–post-test design with independent samples was used, involving 33 pre-service educators in the pre-test and 31 in the post-test (all female). As individual responses could not be matched, the results reflect group-level differences rather than individual change. The findings indicate a statistically significant positive difference in attitudes toward educational robotics, while no significant difference was found for attitudes toward IBL at the composite scale level. These results suggest that practice-oriented, inquiry-based courses may support the development of positive attitudes toward educational robotics in initial teacher education, while also highlighting the need for further research with larger and more diverse samples.
|
J. Kordek (Zagreb University of Applied Sciences, Zagreb, Croatia), Z. Lavrić (Ericsson Nikola Tesla, Zagreb, Croatia), M. Miletić (Zagreb University of Applied Sciences, Bjelovar, Croatia)
Design and Development of the Electronic Board Game Man, Don’t Get Angry for Embedded Systems Showcase
This paper presents the design and development process of the electronic board game "Man, Don't Get Angry." The device is primarily used during an open day event at the Zagreb University of Applied Sciences to promote the embedded systems laboratory. Embedded systems are showcased in an interesting way, combining a new concept with something that the visitors are already familiar with. The paper details the challenges encountered during development, including an experimental procedure that was used for determining the color, number, and height of 3D printed layers required to achieve a satisfactory LED diffusion effect on the enclosure’s top surface. It also covers the system design and the design of double-sided printed circuit boards (PCBs). The device comprises multiple PCBs, including one main PCB and several smaller PCBs with touch-sensitive control surfaces. The enclosure design was done in a 3D CAD environment using 3D models of previously developed PCBs and fabricated via 3D printing. The complete documentation is available on GitHub to share knowledge with the community and encourage further engagement.
|
D. Valenčić (University of Applied Sciences Velika Gorica, Velika Gorica, Croatia)
Implementation of Cisco Network Academy in IT University Education
Cisco Systems is one of the world's leading manufacturers of Internet solutions. Most of the data that travels over the Internet passes through Cisco equipment (routers, switches, firewalls or other equipment) at some point. Cisco Networking Academy (NetAcad) is a global educational program that Cisco launched in 1997. It is a non-profit program that cooperates with schools, universities and educational centers around the world. The basic characteristics of the program are the acquisition of both theoretical and practical knowledge, education in various IT areas (computer networks, cybersecurity, IoT, programming, ...) and preparation for IT certifications. The paper will present the basic characteristics of the Cisco Networking Academy, the types of courses offered, the methods of delivering the courses, the connection of the courses with industry certificates, etc. It will also present the possibilities of applying NetAcad courses in higher education in IT studies (methods of application, advantages, challenges, ...)
|
I. Valova, T. Kaneva, G. Kanev, N. Todorova (Ruse University "Angel Kanchev", Ruse, Bulgaria)
AI Agents in University Education: A Benefit or an Obstacle?
This study presents a literature review on the application of AI agents in education and analyzes the benefits and challenges reported in the scientific literature related to their use in an academic environment. The review of publications shows that AI agents are considered effective tools for personalized learning, automated assistance and facilitated access to information. At the same time, many authors emphasize risks such as over-reliance on systems, reduced critical thinking, inaccurate or misleading answers, as well as difficulties in judging when and how to use artificial intelligence assistance. As part of the study, an experiment was conducted among students who solved an identical learning task in two variants: independently and with access to an AI agent. The comparative analysis of the results shows that despite the expectations of significant improvement, the use of an AI agent does not lead to a significant increase in the quality of the final solutions. The data reveal that the main difficulties of students when using AI agents include incorrect interpretation of the received information, mechanical following of the generated suggestions, lack of skills for checking the credibility, as well as limited understanding of the context of the task.
|
| Radovi
Predsjedatelj: Frano Škopljanac-Mačina
|
F. Škopljanac-Mačina, S. Horvatić, D. Popović (University of Zagreb Faculty of Electrical Engineering and Computing, Zagreb, Croatia)
Exploring the Use of Generative Artificial Intelligence Tools for Solving Electrical Engineering and Physics Problems
This paper explores the effectiveness of generative artificial intelligence chatbot tools, such as ChatGPT, DeepSeek and Gemini, in solving introductory electrical engineering and physics problems at undergraduate engineering level. The experiments were performed using representative sets of written exams from freshman year courses Fundamentals of Electrical Engineering and Physics at University of Zagreb Faculty of Electrical Engineering and Computing. These exams consisted of textual multiple choice questions and computational problems, both usually accompanied by figures, such as circuit diagrams or graphs. The basic testing setup consisted of a single prompt instructing that the selected generative AI tool solves an actual past exam which is then uploaded as a single PDF file. After iterative testing using all selected exams the average score for each tested generative AI tool is calculated. For Physics exams Gemini achieved the best average score (87.5%), followed by ChatGPT (68%) and DeepSeek (61.5%). Interestingly, for Fundamentals of Electrical Engineering the performances of all tools were substantially lower, but Gemini was again the best with average score of 58%, followed by DeepSeek (47%) and ChatGPT (44%). In the paper we also describe different more advanced testing setups which can be used to improve the chatbots' performance.
|
W. Werth, C. Ungermanns (CUAS, Villach, Austria)
From Data to Models: System Identification Education Using a Laboratory System in a Computer-Supported Learning Environment
System identification is a key competence in Control Engineering education, bridging theoretical modeling methods and real-world system behavior. This paper focuses on a practical approach for teaching system identification within a bachelor-level Systems Engineering program, using an industrial Festo laboratory system as a central learning platform.
The proposed approach integrates simulation-based preparation, experimental data acquisition, and model validation. Students identify linear dynamic models from measured input–output data and apply the resulting models to controller design and analysis tasks. The Festo lab system serves as a representative mechatronic platform, enabling realistic experimentation while maintaining a structured learning environment.
The paper outlines the didactic integration of the identification workflow, including experiment design, parameter estimation, and model assessment, and discusses its role in fostering a deeper understanding of system dynamics and modeling assumptions. Initial observations from course implementation indicate that the use of this modern laboratory system enhances student engagement and supports the transfer of theoretical concepts to practical applications. This contribution demonstrates how system identification using industrial-like laboratory equipment can be effectively und successfully embedded into undergraduate Control Engineering education.
|
C. Ungermanns, W. Werth, M. Ungermanns (Carinthia University of Applied Sciences , Villach, Austria)
From Chaos to Kits: A Lightweight Sorting Approach for Construction Bricks in STEM Learning
The sorting of used modular interlocking plastic elements, commonly known from educational and recreational building sets (hereafter referred to as Construction Bricks), has become increasingly relevant in the context of recycling, reuse, and the resale of individual components. Due to the wide variety of shapes, colors, and weights, manual sorting is time-consuming and error-prone. While existing automated systems typically rely on computationally intensive AI algorithms and cloud-based image processing, this project explores a resource-efficient, locally executable alternative.
The proposed system is based on a combination of mechanical separation via vibration, weight-based identification, and basic image processing. It is currently under prototypical development and aims to identify individual bricks, determine key physical attributes, and compare them against the part lists of known construction sets. This makes it suitable not only for reassembling original kits but also for supporting the targeted selection of parts for MOCs (My Own Creations) – custom models frequently built by hobbyists and educators.
Designed around low-cost hardware such as the Raspberry Pi and using minimal computational resources, the system offers an accessible platform for hands-on learning in automation, sensor technology, and algorithmic decision-making. It thus serves as both a practical exploration of automated classification and a didactic tool for STEM education.
|
I. Sekovanić, T. Adamović (Veleučilište u Bjelovaru, Bjelovar, Croatia), D. Vidić (Gimnazija Bjelovar, Bjelovar, Croatia)
Automated Assessment of University Programming Exams Using Gemini API
Manual assessment of programming assignments in higher education is a time-consuming and in some cases inconsistent process. With the development of Large Language Models (LLMs), there is a significant opportunity to automate this workflow. This paper presents a Python-based system utilizing the Google Gemini API to grade student programming exams automatically. To achieve optimal results the LLM is instructed by structured prompt which tells it to act as "strict assistant" and compare student code against a reference solution. To evaluate the effectiveness of this approach, the performance of different LLM variants is compared in terms of grading accuracy, processing speed and cost. Also, the correlation between the automated grades and traditional human assessment on a dataset of real student exams is analysed. The results identify an optimal balance between computational cost and grading reliability, suggesting that LLM-based tools can effectively serve as assistants in engineering education.
|
M. Bednjanec, M. Kovač (Aspira University of Applied Sciences, Zagreb, Croatia)
Application of Software Solutions for Monitoring and Controlling Student Activity
The widespread adoption of digital technologies in education has significantly reshaped the teaching and learning process, introducing both new opportunities and challenges related to classroom management, student engagement, and online safety. Software solutions for monitoring and controlling student activity provide educators with the ability to supervise digital device usage in real time, regulate internet access, block selected websites or applications, and maintain alignment between technology use and instructional goals. These systems contribute to reducing distractions, enhancing cybersecurity, and fostering a more structured digital learning environment.
This paper presents an analysis of the implementation and impact of such software solutions in educational settings, supported by an empirical survey conducted among teachers and students. The survey examines perceptions of effectiveness, usability, and ethical implications, with particular emphasis on privacy, transparency, and trust. The results indicate differing perspectives between teachers and students, highlighting the need for clearly defined usage policies and informed consent. Overall, the study concludes that software-based monitoring tools, when implemented responsibly and in accordance with ethical and legal standards, can improve instructional efficiency and learning outcomes while preserving students’ rights.
|
K. Pavlina, A. Pongrac Pavlina, A. Modrušan (University of Zagreb, Zagreb, Croatia)
Teachers’ use and perceptions of digital educational resources developed within the e-Škole project
The integration of digital educational resources plays a crucial role in supporting contemporary teaching practices and enhancing teachers’ digital competences. This paper examines teachers’ use and perceptions toward digital educational resources developed within e-Škole project which represents a comprehensive national initiative designed to support digital transformation of primary and secondary education in Croatia. Data was collected using a structured questionnaire administered to primary school teachers. Descriptive statistical methods were applied to analyze teachers’ responses, including frequencies, arithmetic means and standard deviations.
The results indicate that teachers demonstrate varying levels of engagement with different types of digital resources. Professional development activities, such as online courses and webinars, are used more frequently than other resources, while tools such as e-Laboratory, teaching scenarios, and Loomen learning management system show lower usage levels. Teachers generally perceive the digital resources as useful and adequately organized, although notable variability in responses suggests differences in individual experience and contextual factors.
The findings highlight the importance of usability, accessibility and structured support in promoting the effective use of digital educational resources. The study contributes to a better understanding of teachers’ interactions with educational digital resources and provides evidence-based insights for improving their design, dissemination and integration into everyday teaching practice.
|
D. Bele (Algebra Bernays University College, Zagreb, Croatia), I. Znika, K. Bilić (Zagreb University of Applied Sciences, Zagreb, Croatia)
Hybrid Human-AI Framework for Personalized Feedback and Student Learning Enhancement via Exercise Assessment in Programming Education
Effective learning requires timely identification of individual student strengths and weaknesses to support targeted formative feedback. However, human-only assessment is often limited by scalability and consistency, while fully automated systems may lack pedagogical transparency and contextual understanding. This paper presents a hybrid human-AI framework for assessing student exercises that integrate machine learning based analysis with structured human oversight to enhance personalized feedback and learning outcomes. The proposed framework analyzes student generated data, including programming code submissions and short textual explanations, to construct individualized assessment profiles that guide learning progression. Automated techniques based on natural language processing and code analysis provide scalable evaluation, while human evaluators validate results and address cases requiring nuanced pedagogical judgment. The framework is evaluated using a dataset of 1,200 undergraduate programming exercise submissions collected from a Java programming course. A comparative analysis of human-only, AI-only, and hybrid assessment approaches is conducted to examine differences in consistency, fairness, and feedback quality. The results indicate that hybrid assessment outperforms single modality evaluation by improving the reliability and actionable value of feedback provided to students. The findings suggest that structured human-AI collaboration represents an effective approach for personalized assessment in programming education, supporting scalable evaluation while preserving pedagogical integrity. The study is explicitly situated in introductory computer science education and reports that hybrid assessment achieved higher overall scores (0.861 for 30% human review and 0.891 for 50% human review) than human review only (0.643) and AI-only (0.709) assessment, while preserving timely feedback delivery for Java programming exercises. These results demonstrate that ambiguity-driven human-AI collaboration enables scalable, reliable, and pedagogically meaningful assessment in programming education.
|
J. Blanco-Guzmán, C. Ceballos-Hernández, E. de la Torre Monge, M. García Gallo (University of Seville, Sevilla, Spain)
From Mobile Learning to Screen-Free Education? A Bibliometric Analysis of a Post-Pandemic Shift
Over the last decade, mobile learning has evolved from an emerging educational innovation to a main-stream pedagogical practice, particularly accelerated during the COVID-19 pandemic. However, recent years have witnessed a noticeable slowdown in mobile learning research alongside a rapid rise in academic interest in screen time reduction and screen-free educational approaches, especially concerning children and adolescents’ wellbeing. This study investigates whether bibliometric evidence supports the existence of a post-pandemic shift in research agendas from mobile learning toward screen-free and digital wellbeing discourses.
Using a comparative bibliometric analysis of publications indexed in Scopus and Web of Science between 2010 and 2025, two parallel corpora were constructed: mobile learning and screen-free education. Bibliometric indicators related to publication volume, growth rates, citation impact, and temporal trends were analyzed.
Results reveal a clear inflection point around 2021–2022, marked by a decline in mobile learning publications and a sustained acceleration of screen-free research. While this temporal coincidence does not imply causality, it reflects a broader reorientation of educational research priorities toward digital balance and wellbeing. The findings highlight a shift from technology adoption narratives toward more critical and contextualized uses of digital learning technologies.
|
A. Christopoulos (University of Turku, Turku, Finland), S. Mystakidis (University of Patras, Patras, Greece), M. Laakso (University of Turku, Turku, Finland)
The Experiential Quantum Framework: Design Principles for Immersive Virtual Reality in Quantum Mechanics Education
Quantum mechanics presents unique pedagogical challenges: phenomena occur at imperceptible scales, evolve on inaccessible timescales, and contradict classical intuition. Virtual Reality offers promising affordances for addressing such challenges yet existing applications are developed based on technological capability rather than pedagogical rationale. The absence of theoretical guidance risks producing environments that engage learners without supporting actual learning. The aim of the current paper is to propose the Experiential Quantum Framework, a theoretically-grounded design framework for immersive Virtual Reality learning environments in Quantum Mechanics education. The framework is grounded in the observation that Virtual Reality’s distinctive capabilities—scale transformation, temporal manipulation, embodied interaction—align with quantum mechanics’ distinctive pedagogical barriers. The framework identifies four Virtual Reality affordances, maps them to six documented quantum learning challenges, and derives six design principles that translate theoretical rationale into actionable guidance. Each principle follows a standardized template connecting VR capabilities to quantum-specific requirements. Targeted learning outcomes include conceptual understanding, misconception remediation, and transfer. The framework addresses the current theory-practice gap in Virtual Reality-based quantum instruction and generates testable propositions for empirical validation.
|
S. Kocijančič, R. Gabrovšek (Faculty of Education, Ljubljana, Slovenia)
Teaching Smart Devices in Primary School through Simulation and Large Language Models
This paper examines the 2025 Slovenian primary school curriculum revision, focusing on the new "Automation and AI" module. It presents an innovative instructional design that integrates Large Language Models (LLMs) with the SimulIDE simulation environment to teach mechatronics to students aged 12–14. The main contribution is a demonstrated pedagogical framework that employs AI to overcome the "syntax barrier" of text-based programming, shifting the learner’s cognitive load from rote code synthesis to higher-level analysis and evaluation in line with Bloom’s Taxonomy. Through two case studies—an "automatic parking barrier" and a "smart thermometer" pilot study— the research assesses the transition from virtual simulation to physical implementation. Findings show that while AI significantly accelerates the design and coding phases, the "physical friction" experienced during hardware assembly remains essential for developing technical troubleshooting skills. The results confirm that AI-assisted simulation fosters technological agency, enabling students to become informed creators of smart mechatronic systems rather than passive users. This framework provides a scalable model for engineering education in the era of generative artificial intelligence.
|
K. Bedi (Graditeljska škola Čakovec, Čakovec, Croatia)
Artificial Intelligence and Student Writing: A Comparative Case Study
This paper examines the impact of artificial intelligence (AI) tools on students’ writing processes and learning outcomes within the course Media Projects, with the aim of analysing the influence of artificial intelligence on students’ writing processes and creative expression. A comparative study was conducted across two school semesters. During the first semester, students independently produced written texts without AI assistance. In the second semester, students were allowed to use AI tools as supportive aids for idea generation, text structuring, and language refinement. The study analyses differences in text quality, motivation, engagement, and students’ perceptions of AI as an educational tool. The findings indicate that the integration of artificial intelligence into teaching contributes to increased student motivation, higher-quality and more structured texts, and the development of digital and creative competences.
|
B. Adjiovski, D. Capeska Bogatinoska (University of Information Science and Technology “St. Paul the Apostle”, Ohrid, Macedonia), K. Nikolovska (Dyslexia Association Einstein, Skopje, Macedonia), R. Malekian (1)Department of Electrical, Electronic and Computer Engineering, University of Pretoria, 0083, Preto, Yerevan, Armenia)
A Teacher-Centered Mobile Application for Screening Learning Disabilities to Support Inclusive Education
Inclusive education requires accessible tools for identifying learning disabilities, yet educational systems often lack resources for early screening despite these conditions affecting approximately one in five students. This research addresses the critical need for practical screening tools in under-resourced settings through the development and validation of a mobile application designed to empower teachers in supporting students with dyslexia, dyscalculia, Attention Deficit Hyperactivity Disorder (ADHD), and auditory processing disorder. Building on our previous quasi-experimental study demonstrating significant improvement when teachers adapted instruction based on screening results, we conducted expert pedagogical validation with specialists from the Einstein Dyslexia Institute and Educational Therapy Center, followed by preliminary clinical testing with four formally diagnosed students. The expert review process identified essential refinements, including culturally appropriate assessment design, age-differentiated question complexity, and a multi-informant architecture that combines student self-assessment with teacher observation, mirroring professional diagnostic practices. Clinical pilot results provided preliminary confirmation of screening accuracy. The refined application serves dual purposes: identifying students who may benefit from adapted instruction and providing teachers with concrete pedagogical strategies for each learning profile. This work contributes a reproducible framework for developing contextually adapted educational screening tools and demonstrates the effectiveness of expert-driven iterative refinement in educational technology development.
|
M. Slobođanac (Mono Software, Osijek, Croatia), M. Mićunović, A. Papić (Faculty of Humanities and Social Sciences , Osijek, Croatia)
Exploring LIS Students’ Preferences towards Artificial Intelligence Tools
Lately, the use of artificial intelligence (AI) in higher education has become a prominent topic due to the advantages it offers for teaching, learning and administration, as well as the challenges that emerge from its rapid and often uncoordinated integration into educational environment. Focusing on generative AI and tools such as ChatGPT, this paper examines how students in the Department of Information Sciences at the Faculty of Humanities and Social Sciences in Osijek perceive and use AI tools. Using survey method, the research examines students’ familiarity with AI, their perception of AI in relation to academic work, their preferences regarding the use of AI for study-related tasks, and how AI affects their academic performance. The findings indicate that library and information sciences (LIS) students are highly aware of AI technologies and have favorable preferences for using AI in their studies. They recognize the advantages of AI, but also acknowledge its challenges and risks in higher education. The results emphasize the need for training on ethical and responsible use of AI, as well as clear institutional guidelines. Overall, AI has a significant role in higher education, provided it is integrated responsibly and supports, rather than replaces, human abilities, skills and knowledge.
|
Y. Rudenko , T. Stepanova, S. Ahadzhanova , K. Ahadzhanov-Honsales (Sumy National Agrarian University, Sumy, Ukraine), N. Dehtiarova, D. Davidenko (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Applying Orange Data Mining for Exploratory LMS Analytics: A Comparative Case Study of Chinese and Ukrainian Students’ Learning Behaviors
Learning Management Systems (LMS) accumulate substantial amounts of data on student interactions, creating opportunities for exploratory learning analytics and the identification of behavioral trends. This study applies Orange Data Mining as a low-code analytical environment to conduct an exploratory comparative analysis of LMS engagement among Chinese and Ukrainian students. The dataset includes records from 1300 students of Sumy National Agrarian University (700 from China and 600 from Ukraine), enabling the examination of general patterns of interaction within the digital learning environment.
The methodological framework combines descriptive statistics, clustering, and statistical significance testing to assess differences between the two student groups using appropriate tests for independent samples. This approach made it possible to identify distinctions in session frequency, inactivity intervals, number of quiz attempts, and diversity of learning actions within LMS, while indicators such as mobile device usage, digital literacy, and prior academic performance remained similar across groups. The clustering results outlined groups of students with comparable activity profiles, reinforcing the findings regarding differences in the rhythm and intensity of learning engagement.
The study demonstrates the potential of low-code tools for conducting initial LMS analytics and highlights the importance of statistically validating observed tendencies.
|
N. Dehtiarova (Sumy State Pedagogical University named after A.S. Makarenka, Sumy, Ukraine), O. Zhmud, M. Medvedieva, L. Titova (Pavlo Tychyna Uman State Pedagogical University, Uman, Ukraine), O. Viunenko (Sumy National Agrarian University, Sumy, Ukraine), M. Pronikova (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Development of Competence of Future Computer Science Teachers for Working with High-Potential Students
The Olympiad movement varies in content and implementation methods. This activity allows educational institutions to identify talented youth while they are still in school. For the students themselves, it provides an awareness of their abilities, readiness to work under stressful conditions, and opportunities to demonstrate their potential. At the same time, it is increasingly challenging for teachers to identify and motivate such students. Teachers need to understand the specific features of working with students with high intellectual potential. They must also be able to create conditions that foster students’ self-organization and self-discipline.
The aim of this study is to identify the methodological features of preparing future computer science teachers to create conditions for fostering self-organization and self-development in senior school students. The research focused on developing future teachers’ abilities to identify high-potential students, work effectively both individually and with small groups, and create conditions that support student motivation. The study was conducted across three universities, and the results were statistically analyzed. The findings provide a basis for updating methods of developing methodological competence in future computer science teachers, particularly in engaging students in collaboration, interaction within groups, and fostering their skills and readiness to demonstrate initiative and responsibility in a social context.
|
O. Zhmud (Pavlo Tychyna Uman State Pedagogical University, Uman, Ukraine), T. Hryhorenko (National University of Life and Environmental Sciences of Ukraine, Kyiv, Ukraine), O. Pidhornyi , I. Kovtaniuk (Pavlo Tychyna Uman State Pedagogical University, Uman, Ukraine), I. Malyk (Lviv Polytechnic National University, Lviv, Ukraine), N. Dehtiarova , V. Pleskan (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Formation of methodological competence of future teachers in the selection and use of digital learning environments in conditions of saturation with digital tools
In the contemporary context of digitalization and globalization, teachers have access to a wide range of electronic tools. The global network offers digital learning environments, platforms, and resources in great variety. For novice teachers, it is often difficult to make a conscious selection of specific tools and to justify this choice from a methodological perspective. This situation requires the development of methodological reflection and the ability to optimally combine technical teaching tools with students’ active engagement in learning.
Moreover, the saturation of digital tools in students’ learning experience may lead to an opposite effect - loss of interest. The study identifies approaches to the preparation of future teachers aimed at enabling them to select and critically evaluate digital tools, integrate them into lessons without overloading students, and develop readiness to work under conditions of continuous renewal of digital resources.
The research methodology included a test assessing the teacher`s ability to analyze and compare the advantages and disadvantages of specific digital tools, as well as their feasibility for students’ use at home; the ability to work according to an algorithm and to construct an algorithm for defining criteria for evaluating digital applications; analysis of academic achievement and the level of student autonomy; and a questionnaire survey and an experimental study on the effectiveness of the applied methodology.
The results of the study provide grounds to assert that the use of clearly defined criteria for selecting digital tools, combined with students’ active learning activities and the creation of conditions for the development of communication and social skills, is important and contributes to students’ development. This, in turn, supports the formation of teachers’ own instructional methodology and the realization of learners’ potential.
|
| Radovi
Predsjedateljice: Marina Mirković, Kristinka Maček Blažeka
|
M. Šitum (OŠ Granešina, Zagreb, Croatia)
Razvoj algoritamskog razmišljanja u razrednoj nastavi kroz metodu MEMA
Ovaj rad prikazuje primjenu metode MEMA u nižim razredima osnovne škole u okviru projekta MULTISTEM, s ciljem razvoja algoritamskog razmišljanja prije uvođenja rada na računalu. Metoda se temelji na fizičkom modeliranju računalne memorije, pri čemu pretinci predstavljaju varijable, a njihove vrijednosti se prikazuju manipulativnim sredstvima (npr. grah, perlice, kartončići). Time se apstraktni računalni procesi čine vidljivima i razumljivima učenicima. Radionice provedene tijekom prve godine projekta obuhvaćale su aktivnosti: izradu modela, ulaz i izlaz podataka, obradu podataka te završno natjecanje. Učenici su razvijali razumijevanje pojma varijable, slijeda koraka i osnovnih logičkih procesa kroz zadatke zbrajanja, oduzimanja i zamjene vrijednosti. U višim razinama uvedeni su elementi iteriranog zbrajanja i osnovni logički operatori. Napredak učenika praćen je formativno prema unaprijed definiranim kriterijima. Rezultati ukazuju na poboljšanje u matematičkom zaključivanju i razumijevanju osnovnih koncepata programiranja. Uočena je visoka razina suradnje među učenicima, dok su najveći izazovi zabilježeni u fazi obrade i zamjene vrijednosti. Zaključno, metoda MEMA pokazuje potencijal kao učinkovit i pristupačan pristup ranom razvoju algoritamskog razmišljanja.
|
T. Ređep, T. Pavičić (I. Osnovna škola Varaždin, Varaždin, Croatia), B. Marčinković (Osnovna šola Bistrica ob Sotli, Bistrica ob Sotli, Slovenia)
Stavovi i mišljenja učitelja o primjeni generativne umjetne inteligencije u izradi prezentacija
Implementacija alata generativne umjetne inteligencije (GenAI) donosi promjene u raznim aspektima života, pa tako i na području obrazovanja. Jedan od načina primjene GenAI alata je stvaranje prezentacija koje su još uvijek najčešći način putem kojeg učitelji nastoje nastavni sadržaj učiniti pristupačnijim i jasnijim. Budući da se za stvaranje prezentacije troši značajna količina vremena i truda, uz GenAI alate mnogi zadaci povezani sa stvaranjem slajdova mogu se automatizirati, čime se otvaraju nove mogućnosti za bržu pripremu nastave i kreativniju obradu sadržaja. U radu su prikazani rezultati istraživanja o iskustvima i stavovima učitelja vezano uz primjenu GenAI alata u izradi prezentacija. Cilj istraživanja bio je ispitati njihovu razinu upoznatosti i praktična iskustva u primjeni ovih tehnologija, analizirati njihove mogućnosti, prednosti i ograničenja u nastavnoj praksi te procijeniti potencijal GenAI alata za izradu prezentacija u obrazovnom kontekstu. Dobiveni rezultati pružaju uvid u korisnost i praktičnu primjenjivost GenAI alata u svakodnevnom radu učitelja i ukazuju na način na koji ove tehnologije mogu doprinijeti kvaliteti nastavnih materijala.
|
E. Grmić (Tehnička škola Bjelovar, Bjelovar, Croatia), A. Tihomirović (Tehnička škola Zagreb, Zagreb, Croatia)
Izazovi strukovnog kurikula Tehničar za računarstvo
Ovaj rad proučava kurikul za stjecanje kvalifikacije tehničar za računarstvo te predlaže promjene koje olakšavaju vertikalnu prohodnost učenika unutar kvalifikacije. U radu se poseban naglasak stavlja na nedostatak horizontalne korelacije među modulima Digitalna logika, Građa računala i Uvod u programiranje. U modulu Građa računala, za postizanje ishoda učenja „Usporediti vrste procesora i procesorskih sustava na primjerima u simulatoru“ iz skupa ishoda učenja „Arhitektura procesora i sabirnički sustavi“ očekuje se da učenik razlikuje načine pohrane prirodnih, cijelih i realnih brojeva u računalu, koji se proučavaju unutar modula Digitalna logika. Modul Uvod u programiranje kao jedan od ciljeva postavlja stjecanje znanja i vještina u uporabi osnovnih tipova podataka. Navedeni cilj također podrazumijeva potpuno razumijevanje ovog segmenta gradiva, koji učenici proučavaju unutar modula Digitalna logika. Ovaj rad ukazuje na posljedice koje nastaju ako ne postoji suradnja među nastavnicima u aktivnostima koje omogućavaju zajedničko postizanje povezanih ishoda te opisuje nadopunu skupova ishoda učenja postojećih modula pomoću kojih se mogu izbjeći te posljedice, te postići viša razina ishoda učenja s ciljem poboljšanja vertikalne prohodnosti u kvalifikaciji tehničar za računarstvo.
|
R. Soldo (Strojarska tehnička škola Fausta Vrančića, Zagreb, Croatia)
Umjetna inteligencija i kritičko mišljenje u nastavi matematike
Umjetna inteligencija može biti koristan alat u rješavanju matematičkih zadataka, no pritom je ključno da se ne zanemari uloga ljudskog kritičkog razmišljanja. Razvijanje kritičkog mišljenja podrazumijeva poticanje učenika da samostalno analiziraju informacije, logički razmišljaju te donose zaključke utemeljene na provjerenim činjenicama. Tako mogu lakše prepoznati ograničenja i moguće pogreške koje AI ponekad napravi pri rješavanju matematičkih problema. Prikazana metoda analiziranja pogrešaka u rješenjima zadataka koje napravi AI u ovom radu, ima cilj potaknuti korištenje umjetne inteligencije u nastavi matematike, ali kroz raspravu s učenicima potičući njihovo kritičko mišljenje, kako bi uočili prednosti i slabosti korištenja umjetne inteligencije. Ovakav pristup razvija kritičko razmišljanje, omogućuje bolje razumijevanje procesa rješavanja zadataka kroz različite strategije te olakšava shvaćanje apstraktnih matematičkih pojmova. Stoga je tehnologiju vrijedno koristiti, ali uvijek uz kritičko promišljanje kako bismo mogli procijeniti točnost i pouzdanost informacija koje AI pruža.
|
V. Tasić (Mining and Metallurgy Institute Bor, Bor, Serbia), A. Božilov (University of Niš, Faculty of Occupational Safety in Niš, Niš, Serbia), R. Kovačević, B. Radović (Mining and Metallurgy Institute Bor, Bor, Serbia), Z. Damnjanović (Građanska čitaonica Evropa, Bor, Serbia)
Primjena niskobudžetnog IoT sustava za praćenje kvalitete zraka u unutarnjem prostoru
Ovaj rad predstavlja mogućnosti za primjenu niskobudžetnog IoT sustava za praćenje kvalitete zraka u obrazovnim ustanovama, s posebnim naglaskom na unutarnje prostore poput učionica. Predstavljeno je rješenje prijenosnog uređaja za mjerenje koncentracija lebdećih čestica, temperature, relativne vlažnosti zraka i dodatnih parametara (ugljični-dioksid, formaldehid). Poseban naglasak stavljen je na sustav komunikacije i prijenos podataka do korisnika putem Wi-Fi i GSM/SMS modula, što omogućava kontinuirani nadzor kvalitete zraka u realnom vremenu. Prikazana je blok shema povezanih modula te mogućnosti primjene uređaja u IoT okruženjima unutar škola. Prikazani rezultati ukazuju na praktičnu i edukativnu vrijednost implementacije ovakvih sustava u školama.
|
D. Medvedović (OSNOVNA ŠKOLA ROVIŠĆE, ROVIŠĆE, Croatia)
Umjetna inteligencija u učionici: problemsko učenje i primjeri iz prakse
Rad prikazuje niz inovativnih i pedagoški utemeljenih aktivnosti kojima se učenicima od petog do osmog razreda osnovne škole omogućuje razvoj AI pismenosti kroz iskustveno, kreativno, kritičko i problemski orijentirano učenje. Aktivnosti su organizirane kroz korake problemskog učenja — od postavljanja pitanja i ideacije do izrade prototipa, provjere i refleksije. Obuhvaćeni su primjeri iz prakse koji se uklapaju u zasad manji broj objavljenih primjera s osnovnoškolske razine, osobito ondje gdje se, uz tehničke vještine, sustavno razmatraju i osobni te društveni učinci tehnologija. Aktivnosti obuhvaćaju istraživanje utjecaja umjetne inteligencije na svakodnevni život i dobrobit učenika, razlikovanje pozitivnih i negativnih učinaka tehnologija u nastajanju, izradu digitalnih sadržaja temeljenih na umjetnoj inteligenciji te korištenje AI alata za podršku učenju. Učenici pritom eksperimentiraju s načinima na koje AI sustavi rade i usavršavaju vještinu postavljanja učinkovitih upita, kako bi bolje razumjeli rezultate i donosili informirane odluke. Posebna je pozornost usmjerena na odgovornu i sigurnu upotrebu tehnologije, kritičko vrednovanje generiranih sadržaja te jasno razlikovanje vlastitog doprinosa od doprinosa alata. Rad pokazuje kako problemski orijentirani pristupi, u kombinaciji s primjereno odabranim AI alatima, potiču kreativnost, suradnju i digitalnu pismenost te nude poticajne primjere za razvoj dobre prakse u ovom području.
|
M. Mirković (Tehnička škola Požega , Požega, Croatia), D. Možnik (Tehničko veleučilište u Zagrebu, Zagreb, Croatia)
Primjena mjera i standarda kibernetičke sigurnosti u eri umjetne inteligencije: civilni i vojni konteksti
U 21. stoljeću umjetna inteligencija (UI) preoblikuje način na koji društva funkcioniraju - mijenja zdravstvenu skrb, prijevoz, komunikacije i nacionalnu obranu. Uz svoje prednosti, UI uvodi nove rizike za kibernetičku sigurnost i mijenja način primjene postojećih mjera i standarda. Zbog promjena, mjere i standardi kibernetičke sigurnosti moraju se razvijati. Mnogi postojeći standardi kibernetičke sigurnosti primjenjuju se i na UI sustave, no pojavljuju se nove smjernice koje odgovaraju jedinstvenim potrebama UI-a. Vojni sustavi slijede stroge sigurnosne zahtjeve jer često uključuju i klasificirane podatke. U radu su analizirane sličnosti i razlike između civilnog i vojnog pristupa. Oba sektora zahtijevaju snažnu zaštitu podataka, kontrole pristupa i sigurne prakse razvoja. Rad je rađen u suradnji Tehničke škole, Požega i Hrvatskog vojnog učilišta “Dr. Franjo Tuđman”, Zagreb. U Tehničkoj školi u Požegi drugu godinu provodi se s manjom grupom učenika fakultativni predmet “Umjetna inteligencija: od koncepta do primjene”. U anketi su uz ove učenike sudjelovali i drugi učenici škole kao i kadeti Hrvatskog vojnog učilišta “Dr. Franjo Tuđman” u Zagrebu. Učinkovita primjena mjera i standarda kibernetičke sigurnosti ovisi o kombinaciji robusnih tehničkih kontrola, etičkim okvirima i međunarodnoj suradnji. Za učenike koji ulaze u karijeru u tehnici ovo područje pruža priliku i obvezu.
|
K. Brleković, D. Ivanović Ižaković (Elektrotehnička i prometna škola Osijek, Osijek, Croatia)
Etičke implikacije umjetne inteligencije
Integracija kurikuluma „Umjetna inteligencija: od koncepta do primjene“ ključna je za suvremeno obrazovanje, budući da se primjenjuje ne samo u izvanastavnim aktivnostima, već i u okviru različitih predmeta poput informatike, matematike, hrvatskog i engleskog jezika te kemije. Učitelji i učenici često koriste AI alate za rješavanje zadataka, oslanjajući se na gotova rješenja i kodove, što može dovesti do superficialnog razumijevanja materijala. Stoga je imperativno potencirati pitanje etike umjetne inteligencije kao zajedničkog izazova koji se tiče svih sudionika obrazovnog procesa – od nastavnika do učenika.Kroz ključne teme poput strojnog učenja, obrade prirodnog jezika (NLP) i sigurnosti korištenja AI, pružaju se konkretni primjeri kako se ova tehnologija može integrirati u nastavu, obogaćujući obrazovne procese i potičući praktične vještine. Na primjer, učenici razvijaju jednostavne projekte poput igara u Scratchu, istovremeno razmatrajući etičke dileme.Istraživanje provedeno među učenicima pokazuje njihovu zrelu svijest o etičkim implikacijama, naglašavajući potrebu za osviještenim pristupom. Ovaj pristup potiče razvoj kritičkog razmišljanja i etičkih stavova, pripremajući mlade za odgovorno korištenje AI u budućnosti.
|
K. Brleković (Elektrotehnička i prometna škola Osijek, Osijek, Croatia)
Što je M za STEM?
STEM pristup u odgoju potiče djecu na istraživanje, logičko promišljanje, učenje kroz greške i princip „uradi sam“, kombinirajući savladano znanje. Međutim, unatoč radionicama o tehnologiji, ekologiji i opremi, interes slabi u višim razredima osnovne škole jer se fokusira na opremu umjesto pristupa.
Radionice postaju spektakularne, ali bez dubinske analize i realne povezanosti s životom, što stvara dojam irelevantnosti. Tradicionalne metode, poput predavanja bez interakcije, ne potiču argumentiranu raspravu niti toleranciju na greške, već strah od neuspjeha. Nedostatak kvalificiranih STEM nastavnika u Hrvatskoj dodatno otežava integraciju.
Matematika nije samo brojevi, već alat za rješavanje stvarnih problema: mjerenje obujma olovke vodi do razumijevanja geometrije, a testiranje njene čvrstoće potiče statistiku i hipoteze. Bez nje, STEM gubi logičku nit, jer matematika razvija toleranciju na pogreške i "uradi sam" princip, ključne za inovacije.
Možemo li interdisciplinarnim metodama oživjeti STEM interes?
STEM nije oprema, već pristup: zadaci ne moraju biti maštoviti ili zahtjevni, već jednostavni poput pitanja o olovci (materijali, priroda, težina, upijanje vode, obujam). Matematika surađuje s fizikom, kemijom, geografijom i informatikom, potičući logičko razmišljanje i „uradi sam“ principe.
|
L. Ille, M. Varga, R. Fic (Graditeljska škola Čakovec, Čakovec, Croatia)
Od ispitanika do istraživača: Radioemisije kao participativna istraživačka metodologija u srednjoškolskom obrazovanju
U radu je predstavljena inovativna metoda u kojoj radioemisija postaje sredstvo pomoću kojeg se vrši parcipativno istraživanje učenika srednjih škola. Za razliku od tradicionalnih istraživanja u kojima su učenici pasivni subjekti, ovakav pristup podrazumijeva novi istraživački instrument kojeg kreiraju učenici, što ih čini aktivnim istraživačima. Učenici sami osmišljavaju i provode intervjue, a rezultate prezentiraju kroz radioemisije koje potom i produciraju. Ovakvo istraživanje u Graditeljskoj školi Čakovec po prvi put je provedeno tijekom školske godine 2024./2025. kada su učenici proveli anketu o percepciji utjecaja društvenih mreža na zdravlje mladih. Proveli su kvalitativne intervjue sa učenicima, roditeljima i stručnjacima te producirali dvije radioemisije, objavivši ih na platformi YouTube. Učenici su, osim sposobnosti intervjuiranja, pokazali i svoje digitalne kompetencije kroz korištenje digitalnih alata za prikupljanje podataka (Google Forms, Microsoft Forms), uporabu audio tehnologije za montažu (Audacity/Adobe Audition), kao i poznavanje platforme za diseminaciju rezultata. Rezultati ovakvog načina istraživanja pokazuju kako su prednosti za učenike višestruke: veći je angažman učenika, a time i bolji istraživački rezultati, jačaju se njihove digitalne i istraživačke kompetencija te kroz istraživanje razvijaju samorefleksiju o digitalnim navikama. Radioemisije, kao sredstvo prezentacije rezultata istraživanja, doprle su do publike znatno šire od školske zajednice, čime se postigla stvarna društvena relevantnost.
|
A. Tihomirović, S. Šišić (Tehnička škola Zagreb, ZAGREB, Croatia)
Prozor u svijet rješavanja problema
Primijenjena matematika je izborni predmet na trećoj i četvrtoj godini obrazovanja za smjer tehničar za računalstvo. Učenici su prethodno u okviru svojega obrazovanja savladali predmet Algoritmi i programiranje, te se pretpostavlja da solidno barataju barem jednim programskim jezikom, te da imaju razvijene vještine logičkog zaključivanja, analize problema i sustavan pristup njegovom rješavanju. U ovom radu bit će analizirani radovi učenika koji obuhvaćaju različite matematičke alate kojima su primjenom programiranja rješavani različiti problemi iz elektrotehnike, optimizacije složenijih logičkih sklopova, kao i situacije kojima smo svakodnevno okruženi. Rad pokazuje horizontalnu i vertikalnu korelaciju različitih predmeta koje savladavaju učenici na svom putu do svjedodžbe tehničara za računalstvo. Poseban naglasak rad stavlja na neuspješne radove i analizira razloge zbog kojih su isti nastali. Rad opisuje korake u razvoju učenika k inženjerskom načinu razmišljanja, te kao takav može poslužiti za razvoj specijaliziranih modula u reformi strukovnog obrazovanja.
|
M. Dergez, D. Fraj, M. Lučan (Tehničko veleučilište u Zagrebu, Zagreb, Croatia)
Pojednostavljena interpretacija procesnih računala realizirana mikroupravljačem
Cilj ovoga rada je bio osmisliti i testirati koncept svestranog sustava za sklapanje automatiziranih sustava spremnih za rad u industrijskome okruženju korištenjem mikroupravljača. Pomoću programa Circuitmaker projektirane su četiri tiskane pločice koje su dio dvaju različitih modula. Moduli su izrađeni te je obavljeno ispitivanje pojedinih dijelova tiskanih pločica kako bi se ustanovila njihova ispravnost korištenjem mjernih instrumenata i prateći stanja varijabli mikroupravljača u programu STM32CubeIDE. Svestranost ovog sustava proizlazi iz činjenice da je osmišljen na način da se pomoću njega mogu sastaviti raznorazni sustavi, tj., sustav nije napravljen za neku specifičnu namjenu, niti je programski ograničen, već je moguća potpuna kontrola nad svakim dijelom. Sustav je zamišljen da radi sa više jedinica koje mogu zasebno raditi ako je potrebno te da ih je lako zamijeniti sa jedinicama iste namjene ili potencijalno s učinkovitijim ili kvalitetnijim komponentama.
|
N. Mesaroš Grgurić (OŠ CENTAR , Rijeka, Croatia)
Učionica koja spaja jezike, igru i tehnologiju
U kontekstu sve veće jezične i kulturne raznolikosti u razrednim odjelima, učitelji su suočeni s izazovom kako istovremeno osigurati uključivanje svih učenika, razvijati njihove jezične kompetencije i primjenjivati suvremene tehnologije na smislen i siguran način. Ovo predavanje i prikaz dobre prakse nudi konkretan odgovor kroz integraciju igre, tehnologije i diferenciranog pristupa učenju.
Kroz primjere iz neposredne nastavne prakse bit će prikazano kako korištenjem edukativnih robota poput Bee-Bota, digitalnih alata i elemenata umjetne inteligencije poticati razvoj vokabulara, komunikacijskih vještina i logičkog mišljenja kod učenika. Aktivnosti su osmišljene tako da omogućuju istovremeno sudjelovanje inojezičnih učenika, darovitih učenika i učenika kojima je potrebna dodatna podrška, čime se ostvaruje individualiziran i uključiv pristup učenju.
Sudionici će imati priliku sudjelovati u strukturiranim, interaktivnim aktivnostima koje demonstriraju kako igra i tehnologija mogu postati snažan alat za razvoj digitalne pismenosti, kritičkog mišljenja i suradnje među učenicima. Poseban naglasak stavlja se na odgovorno i svrhovito korištenje digitalnih tehnologija u sigurnom okruženju.
Radionica osnažuje učitelje za primjenu inovativnih i inkluzivnih metoda poučavanja te nudi konkretne strategije koje je moguće odmah implementirati u svakodnevni rad u razredu.
|
K. Maček Blažeka (Tehnička škola Ruđera Boškovića, Zagreb, Croatia)
Poteškoće početnika u oblikovanju algoritama pri rješavanju jednostavnog problemskog zadatka iz programiranja
U radu se analizira način na koji učenici početnici oblikuju algoritamska rješenja pri rješavanju jednostavnog problemskog zadatka iz programiranja, s posebnim naglaskom na vrste, učestalost i obrasce pogrešaka. Istraživanje je provedeno na uzorku od 45 učeničkih rješenja zadatka koji zahtijeva primjenu strukture grananja if–else u programskom jeziku C++. Pogreške su klasificirane u šest kategorija: konceptualne, interpretacijske, matematičke, algoritamske, logičke i sintaktičke. Rezultati pokazuju da se pogreške rijetko pojavljuju izolirano te da znatan dio netočnosti proizlazi iz nerazumijevanja teksta zadatka i strukture algoritma, a ne isključivo iz nepoznavanja sintakse programskog jezika. Analizom su identificirani tipični obrasci algoritamskog razmišljanja učenika, a na temelju rezultata iznesene su metodičke implikacije relevantne za poučavanje programiranja početnika.
|
S. Bunjevac Nikodem (OŠ Dragutina Tadijanovića, Zagreb, Croatia)
Korištenje interneta i stavovi o učenju matematike kod zagrebačkih osnovnoškolaca
U radu se analizira povezanost korištenja interneta i stavova o učenju matematike. Rad se temelji na analizi rezultata empirijskog istraživanja o korištenju interneta, stavovima o učenju i školskom uspjehu provedenog od listopada do prosinca 2025. godine u četiri osnovne škole u Zagrebu. Istraživanje je provedeno anketom (grupno anketiranje) na učenicima viših razreda (N=457). Korištenje interneta analizira se pomoću sljedećih pokazatelja: prosječno svakodnevno vrijeme provedeno na internetu tijekom radnog tjedna i tijekom vikenda, prosječno svakodnevno vrijeme provedeno na mobitelu i različiti načini korištenja interneta u edukativne svrhe u školi. Stavovi o učenju matematike analiziraju se pomoću jednog instrumenta koji sadrži devet čestica. U radu se polazi od opće hipoteze da korištenje interneta predstavlja važnu aktivnost u svakodnevnom životu učenika. Temeljem dosadašnjih spoznaja i rezultata prethodno provedenih istraživanja, kako nacionalnih tako i međunarodnih, u radu se postavljaju i tri radne hipoteze: 1. Češće korištenje interneta tijekom radnog tjedna i tijekom vikenda negativno je povezano sa sklonošću prema učenju matematike; 2. Češće korištenje mobitela negativno je povezano sa sklonošću prema učenju matematike; 3. Češće korištenje interneta u edukativne svrhe u školi negativno je povezano sa sklonošću prema učenju matematike. Rezultati na osnovnoj razini pokazuju da učenici značajno više koriste internet tijekom vikenda nego tijekom radnog tjedna, a oko trećine ih svakodnevno koristi mobitel više od tri sata. Internet se u edukativne svrhe u školi najčešće koristi za provjeru informacija na školskim internetskim stranicama ili platformama za učenje, te za čitanje i pisanje. Stavovi o učenju matematike su uglavnom podijeljeni, npr. oko 30% ispitanih učenika navodi da voli matematiku, a oko 40% ih smatra da je matematika dosadna. Rezultati daljnje analize potvrđuju postavljene hipoteze.
|
| Radovi
Predsjedatelji: Snježana Babić, Diego Tich, Aleksandra Tonković
|
D. Čakija, E. Ivanjko, M. Gregurić (Fakultet prometnih znanosti, Zagreb, Croatia), N. Hlupić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
Izazovi i pristupi poučavanju programiranja studenata bez prethodnog iskustva u programiranju
Poučavanje programiranja studenata koji se s programiranjem susreću po prvi put predstavlja značajan nastavni izazov. Studenti dolaze na fakultet iz različitih srednjih škola i predznanja, često bez potrebnog informatičkog znanja, pa se nemali broj na prvoj godini prijediplomskog studija s programiranjem susreće prvi put. Ova različitost predznanja izravno utječe na mogućnost praćenja nastave te nameće potrebu pomnog osmišljavanja i prilagodljivog provođenja nastavnih procesa. U ovom radu prikazuju se izazovi u poučavanju programiranja studentima na Sveučilištu u Zagrebu, Fakultetu prometnih znanosti, s posebnim naglaskom na teškoće studenata u razumijevanju postavljenih problema i algoritamskom razmišljanju koje bi dovelo do idejnog rješenja postavljenog problema i omogućilo implementaciju rješenja u odabranom programskom jeziku. Rad donosi analizu iskustava iz višegodišnje nastavne prakse i razmatra učinkovitost izloženih metoda te se potom predlažu smjernice za unaprjeđenje nastave programiranja s ciljem ubrzanja i olakšavanja usvajanja gradiva studentima bez prethodnog iskustva u programiranju.
|
D. Tich, A. Tonković (OŠ Gornja Vežica Rijeka , Rijeka , Croatia)
Multidisciplinarni pristup u radu s učenicima prvih razreda OŠ Gornja Vežica u Rijeci
Mema je igračka pomoću koje mala djeca mogu naučiti programirati bez upotrebe računala. Služi za razvoj kreativnog razmišljanja i logike programiranja kod djece mlađeg uzrasta. Provedene su radionice u Osnovnoj školi Gornja Vežica u Rijeci s učenicima prvih razreda na uzorku od 50 učenika i učenica podijeljenim u 3 heterogene grupe primjenom metode Mema ispitali smo uspješnost usvajanja novih znanja iz programiranja primjenom ranije usvojenih odgojno-obrazovnih ishoda, poznatih operacija zbrajanja i oduzimanja prirodnih brojeva poznatim iz nastavnog predmeta matematike. Među zainteresiranim učenicima bilo je više učenica nego učenika. Multidisciplinarni pristup u podučavanju i popularizaciji STEM-a je ključan jer učenicima omogućuje razvijanje šireg spektra vještina već od najmlađe dobi, kao što su kritičko razmišljanje, timski rad i razvijanje komunikacijskih sposobnosti, što ih kasnije čini sposobnijima za ulazak u izazove koje od njih traži svakodnevnica. Multidisciplinarnost kroz STEM područja predstavlja integraciju različitih disciplina - matematike, umijeća programiranja i informatike. Pomoću provedenih upitnika i kvizova znanja došli smo do pozitivnih rezultata u upotrebi ove metode.
|
L. Rogina, D. Fraj, A. Penđer, M. Lučan (Tehničko veleučilište u Zagrebu, Zagreb, Croatia)
Elektronička društvena igra „Trčeće svjetlo“
U članku je prikazana izrada prototipa elektroničke društvene igre „Trčeće svjetlo“ u okviru završnog rada na prijediplomskom stručnom studiju Mehatronike. Cilj igre je pritiskom na tipkalo zaustaviti trčeće svjetlo na ciljanoj lampici. Potrebno je koristiti vizualnu i motoričku koordinaciju u ostvarivanju što većeg broja bodova. U radu je opisan postupak odabira elektroničkih komponenti, projektiranje tiskane pločice u programu CircuitMaker kao i lemljenje iste. Kućište igre dizajnirano je u programu Onshape te ispisano uz pomoć 3D pisača. Objašnjena je logika igre kao i način rada programskog koda napisanog u Arduino integriranom razvojnom okruženju.
|
D. Kurtić (Agencija za odgoj i obrazovanje, Zagreb, Croatia)
Pristupi poučavanju i učenju u nastavi glazbene kulture i glazbene umjetnosti: video lekcija u nastavi na daljinu
Pojavom pandemije Covid-19 u Republici Hrvatskoj pristupi poučavanja i učenja u školama se mijenjaju. Zbog dobrobiti učenika i redovitog održavanja nastave, prema naputku Ministarstva znanosti i obrazovanja izrađuju se video lekcije koje služe kao digitalno metodičko nastavno sredstvo u nastavi na daljinu. Izrada video lekcije osim stručno-metodičkih nastavnih kompetencija od učitelja i nastavnika zahtjeva visoku razinu digitalne i medijske pismenosti. Značenje video lekcije kroz povijest se mijenja. Video lekcija postaje digitalni medij kojim se omogućuje interakcija između učenika i sadržaja učenja koju kontrolira i kreira učitelj/nastavnik u svrhu obrade nastavnih sadržaja i usvajanja ishoda sukladno kurikulu nastavnog predmeta. U ovom radu prikazat će se postupak izrade video lekcije za predmet Glazbena kultura i Glazbena umjetnost te namjena njezinih sastavnica u svrhu poučavanja i učenja na daljinu.
|
D. Kurtić (Agencija za odgoj i obrazovanje, Zagreb, Croatia)
Stavovi i mišljenja učitelja glazbene kulture, nastavnika glazbene umjetnosti i radne skupine „i-nastava“ o zadovoljstvu korištenja video lekcija u nastavi
Video lekcija upotrebljava se kao stručno-metodičko nastavno sredstvo za poučavanje i učenje u vrijeme pandemije korona virusa i izvođenja nastave na daljinu u svim školama u Republici Hrvatskoj. Izradu video lekcija početkom pandemije korona virusa, za sve škole na nacionalnoj razini isključivo pripremaju članovi radne skupine „i-nastava“. Članovi radne skupine „i-nastava“ su učitelji i nastavnici napredovani u zvanje i s vrlo dobro razvijenim stručno-metodičkim i digitalnim kompetencijama. Izabrani su u radnu skupinu sukladno prijedlogu nadležnog Ministarstva i Agencije za odgoj i obrazovanje. Video lekcija se može koristiti i u klasičnoj nastavi (uživo) u cijelosti ili dijelovima prema potrebi i procjeni učitelja/nastavnika. U ovom radu prikazat će se stavovi i mišljenja učitelja glazbene kulture, nastavnika glazbene umjetnosti i radne skupine „i-nastava“ o zadovoljstvu korištenja video lekcija u nastavi. Podatci su dobiveni ispunjavanjem upitnika učitelja glazbene kulture, nastavnika glazbene umjetnosti i članova radne skupine “i-nastava“ tijekom 2023. godine.
|
T. Adamović, I. Sekovanić, A. Petrović (Veleučilište u Bjelovaru, Bjelovar, Croatia)
Sustav za kontinuirano praćenje i automatsko testiranje studentskog rada u Oracle okruženju
U ovom radu predstavlja se arhitektura i inicijalni rezultati sustava za kontinuirani nadzor studentskog rada na predmetu Relacijske baze podataka na Veleučilištu u Bjelovaru.Studenti tijekom cijelog semestra rade na jednoj centralnoj Oracle bazi podataka, što u kombinaciji s Oracleovim sistemskim tablicama i okidačima omogućuje nadzor i bilježenje svih relevantnih aktivnosti te centralizirano upravljanje sustavom.
Primjenom pohranjenih Oracle procedura i jobova razvijen je asinkroni sustav za testiranje koji automatizirano pokreće i izvršava testove nad studentskim objektima u realnom vremenu.
Svaka značajna akcija studenta tijekom laboratorijskih vježbi bilježi se u dnevnik aktivnosti i, prema unaprijed definiranim pravilima, može pokrenuti jedan ili više testova.
Testovi su skalabilni, deklarativno opisani i povezani s ishodima učenja te čine osnovu za kvantitativno praćenje kvalitete studentskog rada.
U radu se raspravljaju prva iskustva primjene sustava te prikazuju načini kako se iz prikupljenih podataka mogu definirati konkretne metrike za praćenje napretka i sposobnosti studenata na kolegijima iz programiranja.
|
S. Kakuk Fridl (Filozofski fakultet Osijek, Osijek, Croatia)
Pedagoški potencijali i rizici generativne umjetne inteligencije u visokom obrazovanju
Brzi razvoj generativne umjetne inteligencije (GenAI), osobito alata temeljenih na velikim jezičnim modelima, sve snažnije utječe na nastavne i obrazovne prakse u visokom obrazovanju. Iako takvi alati nude nove mogućnosti za podršku učenju, istodobno otvaraju i niz pedagoških pitanja, među kojima se posebno ističu površno učenje, kognitivna pasivnost te pouzdanost generiranih informacija. Cilj ovoga rada jest analizirati pedagoške potencijale i rizike generativne umjetne inteligencije u visokom obrazovanju te razmotriti ulogu nastavnika u pedagoškom usmjeravanju njezine primjene. Rad se temelji na narativnom pregledu recentne znanstvene literature objavljene u razdoblju od 2020. do 2025. godine, s naglaskom na obrazovnu primjenu generativne umjetne inteligencije. Analiza je utemeljena na ključnim pedagoškim perspektivama, uključujući konstruktivizam, samoregulirano učenje i metakogniciju. Na temelju pregleda literature identificiraju se glavne pedagoške funkcije GenAI, ali i ključni rizici povezani s njezinom neprimjerenom uporabom u nastavnoj praksi. Zaključno se predlaže konceptualni okvir koji naglašava potrebu za odgovornom i pedagoški vođenom implementacijom generativne umjetne inteligencije u visoko obrazovanje.
|
D. Peras, Z. Stapić, D. Ljubas (Sveučilište u Zagrebu, Fakultet organizacije i informatike, Varaždin, Croatia)
Primjena alata temeljenih na umjetnoj inteligenciji kao podrške nastavnicima u analizi tehničkog duga studentskih softverskih projekata
Evaluacija studentskih projekata vremenski je zahtjevna i često subjektivna aktivnost, što predstavlja izazov, posebice kada kolegij sluša velik broj studenata. Alati temeljeni na umjetnoj inteligenciji nude mogućnost automatiziranog uvida u različite dimenzije kvalitete softvera, uključujući i informacije koje proizlaze iz analize tehničkog duga.
Rad istražuje kako se alati temeljeni na umjetnoj inteligenciji mogu koristiti kao podrška nastavnicima u analizi tehničkog duga studentskih softverskih projekata. Analiza tehničkog duga se pritom ne promatra kao krajnji cilj, već kao izvor uvida o jasnoći, strukturi, održivosti i kvaliteti koda. U istraživanju se koriste tri različita pristupa: metrički pristup (tehnike strojnog učenja), analitički pristup (alat za analizu tehničkog duga) te jezično-kognitivni pristup (veliki jezični model). Rad istražuje potencijal različitih tipova alata temeljenih na umjetnoj inteligenciji da olakšaju nastavnicima evaluaciju, smanje opterećenje ručnog pregleda koda te pruže vjerodostojnu povratnu informaciju o studentskim projektima.
Doprinos rada je identifikacija načina na koji navedeni alati mogu podržati nastavnike u evaluaciji studentskih softverskih projekata, te analiza njihove korisnosti u nastavi softverskog inženjerstva. Rezultati mogu služiti kao smjernice za unapređenje evaluacijskih praksi u nastavi softverskog inženjerstva te potaknuti daljnju integraciju umjetne inteligencije u obrazovne procese.
|
D. Levstek (KONČAR - Električna vozila d.d., Zagreb, Croatia), D. Fraj, M. Lučan (Tehničko veleučilište u Zagrebu, Zagreb, Croatia)
Probnica za ispitivanje hidro uređaja za kočenje na tramvajima serije TMK2200
U ovom radu opisana je izrada probnice za ispitivanje hidro uređaja za kočenje na tramvajima serije TMK2200. Opisan je rad hidro uređaja te postupak njegovog ispitivanja. Sukladno tome projektirano je tehničko rješenje tj. tiskana pločica koja služi kao probnica za ispitivanje navedenih hidro uređaja. Tiskana pločica bazirana je na STM32 mikroupravljaču te je opisan odabir komponenti za izradu probnice kao i samo ispitivanje rada iste. Također, dodana je i druga tiskana pločica temeljena na ESP32 mikroupravljaču, koja omogućuje unos serijskog broja ispitivanog uređaja putem tipkovnice te automatsko slanje rezultata ispitivanja na Flask server. Podaci se spremaju u bazu podataka te su dostupni putem web sučelja, gdje korisnici mogu pregledavati statistiku, analizirati kvarove, pretraživati ispitane uređaje prema serijskom broju i brisati pojedinačne zapise.
|
U. Ramadani, I. Lazović (Institut za nuklearne nauke Vinča, Institut od nacionalnog značaja za Republiku Srbiju, Univerzitet , Belgrade, Serbia), S. Radojević (Mašinski fakultet, Univerzitet u Beogradu, Bor, Serbia), M. Živković, D. Radivojević, R. Jovanović (Institut za nuklearne nauke Vinča, Institut od nacionalnog značaja za Republiku Srbiju, Univerzitet , Belgrade, Serbia), V. Tasić (Institut za rudarstvo i metalurgiju Bor, Bor, Serbia)
Softverski sustav za prikupljanje, pohranu i vizualizaciju podataka iz mreža senzora kvalitete zraka
Ovaj rad predstavlja ILAQ-NET, softversku platformu za integrirano upravljanje mrežama senzora za mjerenje kvalitete zraka na mikro i makrolokacijskoj razini. Poseban naglasak stavljen je na pouzdano prikupljanje, obradu i vizualizaciju mjernih podataka te na nadzor rada sustava u stvarnom vremenu. Platforma je razvijena za rad s niskobudžetnim senzorima kvalitete zraka i omogućuje dugoročno prikupljanje, pohranu i analizu vremenski ovisnih rezultata mjerenja. Arhitektura sustava temelji se na kombinaciji HTTP i MQTT komunikacije s TLS enkripcijom, čime se osigurava siguran prijenos podataka između senzorskih čvorova, serverskog dijela sustava i korisničkog sučelja. Serverska logika implementirana je korištenjem FastAPI okvira. Za pohranu vremenskih serija mjernih podataka koristi se InfluxDB, dok se metapodaci, konfiguracije uređaja i korisničke informacije pohranjuju u PostgreSQL bazu podataka. Korisničko web-sučelje razvijeno je korištenjem Vue 3 frameworka te omogućuje pregled mjerenja u stvarnom vremenu, analizu povijesnih podataka, upravljanje mrežama i uređajima te slanje konfiguracijskih i upravljačkih naredbi. Sustav je projektiran modularno, s jasno razdvojenim funkcionalnim cjelinama za prikupljanje podataka, njihovu pohranu i vizualizaciju, čime se osiguravaju skalabilnost i prilagodljivost različitim scenarijima primjene. U radu je prikazan primjer primjene platforme u kućanstvima i školskim prostorima, gdje ILAQ-NET omogućuje kontinuirano praćenje unutarnje kvalitete zraka i osnovni nadzor stanja sustava.
|
| Radovi
Predsjedatelji: Dana Palova, Martin Vejačka
|
G. Bubaš (University of Zagreb Faculty of Organization and Informatics, Varaždin, Croatia)
Large Language Models in Teaching About Decision‑Making Errors: Potential Applications in Academic Courses on Health Communication, Political Science, and History
LLMs like ChatGPT have been frequently used in medicine for assistance in diagnosis and decision making, as well as in media and politics for detecting misinformation and fake news in information sources. In cognitive psychology, they have been utilized for analyses of cognitive distortions. In general, LLMs can assist human decision making processes. However, there is limited research on the use of LLMs in teaching within the specific field of decision making errors. In our study, a comprehensive list of decision making errors was applied in combination with ChatGPT and Gemini LLMs, using the Deep Research option to identify decision making errors related to three fields of academic education: health communication, political science, and history. Illustrative case studies are presented, including examples from health communication and health promotion projects in the U.S., political decision making in the U.S. in 2025, and World War II history in Europe. The results of these case studies—using LLMs to search for, analyze, and illustrate decision making errors—are interpreted in the context of student education in university courses. Finally, a review of recent literature is provided on similar uses of LLMs in business and management, legal, medical, and engineering education.
|
S. Babić (Juraj Dobrila University of Pula, Faculty of informatics, Pula, Croatia), M. Čičin Šain (Cybernetics society Rijeka, Rijeka, Croatia)
The MEMA Method for Early Programming Education Supported by Socratic Dialogue and Generative Artificial Intelligence
This paper examines the use of the MEMA method and its associated didactic toy, combined with the Socratic method supported by generative artificial intelligence, in early programming education. The MEMA method incorporates guided learning by gradually introducing students to basic programming concepts, while generative artificial intelligence provides personalized questions, examples, and feedback. This approach can assist teachers in training to implement the MEMA method and help lower elementary school students develop computational thinking.
|
R. Kubík, P. Voštinár (Matej Bel University in Banská Bystrica, Banská Bystrica, Slovakia)
Vibe Programming at Primary School
In recent decades, the educational environment has been transformed under the influence of the expansion of modern digital technologies, especially generative artificial intelligence. In this paper, we focus on teaching programming using the so called vibe programming approach. We describe our experience with teaching within the subject of informatics, in which primary school pupils, after watching an instructional video, created a web application for obtaining current weather information.
The web application was created using generative artificial intelligence without prior programming knowledge. We taught 140 pupils in this way and collected their feedback through a questionnaire. Based on the questionnaire responses and interviews with the pupils, we can conclude that they enjoyed this form of instruction and would like to have further similar tasks that would help them in real life or in other subjects to create mobile and web applications without having to possess programming knowledge.
|
A. Kovačić, G. Bubaš (University of Zagreb Faculty of Organization and Informatics, Varaždin, Croatia)
Student Insights on the Use of GenAI for the Development of Interpersonal Communication Skills in English as a Foreign Language: A Preliminary Study
Computer-assisted language learning (CALL) is an educational domain in which the impact of Generative Artificial Intelligence (GenAI) has been particularly notable. In CALL literature, diverse applications of GenAI and chatbots have been reported for developing linguistic communicative competence in English as a Second Language (L2). However, the specific area of interpersonal communication skills in the context of teaching L2 and English as a Foreign Language (EFL) remains underresearched. This paper presents examples of student evaluations of the usage of a GenAI tool to promote interpersonal skills in L2 from both practical and theoretical perspectives, based on a small-scale qualitative study. The focus was to examine the potential of GenAI tools such as ChatGPT for enhancing selected interpersonal skills (composure, self-disclosure, conversation skills, and verbal expressivity) in L2/EFL learning within the knowledge–motivation–skills theoretical framework. During the study, participants were required to test pre-formulated prompts and critically evaluate the responses provided by ChatGPT. A qualitative analysis of participants’ assessments is presented and interpreted.
|
M. Pokorný (Trnava University, Trnava, Slovakia)
Comparison of Online Testing and Onsite Testing Results in Combinatorics and Working with Data
The COVID-19 pandemic has led to a boom in online testing. This has also raised the issue of cheating on online tests. Teachers have been trying to find various ways to eliminate the possibility of cheating on online tests. Despite several methods, the use of online testing has decreased significantly since the pandemic. In the paper, we compare the results of students from the Combinatorics and Working with Data course at the Faculty of Education of Trnava University, examining whether students who were tested online achieved better results than students who were tested onsite. We also focus on reasons why students chose online testing or onsite testing.
|
J. Štofa (Faculty of Economics, Technical University of Košice, Košice, Slovakia)
Gender Differences in Solving Analytical and Practical Tasks in an Introductory Computer Science Course
The paper analyzes gender differences in performance on analytical and practical tasks in an introductory computer science course at a faculty of economics. The analysis is based on the results of a midterm written exam in the course Informatics I completed by first-year students. The exam consisted of analytically oriented tasks in the R programming language and practically oriented tasks focused on work with the MS Excel spreadsheet processor. The tasks targeted algorithmic thinking, data processing, and the application of learned procedures. The results were analyzed separately for male (M) and female (W) students. Overall score performance as well as performance in individual task types were compared. The findings indicate differences dependent on task characteristics. Analytical tasks exhibit higher performance variability than practical tasks. In practical tasks, the results were more balanced. The results provide empirical evidence for discussion of gender-related aspects in computer science education and for optimization of the structure and assessment of introductory computer science courses.
|
D. Paľová, J. Zoričaková, M. Vejačka (Technical University of Košice, Košice, Slovakia)
Utilisation of Artificial Intelligence Tools in Teaching Entrepreneurial Skills: Experiences from a Pilot Course
Rapid technological development and changing labour market demands require higher education institutions to adopt innovative pedagogical approaches that support the development of entrepreneurial and digital competencies. One of the emerging responses to these challenges is the integration of artificial intelligence (AI) tools into teaching and learning processes, alongside the growing use of micro-credentials in higher education. This paper presents experiences from a pilot university course, “Introduction to the use of artificial intelligence and applications in business”. The course was designed using a competency-based, student-centred pedagogical framework, emphasising active learning, project-based tasks, and design thinking principles. AI tools were incorporated as supportive instruments for idea generation, analysis and problem-solving in entrepreneurial contexts. The paper analyses the course design, teaching methodology and preliminary student feedback collected through questionnaires and qualitative evaluation. The findings indicate that the use of AI tools positively influenced student motivation, engagement and perceived development of entrepreneurial competencies. The results suggest that AIsupported teaching, when appropriately embedded within a pedagogical framework, can enhance learning outcomes and represent a promising approach to implementing microcredentials in higher education.
|
M. Vejačka, D. Paľová, J. Zoričaková (Faculty of Economics, Technical university of Košice, Košice, Slovakia)
Building Data Analysis Competencies in Economics Education: A Structured Learning Path with Industry Collaboration
The ability to understand, analyze and interpret data has become a key competence for graduates of economics-related study programmes. However, traditional curricula often fail to develop these skills systematically. This paper presents the design and implementation of a structured Data Analysis Learning Path aimed at students of Finance, Banking and Investment, as well as Public Administration and Regional Management. The proposed learning path consists of a sequence of compulsory-elective courses focused on the gradual development of data literacy and analytical skills, from basic data handling to advanced analysis and reporting. A significant component of the approach is the integration of living labs, where students work with real datasets and assignments provided by industry partners, enhancing their practical experience and alignment with labour market needs. The paper further presents a case study of the course Data Analysis and Reporting, which has been implemented for three consecutive years. The evaluation is based on grading results of semester project outcomes and course itself. The results indicate improved analytical skills, higher student engagement, and increased perceived relevance of data analysis competencies for future careers. The paper presents main findings from the perspective of support the development of data analysis competencies in economics education and contribute to smoother transitions from academia to professional practice.
|
J. Zoričaková, D. Paľová, M. Vejačka (Technical University of Košice, Košice, Slovakia)
A Comparative Analysis of Human and AI-Based Assessment of Team-Based Business Simulation Projects
The increasing role of digital technologies in entrepreneurship education raises important questions regarding the assessment of complex, project-based learning outcomes. This paper investigates the potential of artificial intelligence (AI) as a supportive assessment tool for economic and entrepreneurial student projects, with a focus on future integration into micro-credential courses aimed at developing entrepreneurial competencies. The study compares traditional teacher-led evaluation with AI-assisted assessment of student business projects in which learners simulate a small company and design a digital transformation of its business processes, combining technical, business, and entrepreneurial competencies. The projects are assessed using predefined evaluation criteria by the course teacher and independently evaluated by an AI-based large language model following the same assessment rubric. Quantitative differences in scores and qualitative differences in feedback are analysed to identify similarities and discrepancies between human and AI assessment approaches. The findings provide empirical insights into the applicability and limitations of AI-supported assessment for entrepreneurship-oriented projects. The paper discusses pedagogical implications in the context of micro-credentials and European digital credits, emphasizing the role of AI as a supportive tool for enhancing transparency and consistency in the assessment of team-based projects.
|
J. Guniš, Ľ. Šnajder, D. Kotlárová, K. Brinziková (P.J. Safarik University in Kosice, Kosice, Slovakia)
Use of Large Language Models in the Evaluation of Open-Ended Questionnaire Responses: A Case Study
In 2025, we conducted a questionnaire survey focused on the perception of artificial intelligence. The target group consisted of students and teachers from the Czech Republic and Slovakia. More than 3,800 respondents participated in the survey. The questionnaire included closed-ended questions and five open-ended questions.
Manual evaluation of responses to open-ended questions would be time-consuming. We therefore explored the possibilities of using large language models (LLMs) for automated analysis of responses. We tested freely available tools Notebook LM (Google), ChatGPT (OpenAI), and Gemini (Google), and subsequently the paid version ChatGPT Plus.
In this paper, we present our experiences and findings from evaluating respondents’ open-ended answers using LLMs.
|
| Radovi
Predsjedatelji: Hannu Jaakkola, Jaak Henno
|
G. Hajdin, D. Plantak Vukovac (University of Zagreb Faculty of Organisation and Informatics, Varaždin, Croatia), T. Milec (Electrical Engineering School Varaždin, Varaždin, Croatia)
Gamification in Secondary Vocational Education: A Case Study of Student and Teacher Perspectives
This paper explores the implementation and perception of gamification within the secondary vocational education system in Croatia. As digital natives enter the final stages of their secondary education, the disparity between traditional teaching methods and student engagement strategies becomes increasingly evident. The study employs a quantitative case study methodology focused on a vocational school in Varaždin County, analyzing data collected from 90 final-year students and 27 teachers through structured questionnaires. The research contrasts the theoretical benefits of gamification, such as increased motivation and cognitive engagement, against the practical barriers reported by teachers, specifically time constraints and curriculum rigidity. While previous research often generalizes gamification effects, this study provides specific insights into the "mechanic mismatch" between the simple tools teachers employ (primarily quizzes like Kahoot) and the complex game elements students desire (narratives and social relatedness). Contrary to initial expectations, the study finds no significant difference in readiness between STEM and non-STEM teachers, suggesting a homogenized digital culture within the institution. The findings suggest that while the acceptance of gamification is moderate to high among both groups, successful integration requires a shift from sporadic game-based activities to structural gamification embedded in the curriculum, supported by Self-Determination Theory principles.
|
E. Kalaitzopoulou, P. Matthews (University of the West of England, Bristol, United Kingdom), A. Christopoulos (University of Turku-Turku Research Institute for Learning Analytics Docent, Faculty of Science, Turku, Finland)
Exploring the impact of marked formative assessment tasks on student performance
Formative feedback is considered important in Higher Education and it has the potential to improve student work and enhance understanding. However, little is known about how students engage with formative feedback once it is returned and its impact on assessment outcomes. The purpose of this study was to examine the impact of optional formative assessment feedback on learners’ performance and behaviour. In total 138 Postgraduate students participated in the study. Students’ profiles were created by an enrolment questionnaire covering the “Big 5” personality dimensions, the Situational Motivation Scale and demographic and background data. Learners’ engagement with the LMS (Blackboard) was monitored utilising built in reporting tools for activity in main content areas over the module’s ten weeks of teaching. Marks from both the formative and the summative assessments were utilised to complement the development of the student profiles and to gauge the effect that such practices had on their engagement and performance. Results indicated that access and time spent on feedback was significant and helped students improve their marks across portfolio tasks. The findings provided insights in terms of feedback uptake behaviour demonstrating the importance of feedback engagement and show how students’ characteristics correlate with feedback and academic engagement.
|
M. Mihova, P. Georgiev (FINKI, Ss Cyril and Methodius, Skopje, Macedonia)
Student Engagement in Linear Algebra through Programming
Linear Algebra is essential in data-driven and engineering fields, yet students often struggle to connect theory with practice. This paper describes a redesign of a Linear Algebra course that integrates Python-based laboratory exercises alongside standard theoretical content (linear systems, matrix operations, vector spaces, eigenvalues, orthogonality, least squares, and quadratic forms). Labs were implemented in Jupyter/Colab and structured into short tool introductions, theory-aligned tasks, and applied problems with visualization and experimentation. Tutor support encouraged independent problem-solving without providing direct solutions. Student feedback and performance trends suggest increased engagement as reflected in laboratory participation, and improved practical understanding of linear algebra concepts. The approach bridges theory and computation for heterogeneous student populations, complementing rather than replacing formal analytical training.
|
B. Kadežabek (Faculty of Organization and Informatics, Varaždin, Croatia)
Enhancing Visual Accessibility of Projected Educational Content in Teaching Environments Using Software-Based Visual Interaction
Projector supported educational content often presents accessibility challenges, particularly for students seated at a greater distance in lecture halls and classrooms or those with visual impairments. Conventional teaching setups and methods to heavily rely on static slides or limited interaction tools, which restrict the instructor’s ability to dynamically emphasize, clarify or adapt visual information in real time. This paper proposes a modern, visual interaction approach in which an instructor’s computer drives the projected teaching content, while a separate general-purpose tablet device serves as the input medium, aimed at enhancing the visual accessibility of projected content during live instruction. The proposed approach enables instructions to visually annotate, highlight and magnify projected content, fully independent of the content source or format using a separate tablet device with real-time synchronization to the projected display. Unlike hardware-dependent solutions like interactive whiteboards, this approach is fully platformagnostic, adaptable, mobile and requiring only readily available devices. This approach is also intended to support visual emphasis and clarity in teaching environments without requiring changes to existing teaching materials or infrastructure.
|
J. Čulić-Viskota (Independent researcher, Split, Croatia), T. Galeta (Nobula AB, Stockholm, Sweden), S. Maričić (Juraj Dobrila University of Pula, Pula, Croatia)
Developing Active Learning Environments: AI-Enhanced Open-Source VR for Semiconductor Manufacturing
Virtual Reality (VR) models based on open-source software, in combination with CAD/CAM 3D models, are emerging as a possible vital engineering education tool to address the severe semiconductor workforce shortage. These accessible and cost-effective digital twins, often built using platforms like Godot and incorporating AI-powered tutors (LLMs), create an active educational learning environment in virtual cleanrooms. For improved understanding of the complex advanced semiconductor packages - as well as their manufacturing, assembly and testing processes - VR simulations allow users and students to interact with actual products in various stages of design and manufacturing steps. It also allows them to practice complex, precision - demanding procedures like wafer and device handling, photolithography, testing and many other processes involving equipment operation without the risk of contaminating a real, expensive cleanroom. Additionally, VR simulations provide a unique tool for troubleshooting and increasing the quality of designs.
|
K. Aleksić-Maslać, F. Borović, M. Kovačić (Zagreb School of Economics and Management, Zagreb, Croatia)
Great Potential of Generative AI Tools in Education: A Comparative Analysis of Leading GenAI Systems
The rapid development of generative artificial intelligence has led to an increasingly widespread use of GenAI tools in education. This paper analyses and compares the performance of five GenAI tools: ChatGPT, Gemini, Copilot, Perplexity, and DeepSeek, based on ten criteria aligned with the methodological framework applied in our previous research. These criteria include system type, database and update model, accuracy and reliability, prompt-generation speed, language performance, creativity and empathy, plug-in options, input-format recognition, output-format options, and link utilization and reference checking.
A total of N1 = 107 HE educators and N2 = 205 students participated in the study, assessing their satisfaction with individual tools in terms of prompt-generation speed, accuracy and reliability, empathy, plug-in functionalities, and other key elements. Both students and educators currently use ChatGPT most frequently, with students rating usage at 4.2 on a five-point Likert scale and faculty at 3.7. Although all analyzed tools received usage-frequency scores in the 1.4–1.8 range, both groups show an increasing openness toward adopting new GenAI solutions.
|
R. Vrana (Filozofski fakultet Sveučilišta u Zagrebu, Zagreb, Croatia)
AI Literacy in Higher Education: Students’ Perceptions at the Faculty of Humanities and Social Sciences, University of Zagreb on the Role of Artificial Intelligence in the Learning Process
Artificial intelligence (AI) is increasingly becoming a ubiquitous technology across many human activities. Its rapid development has highlighted the importance of established literacies such as digital, media, and data literacy, while also introducing AI literacy as a new competency within the broader framework of information literacy. These literacies are essential for studying, working, and living in a changing information environment. This paper presents the results of an exploratory study of students at the Faculty of Humanities and Social Sciences, University of Zagreb, Croatia, examining their views on AI literacy and the use of AI in studying. The results indicate that respondents perceive artificial intelligence as particularly important in the context of their studies rather than in their personal lives. Most respondents report being self-taught in their use of AI tools and tend to use them intuitively rather than through systematic or formal learning. The findings further show that AI has a positive influence on the speed at which students acquire new learning materials and supports them in understanding the context of the facts they are learning. Respondents most commonly use AI for brainstorming purposes. Overall, they consider themselves adequately prepared to use AI in their future professional work.
|
J. Belščak (Primary school Petrijanec, Petrijanec, Croatia)
Beyond Screens: Teachers’ Perceptions of AI Tools and Educational Robotics in Primary Programming Education
This study examines the integration of generative artificial intelligence (GenAI) tools and educational robotics in primary programming education, with a focus on teachers’ perceptions of their pedagogical value. A survey of 195 primary school teachers explored current instructional practices, including traditional approaches as well as robotics supported, AI supported, and combined technology enhanced instruction. The results indicate that teachers generally perceive AI tools and educational robots as supportive of students’ understanding of programming concepts, creative problem solving, and active engagement. Commonly reported challenges relate to equipment availability, preparation time, and technical reliability, highlighting the need for continued institutional support and targeted professional development.
To complement the survey findings, a classroom case study was conducted to illustrate the practical implementation of educational robotics. The case study demonstrates how hands on interaction with the programmable robot Tobbie II extends programming activities beyond screen based tasks and supports experimentation and embodied problem solving. By combining a large scale teacher survey with a classroom case study, this paper contributes empirical insight into how GenAI tools and micro:bit based robotics can complement each other at different stages of primary programming instruction, with implications for teaching practice, school implementation, and curriculum development.
|
I. Shyshenko, Y. Chkana, O. Martynenko, O. Udovychenko, I. Udovychenko, O. Semenikhina (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Learning Analytics–Driven Personalization of Mathematics and Digital Skills Development for Pre-Service Teachers
This paper presents a model for personalizing the learning pathways of pre-service mathematics and computer science teachers through the use of Learning Analytics (LA). The relevance of the study is driven by the increasing demand for personalized instruction, the need to develop both mathematical and digital competence among teacher education students. The aim of the research is to design and validate an adaptive learning model that uses micro-level activity logs to generate personalized recommendations and identify students’ cognitive struggle points.
The proposed analytical pipeline includes the collection of interaction data from a learning management system, detection of behavioral patterns, clustering of students by learning styles, and a recommendation algorithm based on indicators such as task completion time, number of attempts, revisit frequency, and error typology. The model was tested on a sample of pre-service teachers enrolled in mathematics- and informatics-related courses at a pedagogical university.
The results demonstrate improved mastery of course content and higher-quality argumentative responses, indicating a positive impact on the development of critical thinking. These findings confirm the potential of LA as an effective tool for supporting learning and justify the scalability of the proposed model within teacher education programs.
|
R. Filipović (Zorka Sever Primary School, Popovača, Croatia), I. Filipović (Fran Galović Primary School, Zagreb, Croatia), M. Dumančić ( Faculty of Teacher Education, Zagreb, Croatia)
AI Reliance in Schools: Scale Adaptation, Initial Psychometric Testing, and Insights from a Croatian Teacher Sample
This study investigates how much teachers in the Republic of Croatia rely on artificial intelligence tools in daily work and how that reliance relates to training, frequency of use, access to school equipment, and years of service. A cross-sectional online survey of 113 teachers was conducted in November 2025. We adapted a validated student scale to the school context and measured reliance across three domains: reading and text processing, writing, and numerical and analytic activities. Items were scored on a four-point scale and averaged into a composite measure. Psychometric checks indicated a clear single factor within each domain and high internal consistency. Reliance was moderate overall, highest in writing, then reading, and lowest in numerical and analytic tasks. More frequent use of artificial intelligence in the past year was associated with higher reliance, while training level, equipment access, and years of service showed no meaningful links. The findings suggest that reliance develops through repeated, authentic practice rather than one-off courses or additional hardware. We recommend mentor-guided micro-interventions embedded in real tasks, together with clear ethical and pedagogical protocols on privacy, bias, and transparency. Future longitudinal studies should track domain-specific change and student outcomes.
|
H. Jaakkola (Tampere University, Pori, Finland), J. Henno (Tallinn University of Technology, Tallinn, Estonia), J. Mäkelä (University of Lapland, Rovaniemi, Finland)
The New Era of AI – What is It and Where are We Going?
The current frontier technology, Artificial Intelligence (AI), is based on the use of Large Language Models (LLM) to utilize knowledge of large pretrained datasets. It allows interaction with users to solve problems they define to the AI system. A practical manifestation of the use of current AI are Generative AI (GenAI) models, such as GPT (Generative Pre-trained Transformer), their various user interfaces and growing number of applications. The first, best known, and still most widely used is chatting interface ChatGPT with its one billion weekly users. However, a large number of competing platforms have entered the market, and the original chat has diversified, being able to handle not only verbal communication, but also various forms of media - sound, images and videos - and combinations of these. GenAI has in many ways revolutionized the structures of society, including working life. Although the change it has created can be considered revolutionary, from the perspective of the technology life cycle analysis it is not a fundamental paradigm shift (to use the terms of Freeman & Perez's innovation classification), but rather a change in technological systems. What the next couple of decades will bring – as a path from the current narrow (weak) to general (strong) AI - is a matter worth considering. It is also worth noting that although GenAI has in many ways influenced the use of information technology (IT) in different and new contexts, we do not understand in detail the development paths behind it, nor do we fully understand the limitations associated with it. The purpose of this paper is to examine the development from both a present and future perspective.
|
J. Henno (Tallinn University of Technology, Tallinn, Estonia), H. Jaakkola (Tampere University , Tampere, Finland), J. Mäkelä (University of Lapland , Rovaniemi, Finland)
Programming 3
Ability of Large Language Models (LLMs) to produce fluent text in natural languages has created for many people an illusion that LLMs are intelligent (they are often addressed ambiguously as AI) and can do whatsoever, e.g. produce computer programs. But professional programmers are increasingly avoiding them, claiming that LLMs produce 'code slope'. Proponents of change claim "using AI/LLMs is no longer optional", but using LLMs for software development is described very vaguely; developers nd their teachers have themselves learn through trials and errors. In this paper are described some experiments with LLM-based coding and explained their usability comparing properties of natural and programming languages.
|
H. Singh, M. Dinneen, S. Manoharan (The University of Auckland, Auckland, New Zealand)
Evaluating Kernel Isolation for Performance and Security in CS Assignments
Securely executing untrusted student code is a central requirement of automated grading systems. While lightweight Linux kernel tools—such as namespaces for process and filesystem isolation, seccomp for system call filtering, cgroups for resource management, RLIMITs for process limits, capabilities for privilege control, and SELinux for access policies—provide different levels of isolation, their performance-security trade-offs are understudied in computer science education. This paper empirically evaluates these mechanisms for C programming assignments in both standard and adversarial settings. We measure execution time, memory usage, and overhead across workloads ranging from typical algorithmic problems to resource-intensive and malicious submissions. Alongside this, we test each sandbox’s defenses against common threats, including fork bombs, filesystem breakouts, and privilege escalation attempts. Results show substantial variation in protection levels and runtime costs. We provide practical recommendations for educators and system designers on mechanism selection for safe, efficient autograding.
|
| Radovi
Predsjedatelji: Robert Repnik, Marjan Krašna
|
N. Vindiš (Zavod za usposabljanje, delo in varstvo Dornava, Dornava, Slovenia), M. Krašna (University of Maribor, Maribor, Slovenia)
Artificial intelligence in primary education: Teachers’ and parents’ view
Education is a fundamental pillar of society and must continuously adapt to technological, societal, and pedagogical change. Although a delay between the emergence of new technologies and their integration into education is often perceived negatively, it can allow for critical reflection and informed adoption. Artificial intelligence (AI), particularly in its freely available forms, has lowered barriers to access and is increasingly embedded in everyday educational practices. Following the initial phase of AI-related hype, attention has shifted toward its productive and responsible use. Tasks that can be easily solved by AI are becoming pedagogically obsolete, prompting educators to adapt accordingly. However, educational attitudes toward AI are shaped not only in formal institutions but also within the home environment. This study examines the perspectives of teachers (n = 212) and parents (n = 538) regarding the use of AI in education, based on a survey conducted in the second half of 2025. Results indicate that most respondents use AI when needed and hold predominantly neutral attitudes toward it. Reluctance toward subscription-based AI services was observed, while parents expressed greater concern about data security. Both groups emphasized the need for clearer technical support and guidelines to maintain critical distance when interpreting AI-generated feedback.
|
R. Repnik (Faculty of Natural Sciences and Mathematics, University of Maribor, Maribor, Slovenia), D. Osrajnik (Primary School Radlje ob Dravi, Radlje on Dravi, Slovenia), E. Klemenčič (Faculty of Natural Sciences and Mathematics, and Faculty of Energy Technology, University of Maribor, Maribor, and Krsko, Slovenia)
Analysis of Students’ Guided Self-Reflection after Digitally Supported Fieldwork in Physics Education
Despite widespread school digitalization, the effective use of digital measuring instruments in physics classrooms depends on teachers’ digital competences. These extend beyond operational skills and include evaluating data quality, measurement uncertainty, and the didactic value of digitalization. This paper analyzes guided self-reflection conducted by students after three digitally supported fieldwork modules within the course Digitally Supported Teaching of Physics Contents through Fieldwork. The reflections focused on critically assessing experiences with digital measuring instruments, data recording and export, digital analysis, and interpretation of results using digital tools. The self-reflection framework included four dimensions: operation and reliability of devices, recording and exporting data, digital analysis and interpretation, and the didactic potential of digital tools in future classroom practice. Students’ reflections were analyzed qualitatively to identify patterns in digital competences, awareness of measurement uncertainty, and the ability to relate digital data collection to scientific reasoning. The findings show that guided self-reflection helps students recognize limitations of digitalization in fieldwork and improves readiness to integrate digital tools into physics teaching. This approach also functions as an effective formative assessment tool and promotes digital and science literacy among pre-service physics teachers, supporting future classrooms.
|
N. Ponomarenko (Sumy State University, Sumy, Ukraine), O. Semenog (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), M. Sadivnycha (Sumy State University, Sumy, Ukraine), L. Lukіanyk (Rivne State Humanitarian University, Rivne, Ukraine), D. Davydenko (Sumy State Pedagogical University named after A. S. Makarenko, Sumy , Ukraine), O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Analysis of the Relationship Between ICT Use, Media Literacy, and Media Trauma Among Students in a War-Affected Border Region (Sumy, Ukraine)
Despite the active development of media education programs, there is still limited empirical evidence on how ICT practices, self-assessed media literacy, coping strategies, and subjective symptoms of media trauma are combined in higher education students living in Ukrainian regions close to the front line. This paper presents the results of a pilot survey of higher education students from universities in Sumy. The aim is to describe the student profile in terms of ICT use and media consumption, to assess the prevalence of media-trauma-related symptoms, and to analyze how self-assessed media literacy, ICT practices, and the frequency of symptoms related to media trauma intersect. The empirical basis is an online survey of 252 participants. The questionnaire covered ICT practices, news sources, self-assessed media literacy, strategies for responding to distressing content, as well as the frequency of anxiety, sleep/concentration problems, and emotional numbing that respondents associate with war-related news and social media. The findings confirm that smartphones and Telegram channels are the primary environments for both learning and news consumption. A substantial proportion of respondents frequently experience anxiety and at least episodically report sleep/concentration problems and emotional numbing related to media content. Pairwise analyses reveal that higher news-checking frequency, residing in the border region, and relying on Telegram as a primary news source are associated with more severe symptoms. Differences in coping strategies are observed between groups with different levels of self-assessed media literacy; however, a high level of confidence does not provide an unambiguous protective effect.
|
O. Shukatka (Ivan Franko National University of Lviv, Lviv, Ukraine), N. Оliinyk (Vinnytsia Mykhailo Kotsyubynskyi State Pedagogical University, Vinnytsia, Ukraine), S. Kharchenko (Sumy National Agrarian University, Sumy, Ukraine), A. Yurchenko (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), S. Bohoslavskyi (Sumy State Pedagogical University named after A.S. Makarenko, Sumy , Ukraine), O. Semenikhina (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Hybrid Learning for Pre-Service Physical Education Teachers: Conceptual Model in a Digital Learning Environment and its Expert Evaluation
The article presents the theoretical grounding and expert evaluation of a hybrid learning model for pre-service physical education teachers implemented in a digital learning environment that integrates an LMS, mobile activity trackers, video services, and communication platforms. The aim of the study is to describe the structure and digital components of the model and to analyze the results of its expert appraisal. Seven experts in physical education, higher education, and digital technologies evaluated the model against a set of criteria: relevance, scientific soundness, alignment with standards, clarity and measurability, integrative potential, practical orientation, logic and feasibility, balance of learning formats, coherence of assessment, validity of outcomes, innovativeness, practical value, and overall integrity. A five-point Likert scale was used, descriptive statistics were calculated, and Kendall’s coefficient of concordance was applied. The overall mean score for the model was 3.90; the conceptual and procedural blocks, as well as the logic and technological feasibility of hybrid interaction among participants in the educational process, received the highest scores. More cautious evaluations were recorded for the content and outcome blocks, particularly regarding the integrative potential of combining applied sports, gamification, and digital platforms, as well as the validity of assessing complex learning outcomes. Qualitative analysis of experts’ open-ended responses confirmed the high potential of the model, while also revealing risks related to organizational complexity and highlighting the need for targeted teacher training. The study concludes that further experimental testing of the model, the development of diagnostic tools, and methodological guidelines for its phased implementation in pre-service physical education programs are warranted.
|
L. Sytnyk (Sumy National Agrarian University, Sumy, Ukraine), K. Zhurba (Institute of Problems on Education of the NAES of Ukraine, Kyiv, Ukraine), Y. Rudenko (Sumy National Agrarian University, Sumy , Ukraine), S. Petrenko, M. Soroka (Sumy State Pedagogical University named after A. S. Makarenko, Sumy , Ukraine), O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Low-Code Platforms for Teaching Big Data Analytics: a Comparative Study of Orange Data Mining, Weka, and KNIME
The rapid growth of Big Data technologies necessitates revised instructional approaches to teaching data analytics to students across various majors. Low-code platforms enable the design of learning activities without advanced programming skills, while still providing access to professional data analytics tools and methods. This article presents a comparative study of three low-code tools (Orange Data Mining, Weka, and KNIME) used for teaching Big Data analytics. The experiment involved students of technical, economic, and humanities programs at Ukrainian universities. The comparison criteria included task completion time, number of typical errors, quality of the resulting models, and self-reported interface complexity. The results indicate a substantially higher level of user satisfaction and perceived interface intuitiveness for Orange Data Mining compared to Weka and KNIME, as evidenced by fewer errors and faster task completion. Orange is perceived by students as the most accessible tool at the introductory level and yields the fewest errors at the basic stages of data analysis. Weka appears to be the most suitable tool for tasks of moderate complexity, particularly for familiarizing students with classical machine learning algorithms. KNIME provides the highest model accuracy and the best performance on more complex workflow-oriented tasks; however, it requires more time and effort to master.
|
A. Yurchenko (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), I. Kharchenko (Sumy National Agrarian University, Sumy, Ukraine), M. Ostroha, T. Aleksakhina, I. Gorovoy, O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Digital Multilingualism and IT-Supported Language Education in Higher Education
The paper examines how information technologies can systematically support multilingual language education in higher education, with a particular focus on second-language learning. It argues for a shift from the piecemeal use of individual digital tools to coherent models that integrate technological, pedagogical, linguistic, multilingual, and organizational dimensions. Drawing on a targeted bibliographic overview of publications on computer- and mobile-assisted language learning, virtual exchange, content-and-language integrated learning, English-medium instruction, and the implementation of AI-based tools, the paper proposes a four-dimensional framework for IT-supported multilingual language education. The framework categorizes digital tools, typical pedagogical scenarios, clusters of learning outcomes, and organizational conditions for technology use. Using examples of task-oriented language courses and AI-supported academic writing modules, it illustrates how different configurations of tools and scenarios influence learners’ competencies. The paper discusses how the framework can be used for the design and review of language courses, for aligning an institution’s language policy with platform choices and AI use guidelines, and for setting directions for further research, including design-based studies that iteratively refine multilingual language education programs.
|
О. Kudrina (Sumy State Pedagogical University named after A.S.Makarenko, Sumy, Ukraine), A. Yurchenko (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), I. Chystiakova (Sumy State Pedagogical University named after A.S.Makarenko, Sumy, Ukraine), O. Kovtun (Hryhorii Skovoroda University in Pereiaslav, Pereiaslav, Ukraine), V. Bespalyi, O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
VOSViewer-Based Workshop for PhD Students and the Development of Their Research Skills
The study reports findings from a pilot quasi-experimental study that evaluated the effectiveness of a VOSviewer-based workshop for doctoral (third-cycle) learners in the field of education. The study examines whether, and in what ways, participation in such a workshop supports the development of research skills among prospective PhD candidates. The sample consisted of 31 doctoral learners, comprising a control group (CG, n = 15) and an experimental group (EG, n = 16). Both groups attended compulsory courses in research methodology and academic writing; however, only the experimental group participated in the VOSviewer workshop. Using a pre-test/post-test design, the study assessed two outcomes: (1) skills in working with scholarly literature (the ability to identify key concepts and their relationships; to recognize research methods used in a publication; to articulate research gaps clearly and with justification; and to select up-to-date and relevant sources for a future study within a specified bibliometric database) and (2) the quality of individual research proposals (expert evaluation based on the following criteria: study justification, originality, conceptual coherence, and methodological adequacy). The EG demonstrated statistically significant improvements across the indicators, with effect sizes ranging from medium to above medium, whereas changes in the control group were small and not statistically supported. The study provides initial empirical evidence that a VOSviewer-based workshop may serve as a promising component of research methodology courses in doctoral programs, while also underscoring the need for longer-term and larger-scale evaluations of its impact.
|
D. Turchyn, A. Yurchenko, O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Offline-First PWA for Teaching Informatics in Wartime Conditions: Architecture and Survey Results
The paper focuses on the design and evaluation of an offline-first Progressive Web App (PWA) for teaching informatics in wartime conditions, where online learning takes place amid recurring disruptions to Internet connectivity and power supply. The proposed solution is based on a layered architecture: a client-side PWA with local progress persistence and an event queue; an offline/edge layer built on service workers and caching; and an online layer that supports synchronization, AI-based hints, and teacher analytics. Pedagogical support is implemented as a stepwise scaffolding model that strengthens learners’ autonomy while limiting the delivery of ready-made solutions; the motivational layer emphasizes resilience through non-competitive mechanisms of individual progress. The empirical evaluation was conducted through surveys of pupils (n = 33), teachers (n = 10), and experts (n = 5), using a five-point Likert scale and open-ended questions. The analysis combined descriptive statistics (M, SD) with a thematic synthesis of comments. The results show high ratings for usability under unstable connectivity and for alignment with the needs of war-affected border regions. Pupils rated the multi-level hints and AI support particularly highly, whereas teachers provided more differentiated assessments of this component. Experts confirmed the strengths of the offline-first approach and identified priorities for further improvement, including more intuitive navigation and expanded teacher reporting tools (notably, exporting results).
|
Y. Rudenko , V. Pivtoraiko (Sumy National Agrarian University, Sumy, Ukraine), M. Antonchenko , L. Sierykh (Sumy Regional Institute of Postgraduate Pedagogical Education, Sumy, Ukraine), O. Saenko (Sumy State Pedagogicak University, Sumy, Ukraine), N. Dehtiarova (Sumy State Pedagogicak University named after A.S. Makarenko, Sumy, Ukraine)
Developing Digital Competencies in Learning through Visualized Plant Protection Case Studies
The paper presents the results of a study aimed at developing digital competencies of higher education students through the use of visualized plant protection case studies. The research was conducted with third-year students of Sumy National Agrarian University within the course “Plant Protection.” During the pedagogical experiment, three educational case studies were implemented, involving the use of digital tools for data analysis, interpretation, and visualization based on real professional situations in the field of plant protection.
The scientific novelty of the study lies in substantiating and experimentally validating visualized case studies as an effective pedagogical tool for the targeted development of digital competencies of future agricultural specialists. To assess the level of digital competency development, three criteria were defined: information-digital, analytical-practical, and visual-digital. For each criterion, three levels of competency formation were identified: low, medium, and high.
The analysis of the results was carried out using descriptive statistics and the Pearson chi-square (χ²) test. The findings demonstrated statistically significant positive changes in the levels of students’ digital competencies after the implementation of visualized case studies, including a decrease in the proportion of students with a low level and an increase in those with medium and high levels. In addition, a student survey was conducted, the results of which indicated positive perceptions of the case-based approach, increased learning motivation, strong interest in visualized case studies, and students’ willingness to continue learning using the case method. The study concludes that the use of visualized plant protection case studies is an effective approach to developing digital competencies in the professional training of future agricultural specialists. The proposed approach can be adapted for other agricultural and pedagogical disciplines
|
A. Mirković Moguš, K. Dobi Barišić (Faculty of Education, Osijek, Croatia)
Exploring Students’ Prompt Engineering Practices and Competencies in Higher Education
Artificial intelligence (AI) is gradually being integrated into higher education. Digital competencies, such as formulating effective prompts, referred to as prompt engineering, are gaining increasing importance. Empirical research on students’ familiarity with prompt engineering and their prompt formulation techniques remains limited. This research explores the use of AI tools and concepts of prompt engineering by students, kinds of commonly used prompt strategies, and their perception of skills measured by the Prompt Engineering Competence Scale (PECS). A quantitative survey was distributed to university students to capture their familiarity, usage patterns, and skill levels through Likert-scale and open-ended items. Results showed that students’ knowledge and habits were highly variable. Although the majority of the respondents use AI tools, their comprehension of posted structured prompting was quite inconsistent. Zero-shot prompting and chain-of-thought prompting were the most frequently used prompt techniques. Students' perception of their own competences was, on average, neutral or indifferent. The findings imply significant disparities between the routine use of AI and the use of strategic, evidence-based prompt engineering. This research emphasizes the necessity of incorporating prompt engineering literacy as a part of the university syllabi in order to facilitate socially responsible, effective, and academically sound use of AI.
|
N. Borozenets, A. Rozumenko, A. Rozumenko (Sumy National Agrarian University, Sumy, Ukraine), I. Shyshenko, O. Udovychenko, A. Pidopryhora , O. Semenikhina (Sumy State Pedagogical University named after A.S. Makarenko, Sumy, Ukraine)
Effectiveness of Digital Microlearning Units in the Mathematical Training of Students of Non-Mathematical Specialties
The article examines the effectiveness of digital microlearning units in the mathematical training of students of non-mathematical specialties. The relevance of the study is determined by the need to modernize higher mathematics courses and increase student motivation through the use of contemporary digital technologies and short modular learning units (micro-units). The aim of the study was to determine whether the implementation of microlearning units contributes to improved learning outcomes, increased student engagement, and greater interest in learning.
The research was conducted using a comparative pre–post design, in which traditional instruction was compared with a course structured in the format of digital microlearning units. Data were collected through achievement testing, motivation surveys, and analysis of students’ interaction with digital learning platforms.
The results indicate an improvement in students’ mastery of mathematical content, higher learning motivation, and more active use of digital resources among students who studied using the microlearning model. The findings confirm that the use of digital microlearning units is an effective approach to modernizing mathematical education for non-mathematical specialties and can be scaled to other academic disciplines.
|
T. Penniston (University of Maryland, Baltimore County (UMBC), Baltimore, United States), N. Kadoić, N. Begičević Ređep (University of Zagreb Faculty of organization and informatics, Varaždin, Croatia)
Multicriteria Prioritisation of Innovative Strategic Educational Directions in Higher Education
Today's turbulent environment exposes higher education institutions (HEIs) to accelerating change and disruption; they must respond with greater agility to face these new challenges. In response to these dynamics, a core question emerges: What in higher education teaching and learning should be reconsidered or reshaped in the name of innovation and adaptation, and what merits conservation due to time-tested social or economic value?Based on a literature review, this paper defines innovative strategic educational approaches, while ensuring students are equipped with the future-ready skills demanded by the market. Through the lens of multi-criteria, data-informed decision making grounded in literature-based themes, this research aims to provide a transparent model to support innovations in teaching and learning. It compares traditional models, technology-mediated formats, credential innovations, and applied learning experiences, community centric/social learning against criteria (e.g. cost/ROI; graduate employability impact; institutional scalability; organizational acceptance; reputational value; strategic alignment) to inform decision making. In this research, purposive sampling of expert stakeholders will identify experts to rate the defined approaches and establish a baseline comparison model. The Analytic Hierarchy Process (AHP) is used to rank the approaches based on the experts' judgment to provide a platform for increasing agility and implementing innovative strategic educational approaches in HEIs.
|
M. Duić (University of Zadar, Zadar, Croatia)
Exploration of the Croatian Photographic Heritage in Europeana
The Europeana digital library has a significant number of photographs that were created in different periods in the territory of Croatia. These photographs present contents related to various important topics, people and phenomena about which is taught at Croatian schools and universities. Therefore, these photographs can be very useful in the processes of teaching and creation of educational content, especially since many of them are free for use for educational purposes. This paper will explore the available data in Europeana on photographs from the 19th and 20th centuries that were created in the territory of Croatia. The goal of the paper is to explore various aspects related to these photographs: how many photographs are available through Europeana; from which countries and institutions originate these photos and how many photos there are from individual countries and institutions; what are the copyright status, image size and formats of photographs; which Europeana aggregators mediated between providing institutions and the Europeana library. The research method of content analysis will be used. The research results could be useful for various purposes. For example, acquired insights could support people in finding valuable old photographs and using them in educational and other activities.
|
| Radovi
Predsjedatelj: Ivan Kaštelan
|
I. Vlahović, I. Ogrizek Biškupić, M. Balković (Algebra Bernays University, Zagreb, Croatia)
Enhancing Educational Integrity: Leveraging RAG-Based NotebookLM for Reliable Moodle Question Bank Automation
The rise of generative AI tools for education raises concerns about the accuracy and reliability of their materials due to frequent hallucinations in content created by popular large language models. Tools like Google NotebookLM, which use retrieval-augmented generation (RAG) and new instructional design methods, like Item Response Theory (IRT), for verifying quiz correctness, have potential to enhance lesson planning and student learning. Following UNESCO competency framework guidelines for teachers focused on human-centered approach and ethics of AI, AI foundations and applications, AI education, and AI for professional learning, implementation in question creation could be a solid foundation for integration in popular learning platforms like Moodle. In this paper one possible framework for Quizzes generation in NotebookLM and integration to Moodle LMS workflow and question bank is shown as automation pipeline. Using this pipeline process of learning material creation can contribute to “AI dividend” i.e. saving hours of work on preparing teaching materials and checking knowledge and methods of checking students' mastery of course material. This pipeline could contribute also to integration of generative AI in digital learning (GAIDL framework) of educational institutions as standard for using RAG architecture in lesson creation.
|
N. Bošnjaković (Primary school Vođinci, Vođinci, Croatia), I. Đurđević Babić (Faculty of Education, Osijek, Croatia)
Artificial Neural Network for Identifying Teachers' Attitudes Toward Digital Gamification
Due to its many advantages in education, digital gamification has been the subject of numerous studies in recent years. With strong prevalence at the university level, its benefits, characteristics, and various influences – such as its impact on motivation, engagement, and student success – have been explored. Because teachers' competencies are crucial for the successful implementation of digital gamification, this paper analyzes the potential of artificial neural networks to identify teachers' attitudes toward digital gamification as a means to facilitate their lectures, using attitudes toward the traits and competencies they consider important for the successful application of digital gamification, along with general demographic information. The best multilayer perceptron (MLP) neural network model achieved an overall classification accuracy of 71.05%, confirming the capabilities of neural networks. Sensitivity analysis revealed that, in addition to gender and level of education – which have the greatest impact on the model's effectiveness – playfulness, openness to technology, and exploratory inclination as perceived traits are also among the most influential variables on the model's success.
|
V. Kalajžić, B. Miočić (University of Zadar, Zadar, Croatia)
Education and Artificial Intelligence in Articles on the Hrčak Portal – A Communication Perspective
In the contemporary digital environment, education is undergoing significant transformations driven by technological and social conditions, with artificial intelligence increasingly shaping modes of learning and teaching as well as the organization and functioning of educational systems. Although artificial intelligence is most often examined through its technological and pedagogical implications, its communication dimension in education remains less systematically explored. This paper analyzes the extent to which and the ways in which Croatian scientific and professional journals have addressed artificial intelligence in the educational context, with a particular focus on the communication dimension. The research is based on papers dealing with artificial intelligence and education published in the Hrčak database, the central national portal aggregating Croatian scientific and professional open-access journals. The aim of the paper is to examine, through analysis of the selected corpus, bibliographic characteristics, thematic emphases, and the communication aspects of artificial intelligence in education. The communication aspect encompasses artificial intelligence as a mediator of educational communication, changes in educational interaction, and the understanding and shaping of AI-mediated messages, including issues of accuracy, reliability, and interpretation of information. The study employs an analysis of bibliographic characteristics combined with qualitative and quantitative content analysis of papers selected according to keywords relevant to the fields of artificial intelligence and education. The contribution of this paper lies in providing a systematic overview of domestic scientific and professional output, identifying thematic trends, and highlighting the communication dimension of artificial intelligence in education.
|
M. Bednjanec (Zagreb University of Applied Sciences, Zagreb, Croatia), S. Jovčić (VERN' University, Zagreb, Croatia), M. Nižetić (Vladimir Prelog Science School, Zagreb, Croatia)
Artificial Intelligence as a Time-Efficiency Enabler in Academic Assessment Practices
The increasing integration of artificial intelligence (AI) into higher education has introduced new opportunities for improving teaching efficiency, particularly in assessment and learning management processes. This paper investigates the extent to which teaching staff adopt AI-based tools in their professional practice, with a specific focus on time savings achieved through AI-supported grading, automated feedback, and AI functionalities embedded within learning management systems (LMS).
The study is based on a quantitative research design employing a structured survey administered to teaching staff in higher education institutions. The survey examines the frequency and purposes of AI tool usage, perceived time efficiency in assessment-related tasks, and the impact of AI-driven LMS analytics on monitoring student progress and supporting instructional decision-making. Key dimensions include automated grading, feedback generation, learning analytics, and overall user satisfaction.
The findings are expected to indicate that AI tools significantly reduce the time required for grading and administrative tasks while enhancing the timeliness and consistency of feedback provided to students. Furthermore, the use of AI-enhanced LMS features is anticipated to support more effective tracking of student performance and early identification of learning difficulties. The results contribute to understanding current adoption patterns and provide insights for institutions seeking to implement AI technologies in a responsible and effective manner.
|
M. Bednjanec (Zagreb University of Applied Sciences, Zagreb, Croatia), S. Jovčić, D. Nižetić (VERN' University, Zagreb, Croatia)
Information Systems for Teaching Management in Higher Education
The ongoing digital transformation of higher education has intensified the demand for structured and efficient solutions for managing teaching-related activities. Information systems for teaching management provide integrated support for the planning, organization, monitoring, and evaluation of teaching processes, while reducing administrative complexity and enhancing institutional transparency. This paper presents a conceptual analysis of information systems for teaching management and examines their role in improving the efficiency and quality of teaching processes in higher education institutions.
A quantitative, survey-based methodological model is proposed to assess user perceptions of teaching management systems. The model incorporates key evaluation dimensions, including system usability, administrative efficiency, communication support, and transparency of teaching processes. Perceived teaching efficiency and overall user satisfaction are treated as outcome variables. Data collection is designed through a structured online questionnaire administered to students and teaching staff, using a five-point Likert scale.
The proposed methodological framework enables a systematic, user-centered evaluation of teaching management information systems across diverse higher education contexts. The results are expected to indicate that effective system implementation positively influences teaching organization, stakeholder communication, and user satisfaction. The study provides a foundation for future empirical research and supports evidence-based decision-making in the context of digital transformation in higher education.
|
M. Konecki, M. Konecki, J. Antolos (Fakultet organizacije i informatike, Varaždin, Croatia)
Artificial Intelligence as an Aiding Tool Among Higher Education Students
Artificial intelligence is transforming the existing paradigms and becomes a major change catalysator in digital transformation society. The well-known way to do business and educate changes, as well as certain aspects of society everyday ecosystem. In this kind of technological setting higher education teachers face many challenges that come from students learning and working differently by using artificial intelligence as an aiding tool. One of the important research questions is how frequently and how extensively students use artificial intelligence tools in their learning process and also how confident and critical they are about the provided results generated by artificial intelligence. In this paper the research results related to mentioned research questions are presented and discussed. Also, several recommendations for student use of artificial intelligence based on presented results are given and elaborated.
|
D. Marnika, B. Kovačić (Fakultet informatike i digitalnih tehnologija, Rijeka, Croatia), M. Gligora Marković (Medicinski fakultet, Rijeka, Croatia)
Algorithmic Thinking in the Era of Large Language Models: A Review of Measurement Instruments in Programming Education
Algorithmic thinking is a fundamental competence in programming and digital problem-solving, encompassing abstraction, logical reasoning, and creativity. While developing this skill enables learners to participate actively in technology creation, programming education is often associated with high cognitive load, difficulties in understanding abstract concepts, and reduced self-efficacy. Contemporary pedagogical approaches, such as project-based and problem-based learning, can increase motivation and engagement; however, they also have limitations, particularly in collaborative learning contexts.
Recently, large language model (LLM)–based tools, such as ChatGPT, have attracted increasing attention for their potential to support programming education by providing personalised feedback, structuring knowledge, and guiding learners through problem-solving processes. Despite this growing interest, empirical research examining the impact of LLM-supported learning on the development of algorithmic thinking remains limited.
This paper reviews existing scientific literature that employs research instruments and process-based scales for measuring algorithmic thinking in LLM-supported programming education. Particular emphasis is placed on applications in higher education, as well as in primary and secondary education. The findings provide a foundation for future research aimed at developing valid and comprehensive instruments that capture the key cognitive processes underlying algorithmic problem-solving.
|
K. Stanič (Secondary School of Economics and Gymnasium Maribor, Maribor, Slovenia), A. Špernjak (Faculty of Natural Sciences and Mathematics, University of Maribor, Maribor, Slovenia)
Developing Teachers’ Digital Competencies Through Designing Augmented Reality Learning Materials for Biology Education
The integration of augmented reality (AR) into biology education places new demands on teachers, particularly in relation to the development of advanced digital competencies. Designing AR-based instructional materials requires teachers to move beyond subject-specific biological knowledge and engage in purposeful pedagogical, technological, and didactic decision-making. This paper reports about a pilot design-based study of teachers’ digital competencies developed when creating AR learning materials for biology, with particular emphasis on the Creation phase defined within the Teachers´ Augmented Reality Competencies (TARC) framework. The findings indicate that this process supports the development of advanced digital competencies, including digital content creation, pedagogically informed multimedia design, and the critical selection and adaptation of AR tools in alignment with biological learning objectives. Teachers are also required to demonstrate pedagogical digital competence, such as managing cognitive load, ensuring conceptual clarity, supporting inquiry-based learning, and maintaining the scientific accuracy of augmented representations. In addition, attention must be given to usability, accessibility, and classroom implementation constraints proved to be an integral part of the design process. Overall, the preparation of AR materials fosters higher-level digital competencies, including problem solving, critical evaluation of digital resources, and reflective instructional design. This research highlights that effective AR integration in biology education depends on a balanced combination of biological expertise, pedagogical knowledge, and digital competence, positioning teachers as active designers of meaningful technology-enhanced learning environments. To strengthen these initial insights, a future research phase should complement design reflections with objective measurement instruments.
|
A. Yurchenko, О. Kudrina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), V. Riznyk (Hryhorii Skovoroda University in Pereiaslav, Ukraine, Pereiaslav, Ukraine), Y. Khvorostina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine), Y. Diemientiev (Sumy State Pedagogical University named after A. S. Makarenko, Sumy , Ukraine), O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Selecting Digital Tools for Collaborative STEM Learning in War-Affected Higher Education Based on an Analytic Hierarchy Process Approach
The article addresses the selection of digital tools for collaborative STEM learning in a university operating under war conditions. The aim of this study is to identify which platforms should be considered a priority for organizing students’ collaborative work and to determine the criteria that drive this choice. The empirical basis is the course “ICT in Education” at Sumy State Pedagogical University named after A. S. Makarenko. The expert group consisted of five STEM lecturers with experience in blended and online teaching in a frontline region. The Analytic Hierarchy Process (AHP) was applied to construct a three-level model (“goal – criteria – alternatives”), conduct pairwise comparisons, and obtain group weights. Five tools (Miro, Padlet, Microsoft Teams with Whiteboard, Zoom with interactive whiteboards, BigBlueButton) were evaluated against six criteria: support for real-time collaboration, tracking of individual contribution, pedagogical affordances for STEM tasks, accessibility and technical requirements, integration with the institutional environment, and usability for lecturers and students. The most important criterion was pedagogical affordance, followed by real-time collaboration and accessibility under war-related constraints. Miro ranked first, followed by Microsoft Teams with Whiteboard in second place, and Padlet in third. Sensitivity analysis revealed that Miro's leading position remained stable under moderate changes in criterion weights. The findings can be used to guide the development of the university’s digital learning environment, to design collaborative STEM courses, and to inform further research that combines AHP-based models with empirical data on learning outcomes.
|
M. Drushlyak , D. Budianskyi, V. Shamonia, A. Bobokalo, H. Skoropad, O. Semenikhina (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Moral Dilemmas of Using AI in Academia: What Do Students Think?
The paper presents the results of an empirical study on students’ perceptions of the ethical use of generative artificial intelligence (GenAI) in higher education. The analysis is based on responses from 362 students enrolled in Ukrainian higher education institutions. The data were collected as part of the international project “Students’ Perception of ChatGPT” (https://www.covidsoclab.org/chatgpt-student-survey/). The study examines students’ awareness of institutional AI policies, their attitudes toward ethical and legal aspects of AI use, and their judgments on the acceptability of different AI use scenarios in academic settings. Two composite indices were constructed: one measuring ethical sensitivity, and the other measuring perceived acceptability of AI use. Comparative analyses were conducted by gender and level of education. The results indicate that Ukrainian students generally hold neutral or ambivalent views about academic AI use. Their awareness of institutional policies is low. The findings suggest that ethical AI use in education requires more than regulation and prohibitions. It calls for the development of students’ internal responsibility for their own academic outcomes.
|
M. Drushlyak , T. Lukashova, D. Ielizarenko, O. Nadtochyi, S. Korzh (Sumy State Pedagogical University named after A. S. Makarenko, Sumy, Ukraine)
Digital Tools in the Training of Future Doctors of Philosophy in Mathematics
The paper explores the role of digital tools in the training of future Doctors of Philosophy in Mathematics. It highlights the growing importance of digitalization in research activities, which significantly transforms the process of conducting dissertation studies. The authors emphasize that technological progress facilitates access to scientific sources, organization of research data, and the preparation of scholarly publications. At the same time, the digital environment poses new challenges related to the need for critical evaluation of information, data management, and the development of academic digital literacy. The study analyzes a range of digital tools that support various stages of the research process, including reference management systems, software for data analysis and visualization, cloud collaboration services, and specialized applications such as VOSviewer for bibliometric mapping. Based on the analysis of current practices, the article outlines recommendations for integrating these tools into doctoral training programs in mathematics. The authors argue that effective use of digital technologies not only enhances the efficiency of individual research but also contributes to the development of digital competence, critical reasoning, and methodological culture among young researchers.
|
|
Osnovni podaci:
Voditelji:
Snježana Babić (Croatia), Marina Čičin-Šain (Croatia), Jaak Henno (Estonia), Hannu Jaakkola (Finland)
Voditeljstvo:
Michael E. Auer (Austria), Marjan Krašna (Slovenia), Nadica Kunštek (Croatia), Dusko Lukac (Germany), Robert Repnik (Slovenia)
Programski odbor:
Toni Aaltonen (Finland), Lejla Abazi-Bexheti (Macedonia), Ivan Kaštelan (Serbia), Mario Konecki (Croatia), Božidar Kovačić (Croatia), Gorana Mudronja (Croatia), Dana Paľová (Slovakia), Libuša Révészová (Slovakia), Mika Saari (Finland), Frano Škopljanac-Mačina (Croatia)
Prijava/Kotizacija:
|
PRIJAVA / KOTIZACIJE
|
CIJENA U EUR-ima
|
|
Do 15.5.2026.
|
Od 16.5.2026.
|
| Članovi IEEE |
297 |
324 |
| Članovi MIPRO |
297
|
324
|
| Studenti (preddiplomski i diplomski studij) te nastavnici osnovnih i srednjih škola |
165
|
180
|
| Ostali |
330
|
360
|
Studentski popust se ne odnosi na studente doktorskog studija.
OBAVIJEST AUTORIMA: Uvjet za objavu rada je plaćanje najmanje jedne kotizacije po radu. Autorima 2 ili više radova, ukupna se kotizacija umanjuje za 10%.
Kontakt:
Snježana Babić
Sveučilišta Jurja Dobrile u Puli
Fakultet informatike
Alda Negrija 6
52100 Pula, Hrvatska
E-mail: snjezana.babic.ce@gmail.com
Marina Čičin-Šain
Društvo kibernetičara
Jadranski trg 1/II
51000 Rijeka, Hrvatska
E-mail: cicin.sain.marina@gmail.com
Prilikom prijave rada treba predložiti kategorizaciju rada (znanstveni, stručni).
Ako je sažetak rada na hrvatskom, rad i izlaganje je također na hrvatskom jeziku. Ako je sažetak rada na engleskom, rad i izlaganje rada također treba biti na engleskom jeziku.
Najbolji radovi bit će nagrađeni.
Prihvaćeni radovi bit će objavljeni u zborniku radova s ISSN brojem. Radovi na engleskom jeziku prezentirani na skupu bit će poslani za uključenje u digitalnu bazu IEEE Xplore.
Mjesto održavanja:
Opatija je vodeće ljetovalište na istočnoj strani Jadrana i jedno od najpoznatijih na Mediteranu. Ovaj grad aristokratske arhitekture i stila već više od 180 godina privlači svjetski poznate umjetnike, političare, kraljeve, znanstvenike, sportaše, ali i poslovne ljude, bankare, menadžere i sve kojima Opatija nudi svoje brojne sadržaje.
Opatija svojim gostima nudi brojne komforne hotele, odlične restorane, zabavne sadržaje, umjetničke festivale, vrhunske koncerte ozbiljne i zabavne glazbe, uređene plaže i brojne bazene i sve što je potrebno za ugodan boravak gostiju različitih afiniteta.
U novije doba Opatija je jedan od najpoznatijih kongresnih gradova na Mediteranu, posebno prepoznatljiva po međunarodnim ICT skupovima MIPRO koji se u njoj održavaju od 1979. godine i koji redovito okupljaju preko tisuću sudionika iz četrdesetak zemalja. Ovi skupovi Opatiju promoviraju u nezaobilazan tehnološki, poslovni, obrazovni i znanstveni centar jugoistočne Europe i Europske unije općenito.
Detaljnije informacije se mogu potražiti na www.opatija.hr i www.visitopatija.com.
|
|