Selected publications 2024
[1] B. Szilágyi and R. Lakner, “NLP-enhanced classification of remote employment opportunities,” MAGYAR NYELVŐR, vol. 148, no. 5, pp. 747–760, 2024.
Efficiently filtering large amounts of information is a challenge in online job searching. Addressing these challenges, artificial intelligence tools, especially Natural Language Processing, are becoming increasingly essential. The article presents an automated text classification system for identifying remote work opportunities based on job advertisement texts. Following the collection and preparation of data from online job portals, the method applies rule-based classification and machine learning to analyze the texts. https://nyelvor.mnyknt.hu/wp-content/uploads/148514.pdf |
[2] Trenka, Zoltán, V. Seres, Viktória, and Pogatsnik, Monika, “The impact and potential of artificial intelligence in language education,” MAGYAR NYELVŐR, vol. 148, no. Special Issue in English (2024-5), pp. 682–695, 2024.
Over the past decade, the education community, together with students, could not avoid taking advantage of the digital revolution and integrating different applications into the teaching and learning process. The research explored the potential of using AI-based applications in the teaching of humanities subjects. The focus was on teachers of Hungarian and foreign languages, whose attitudes and knowledge on the topic were measured through a random, anonymous questionnaire survey. The survey confirmed a low level of awareness among the teaching community, partly due to a higher proportion of digital immigrants. However, teachers are open to learning about AI and see positive benefits in using it. In line with previous assumptions, foreign language teachers have higher levels of awareness and are the main users of AI-based applications, while Hungarian language teachers are much less likely to use them. The survey also covered the use of chatbots, and the results showed that a high percentage of teachers recognize the tasks solved using ChatGPT. https://nyelvor.mnyknt.hu/wp-content/uploads/148510.pdf |
[3] Tolner, Nikoletta and Pogatsnik, Monika, “The large language models in the service of education,” MAGYAR NYELVŐR, vol. 148, no. Special Issue in English (2024-5), pp. 629–649, 2024.
The name Artificial Intelligence (AI) appeared in the 1950s, but mankind had a much older desire to perform certain tasks with machines. AI permeates our everyday lives, so the widest possible layers of society must acquire knowledge about Artificial Intelligence. The understanding and acceptance of AI are crucial for social development, equality, and innovation. For these reasons, the Hungarian Artificial Intelligence Coalition was founded in 2018, the main goal of which is to bring technology closer to all actors in society. As a result of the coalition’s work, the Artificial Intelligence Strategy of Hungary was published in May 2020, in which one of the important segments of the priority AI research and development directions is the development of language technology. Our research aims to provide a comprehensive overview of the development and functioning of the large language models, focusing on the specific challenges and opportunities that arise in the field of Hungarian language models. In addition, we would like to show how these models can be used effectively in education and how they can help students improve their language skills, support their studies, or even create intelligent systems for educators. https://nyelvor.mnyknt.hu/wp-content/uploads/148507.pdf |
[4]Á. Wolf, P. Zsoldos, K. Széll, and P. Galambos, “Towards robotic laboratory automation plug & play: Reference architecture model for robot integration,” SLAS TECHNOLOGY, vol. 29, no. 4, 2024.
Supportive robotic solutions take over mundane, but essential tasks from human workforce in biomedical research and development laboratories. The newest technologies in collaborative and mobile robotics enable the utilization in the human-centered and low-structured environment. Their adaptability, however, is hindered by the additional complexity that they introduce. In our paper we aim to entangle the convoluted laboratory robot integration architectures. We begin by hierarchically decomposing the laboratory workflows, and mapping the activity representations to layers and components of the automation control architecture. We elaborate the framework in detail on the example of pick-and-place labware transportation – a crucial supportive step, which we identified as the number one area of interest among experts of the field. Our concept proposal serves as a reference architecture model, the key principles of which were used in reference implementations, and are in line with international standardization efforts. https://www.sciencedirect.com/science/article/pii/S2472630324000505 |
[5]O. Shvets, B. Smakanov, G. Györök, and L. Kovács, “A Driver Fatigue Recognition System, Based on an Artificial Neural Network,” ACTA POLYTECHNICA HUNGARICA, vol. 21, no. 8, pp. 211–226, 2024.
[6]V. Vass, Z. Farkas, and G. Györök, “Environmental Effects of Alkaline Degreasing for Automotive, Boat and Machine Industry Purposes,” ACTA POLYTECHNICA HUNGARICA, vol. 21, no. 10, pp. 379–392, 2024.
[7]Simon, Gyula and G. Zachar, “Fast and Fault-Tolerant Passive Hyperbolic Localization Using Sensor Consensus,” SENSORS, vol. 24, no. 9, 2024.
The accuracy of passive hyperbolic localization applications using Time Difference of Arrival (TDOA) measurements can be severely compromised in non-line-of-sight (NLOS) situations. Consensus functions have been successfully used to provide robust and accurate location estimates in such challenging situations. In this paper, a fast branch-and-bound computational method for finding the global maximum of consensus functions is proposed and the global convergence property of the algorithm is mathematically proven. The performance of the method is illustrated by simulation experiments and real measurements. https://www.mdpi.com/1424-8220/24/9/2891 Partner: Institute for Software Integrated Systems, Vanderbilt University, Nashville, TN 37212, USA |
[8]Modne Takacs, Judit and Pogatsnik, Monika, “A Comprehensive Study on Cybersecurity Awareness: Adaptation and Validation of a Questionnaire in Hungarian Higher Technical Education,” ACTA POLYTECHNICA HUNGARICA, vol. 21, no. 10, pp. 533–552, 2024.
Background: Cybersecurity is an extremely important topic in the 21st Century, especially for students in education. It is essential for career development in technical higher education to know how to defend against digital threats and cyberattacks effectively. Education may enhance digital literacy and security awareness. To measure the success of this development it is essential to have a reliable measurement tool. Objective: This study aims to develop a Hungarian adaptation of the Cybersecurity attitude survey (CS-C), to test the psychometric properties of the survey among students of technical higher education institutions and to analyze the results. Method: The 25-item questionnaire measures cyberawareness on a 5-point Likert scale. A pilot study with 35 participants, who were retested after a few weeks, was conducted in the first round. For a more comprehensive analysis, N=398 participants in higher technical education were included in the second phase of the study. Results: The results of the psychometric analyses demonstrated the internal reliability and validity of the CS-C-H questionnaire and confirmed that it is reliable (Cronbach Alfa =.858) in its application and interpretation along dimensions of cyberspace-related attitudes, especially among students in education. Respondents’ cybersecurity awareness is at an acceptable level, but question-specific differences between groups can be found. Further research into the factors that influence the development of attitudes is, therefore, worthwhile. Conclusion: The use of this diagnostic tool among Hungarian students is justified based on the results of the study. https://acta.uni-obuda.hu/ModneTakacs_Pogatsnik_150.pdf |
[9]X. Tan, Q. Meng, F. Zhao, L. Zhang, X. Hu, and T. Jancsó, “HR-UVFormer: A top-down and multimodal hierarchical extraction approach for urban villages,” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024.
Urban villages (UVs) renovation has been incorporated into the Sustainable Development Goals (SDGs) as a result of the inequality issue among residents garnering substantial social attention. However, existing deep-learning techniques for UVs extraction have been limited to a single spatial scale (e.g., patch-level or pixel-level extraction), leading to inadequate precision and integrity in their extraction outcomes. To overcome this limitation, our study introduces HR-UVFormer, a top-down and multimodal hierarchical extraction approach that extracts UVs from a coarse scale (patch) to a fine granularity (pixel), aiming to enhance the internal completeness and boundary accuracy of the extraction results. The multimodal approach can effectively fuse multimodal features [e.g., building footprints (BFs)] with remote sensing images (RSIs) to enhance UVs extraction. The Shenzhen results indicate that the coarse-scale extraction accuracy achieves an overall accuracy (OA) of 98.79%, and the fine-grained extraction accuracy achieves a mean Intersection over Union (mIoU) of 93.60%. Furthermore, ablation experiments demonstrate a notable 7.14% improvement in mIoU with the hierarchical extraction strategy compared to the traditional pixel-based extraction strategy, and the fusion of BF and RSI yields further improvements of 2.78% and 0.65% in OA and mIoU, respectively. This finding confirms the synergistic effect between RSI and BF in UVs extraction, which has been further analyzed in this study. In addition, the proposed model outperforms other deep learning models and exhibits the potential to support more modal features (e.g., POI). Finally, the experimental dataset and code can be publicly accessed at https://github.com/q1310546582/HR-UVFormer-code https://ieeexplore.ieee.org/abstract/document/10496960?casa_token=gE9DVoI71FoAAAAA:KN2k0aS9GPyGDopvFAH5b77kIe6kFFGvuxopan4h9UbIQLRlRkNc09e5L3xZlP3cKde2nkUt9F0lZg Partner: Aerospace Information Research Institute Chinese Academy of Sciences, Beijing, China |
[10] M. Pogátsnik, P. Tóth, and K. Horváth, “The Development of Engineering Students’ Analogical Reasoning,” ACTA POLYTECHNICA HUNGARICA, vol. 21, no. 8, pp. 169–188, 2024.
Based on the Industry 4.0 strategy, problem-solving and, in this context, analogical reasoning has become crucial for successful placement and positioning in the labor market. Therefore, its development and monitoring is a priority. The goal of our research was to explore the development of analogical thinking in first-year engineering students who have completed their high school studies and are now starting their engineering studies. In the present study, 241 first-year engineering students of Óbuda University participated. The students enrolled in the BSc in Computer Engineering and BSc in Electrical Engineering have more advanced analytical skills and higher thinking speed. Cluster analysis was used to identify and characterize three groups. https://acta.uni-obuda.hu/Pogatsnik_Toth_Horvath_148.pdf |
[11]R. Liza, P. Pereyra, D. Muñoz, V. Viera, M. E. López Herrera, J. Rojas, D. Palacios, F. Díaz, N. Cerna, S. Rojas, and L. Sajo-Bohus, “Comprehensive Study of Natural Radioactivity in Building Materials: A Case Study in Ica, Peru,” ATMOSPHERE, vol. 15, no. 3, 2024.
[12] Modne Takacs, Judit and Pogatsnik Monika, “The Presence of Cybersecurity Competencies in the Engineering Education of Generation Z,” ACTA POLYTECHNICA HUNGARICA, vol. 21, no. 6, pp. 107–127, 2024.
In the context of 21st-century work in cyberspace, soft skills, and cybersecurity competencies are essential for young engineers in preparation for a career in engineering. The primary objective of this pilot study is to assess the effectiveness and level of security awareness training in the context of digital literacy education, considering the soft skills, educational experiences, and attitudes of the youth. The research uses an innovative methodology. The questionnaire-based quantitative survey is complemented by an alternative qualitative method. In addition to the measurement of attitudes supported by a focus group interview mixed with an imagery association technique, the level of cyber-competence development of the N=130 participating engineering students will be measured by a partially adapted questionnaire. The results of the research will provide insights into the level of awareness, knowledge, and ways of dealing with cyberspace threats among young engineering students, as well as highlight the gaps and strengths of education in terms of skills development. In conclusion, although young engineering students are aware of cyberspace threats, they are not well equipped to deal with them, especially in terms of password habits, security settings, and the use of online social platforms. https://acta.uni-obuda.hu/ModneTakacs_Pogatsnik_146.pdf |
[13]Q. Meng, J. Gao, L. Zhang, X. Hu, J. Qian, and T. Jancsó, “Coupled cooling effects between urban parks and surrounding building morphologies based on the microclimate evaluation framework integrating remote sensing data,” SUSTAINABLE CITIES AND SOCIETY, vol. 102, pp. 1–12, 2024.
Urban parks alter regional microclimates and considerably mitigate the urban thermal environment. However, studies on the coupled cooling effects of parks and surrounding building morphologies are limited, and previous microclimate simulations are scarcely generalizable to cities. Therefore, based on an improved microclimate framework integrating remote sensing data with ENVI-met model, this study quantified the spatiotemporal cooling effect and energy saving potential generated by a representative park and surrounding building morphologies during a summer day in Beijing, China. The results showed that the morphologies of the surrounding buildings notably affected the cooling effect. The cooling effect affects the microclimate above the average building height. Additionally, considering the building morphological attributes surrounding all parks in the central urban area of Beijing, the spatial distribution of the cooling effect was not strongly and strongly influenced by the building density and average height, respectively; whereas, the accumulated energy saving potential had the opposite effect. Energy savings at different heights varied with surrounding building morphologies and exhibited a significant non-linear negative correlation with height. This study highlights the practical implications of integrating remote sensing data with microclimate evaluation framework and provides new perspectives and insights into urban thermal mitigation strategies. https://www.sciencedirect.com/science/article/pii/S2210670724000647?casa_token=2S-sslRU8LYAAAAA:IAGM6XThny2OKZZOmLEJoD2RmD-YCAZR9Tc_KlWGhWcLKdtY-t0IgVixwqbmCkVZFaDdyIbMIbmP Partner: Aerospace Information Research Institute Chinese Academy of Sciences, Beijing, China |
[14]J. Qian, L. Zhang, U. Schlink, Q. Meng, X. Liu, and T. Jancsó, “High spatial and temporal resolution multi-source anthropogenic heat estimation for China,” RESOURCES CONSERVATION AND RECYCLING, vol. 203, 2024.
Anthropogenic heat (AH) emissions have rapidly increased in recent decades and are now critical for studying urban thermal environments; however, AH datasets composed of multiple heat sources with fine and accurate spatiotemporal characteristics at large scales are lacking. This study obtained annual, monthly, and hourly AH of multiple heat sources in China for 2019 at 500 m resolution. We first corrected the top-down inventory method for China, which is based on official energy consumption data. Then, we considered features such as the national building height, weighted factory density, and weighted road density to better represent the spatial characteristics of multi-source AH. Based on the above data preparation, a stacking framework was employed to integrate multiple machine-learning algorithms to construct an efficient and accurate AH estimation model. Finally, besides the comparative validation, the results were further tested by participating in a short-term climate numerical simulation for both winter and summer. The resulting data showed a reasonable AH composition and the total amount and composition of AH varied notably from region to region. The spatial and temporal characteristics of the AH from different sources differed greatly and were more detailed and accurate than those reported in previous studies. Air temperature simulations in winter were improved by the AH dataset input, but the uncertainties of climate simulations also limit its validity in AH validation. Because of its large spatial extent and detailed spatiotemporal characteristics, the new dataset strongly supports urban climate research and sustainable development. https://www.sciencedirect.com/science/article/pii/S0921344924000454?casa_token=WD-g4EgSxM4AAAAA:UpJWZ_BMSpTuqwuJ1mLtWppetOBz2sueI8VMW6y4VsMhyjKU4oJaRaK3EmmU3OcIQ0r8yAvF1LpG Partner: Aerospace Information Research Institute Chinese Academy of Sciences, Beijing, China |