Mesterséges intelligencia alapú polgári védelmi lakossági tájékoztató és felkészítő rendszer keretfeltételeinek vizsgálata a digitális generációk válságkommunikációjában
Abstract
A tanulmány egy helyi telepítésű (on-premise), mesterséges intelligencia alapú polgári védelmi lakosság tájékoztató és felkészítő rendszer megvalósíthatósági keretfeltételeit vizsgálja, különös tekintettel a digitálisan aktív Y és Z generációk specifikus válságkommunikációs igényeire, valamint az online oktatási modellek tesztkörnyezeti implementációjára. A rendszer illeszkedik a Magyarország Mesterséges Intelligencia Stratégiája 2025–2030 célkitűzéseihez, amelyek az AI-alkalmazások közszolgálati és katasztrófavédelmi kiterjesztését prioritásként jelölik meg, ugyanakkor a tudomány és a technológia oly gyorsan fejlődik, hogy a stratégia egyes elemei már a kidolgozásuk pillanatában is folyamatos frissítést igényelnek.
References
BM OKF Scientific Council for Disaster Management (2025). New perspectives for population preparedness using artificial intelligence [Prize-winning study summary]. Ministry of Interior, National Directorate General for Disaster Management. [Online]. Available: https://www.katasztrofavedelem.hu/application/uploads/documents/2026-02/86850.pdf (2026.09.10.)
Kiss, G. & Kozma, G. (2024). Communication habits of digital generations in the workplace. Jel-Kép: Kommunikáció, Közvélemény, Média, 2024(1). [Online]. Available: https://www.jelkep.eu/pdf/2024/1/JelKep_2024_1__Kiss_Kozma_Georgina.pdf (2026.09.10.)
Petruska, F. (2025). Connection areas between artificial intelligence and national defence, with particular attention to regulation. Defence-Security Regulatory and Governance Studies, 2025(10). [Online]. Available: https://hhk.uni-nke.hu/document/hhk-uni-nke-hu/VBSZK_MT_2025_10 (2026.09.10.)
DRIVER+ Consortium (2019). DRIVER+ project in a nutshell. European Commission, 7th Framework Programme. [Online]. Available: https://www.driver-project.eu/driver-project/(2026.09.10.)
EC Research and Innovation (2021). The ultimate platform to handle extreme weather events [beAWARE project]. European Commission. [Online]. Available: https://projects.research-and-innovation.ec.europa.eu/en/projects/success-stories/all/ultimate-platform-handle-extreme-weather-events (2026.09.10.)
SecurIT (2023). AI Disaster Emergency Com. SecurIT Project, European Commission. [Online]. Available: https://securit-project.eu/funded-projects/ai-disaster-emergency-com/ (2026.09.10.)
Kis, M., Bódi, A. & Számadó, R. (2025). The process and risks of domestic implementation of NIS2. Hadmérnök [Military Engineering], 19(3), 165–182. [Online]. Available: https://doi.org/10.32567/hm.2024.3.10 (2026.09.10.)
Kirovné Rácz, R. & Scholtz, E. (2025). Application of artificial intelligence and machine learning algorithms in defence against hydrological disasters. Hadmérnök [Military Engineering], 20(4), 125–147. [Online]. Available: https://doi.org/10.32567/hm.2025.4.8 (2026.09.10.)
Vásárhelyi, Ö. (2024). The future of Hungary's cybersecurity in the light of the EU NIS2 Directive. Nemzetbiztonsági Szemle [National Security Review], 12(3), 18–35. [Online]. Available: https://doi.org/10.32561/nsz.2024.3.2 (2026.09.10.)
Guo, S. et al. (2023). Ensemble Reinforcement Learning: A Survey. arXiv:2303.02618. [Online]. Available: https://arxiv.org/abs/2303.02618 (2026.09.10.)
Sutton, R.S. & Barto, A.G. (2018). Reinforcement Learning: An Introduction (2nd ed.). MIT Press. ISBN: 978-0262039246. [Online]. Available: https://web.stanford.edu/class/psych209/Readings/SuttonBartoIPRLBook2ndEd.pdf (2026.09.10.)
Alqithami, S. (2025). Leveraging Public Sentiment for Resource Coordination in Disaster Response: A Multiagent Framework. arXiv:2505.03979. [Online]. Available: https://arxiv.org/pdf/2505.03979 (2026.09.10.)
Mnih, V. et al. (2015). Human-level control through deep reinforcement learning. Nature, 518, 529–533. [Online]. Available: https://doi.org/10.1038/nature14236 (2026.09.10.)
Kennedy, J. & Eberhart, R. (1995). Particle swarm optimization. Proceedings of IEEE International Conference on Neural Networks, 4, 1942–1948. [Online]. Available: https://doi.org/10.1109/ICNN.1995.488968 (2026.09.10.)
Guan, M., Feng, Z., Jiang, S. & Zhou, W. (2024). Tethered Balloon Cluster Deployments and Optimization for Emergency Communication Networks. Entropy, 26(12), 1071. [Online]. Available: https://doi.org/10.3390/e26121071 (2026.09.10.)
Dorigo, M. & Stützle, T. (2004). Ant Colony Optimization. MIT Press. ISBN: 978-0262042192. [Online]. Available: https://mitpress.mit.edu/9780262042192/ant-colony-optimization/(2026.09.10.)
Muhammad, A. & Wu, T. (2023). Examining the evacuation routes of the sister village program by using the ant colony optimization algorithm. Open Geosciences, 15(1), 20220512. [Online]. Available: https://doi.org/10.1515/geo-2022-0512 (2026.09.10.)
Brassai, S.T. (2019). Neurális hálózatok és fuzzy logika [Neural Networks and Fuzzy Logic]. Kolozsvár: Scientia Kiadó. ISBN: 978-606-975-007-1
Riano-Briceno, G. et al. (2013). Comparison of ANFIS and ANN model to categorize patients in the emergency department. PMC/NCBI. [Online]. Available: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3776083/(2026.09.10.)
ScienceDirect (2025). Advancements in data-driven evolving fuzzy and neuro-fuzzy control: A comprehensive survey. Engineering Applications of Artificial Intelligence. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0952197625000739 (2026.09.10.)
Liu, Y., Zhang, K., Li, Y. et al. (2024). Towards trustworthy large language models for disaster management. Patterns, 5(6), 100940. [Online]. Available: https://doi.org/10.1016/j.patter.2024.100940 (2026.09.10.)
Xia, Y., Dittler, D., Jazdi, N., Chen, H. & Weyrich, M. (2024). LLM experiments with simulation: Large language model multi-agent system for simulation model parametrization in digital twins. IEEE ETFA 2024. [Online]. Available: https://doi.org/10.1109/ETFA61700.2024.10710961 (2026.09.10.)
Xie, Z. et al. (2025). Towards Intelligent Emergency Management: A Scenario-Learning-Decision Framework Enabled by Large Language Models. Mathematics, MDPI. [Online]. Available: https://doi.org/10.3390/math13010000 (2026.09.10.)
Al-Shamrani, S. & Hasan, R. (2025). CrisisGuideGPT: An Agentic-AI for Crisis Response Planning. 2025 IEEE World Forum on Public Safety Technology (WF-PST), 200–205. [Online]. Available: https://doi.org/10.1109/WF-PST65083.2025.00037 (2026.09.10.)
Yang, L. et al. (2025). Leveraging Large Language Models for Enhanced Digital Twin Modeling. arXiv:2503.02167. [Online]. Available: https://arxiv.org/abs/2503.02167 (2026.09.10.)
Sudanthiran, A. et al. (2025). Disaster Management in the Era of Agentic AI Systems: A Vision for the Future. arXiv:2510.16034. [Online]. Available: https://arxiv.org/abs/2510.16034 (2026.09.10.)
Harang, P. (2024). How efficient are LLM tokenisers on Hungarian text? Medium. [Online]. Available: https://medium.com/@harangpeter/mennyire-hat%C3%A9kony-az-llm-ek-tokeniz%C3%A1l%C3%B3ja-magyar-sz%C3%B6vegen-ce8481e29f93 (2026.09.10.)
Farquhar, S., Kossen, J., Kuhn, L. & Gal, Y. (2024). Detecting hallucinations in large language models using semantic entropy. Nature, 630, 625–630. [Online]. Available: https://doi.org/10.1038/s41586-024-07421-0 (2026.09.10.)
Xie, Z. & Sun, Z. (2025). Mitigating Hallucination in RAG via Ensemble Voting [VOTE-RAG]. arXiv:2603.27253. [Online]. Available: https://arxiv.org/pdf/2603.27253 (2026.09.10.)
George Mason University (2024). Transforming emergency response training through AI and interactive games. [Online]. Available: https://www.gmu.edu/news/2024-12/transforming-emergency-response-training-through-ai-and-interactive-games (2026.09.10.)
Gurcan, O. (2025). LLM-augmented agent-based modelling for social simulations: Challenges and opportunities. arXiv:2405.06700. [Online]. Available: https://arxiv.org/abs/2405.06700 (2026.09.10.)
Copyright (c) 2026 Defence Science
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
