Találati lista | Tudóstér

001-es BibID:	BIBFORM135421
Első szerző:	Girászi Tamás (informatikus)
Cím:	Enhancing Password Guessing Attacks with GAN-Derived Rule Generation / Tamás Girászi, Norbert Oláh, Andrea Huszti
Dátum:	2025
Megjegyzések:	Many security incidents can be traced back to insecure user authentication mechanisms, which can compromise the confidentiality and integrity of sensitive data. Various alternative authentication methods have been proposed, but the password-based authentication systems remain widely used due to their simplicity and ease of implementation. Recent advances in deep learning, particularly Generative Adversarial Networks (GANs), have introduced new possibilities for password guessing attacks by learning and replicating statistical patterns in real-world password datasets. However, while GANs are capable of generating realistic and novel password candidates, they require substantial computational resources and storage to match the efficiency of rule-based approaches. In this paper, we propose a hybrid method that leverages GANs to generate rulesets, which are then used within traditional rule-based attack frameworks. Our approach combines the generalization capabilities of GANs with the efficiency and interpretability of rule-based attacks. We evaluate multiple state-of-the-art GAN-based models and generate compact rulesets from their outputs using Levenshtein-distance-based transformations. Experimental results on the RockYou dataset demonstrate that the derived rulesets not only outperform the GANs themselves in efficiency but also surpass traditional handmade rulesets. The proposed method is fully automated, requires no expert knowledge, and is suitable for low-resource attack scenarios.
ISBN:	9798331514358
Tárgyszavak:	Természettudományok Matematika- és számítástudományok tanulmány, értekezés könyvrészlet Deep learning Humanities Neural networks Authentication Passwords Generative adversarial networks Hybrid power systems Computational efficiency Security
Megjelenés:	Proceedings of the 2025 IEEE International Conference on Cyber Humanities (IEEE-CH). - 1 (2025), p. 99-105. -
További szerzők:	Oláh Norbert (1991-) (gazdaságinformatikus) Huszti Andrea (1975-) (informatikus)
Pályázati támogatás:	EKOP-25-3-I Egyéb
Internet cím:	Szerző által megadott URL DOI Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon:

001-es BibID:	BIBFORM120420
035-os BibID:	(Scopus)85191151631
Első szerző:	Huszti Andrea (informatikus)
Cím:	Blockchain-Based Messaging for VANETs / Andrea Huszti, Tamás Girászi, Norbert Oláh
Dátum:	2023
Megjegyzések:	An intelligent transport system is indispensable for today`s life. It can help to reduce traffic and even the number of accidents. However, it raises several security issues. Providing travelers` privacy and anonymity is essential. We propose a message broadcast protocol for vehicle communication based on the blockchain technology. Eligible vehicles can anonymously report road conditions (e.g. traffic jams, accidents, etc.). We give an efficient solution applying Solana blockchain that provides fast block generation time and supports smart contracts. Blockchain maintains immutability of messages, enables anonymous message submission, moreover assures message unlinkability. Our scheme is identity-based, applying bilinear pairings batch verification of messages is accomplished.
ISBN:	979-8-3503-2760-1
Tárgyszavak:	Természettudományok Matematika- és számítástudományok konferenciacikk folyóiratcikk Identity-Based Cryptography blockchain VANET incentive-punishment mechanism Solana
Megjelenés:	IEEE. - 1 (2023), p. 2443-2450. -
További szerzők:	Girászi Tamás (1999-) (informatikus) Oláh Norbert (1991-) (gazdaságinformatikus)
Pályázati támogatás:	TKP2020-NKA-04 Egyéb
Internet cím:	Intézményi repozitóriumban (DEA) tárolt változat DOI Szerző által megadott URL
Borító:
Saját polcon: