Author:
Author:
Hedvig Szabó
szabohedv@gmail.com
https://orcid.org/0009-0000-8403-8943
Hedvig Szabó
Nemzeti Közszolgálati Egyetem, Budapest , Magyarország
Search for other papers by Hedvig Szabó in
Current site
Google Scholar
PubMed
Search for other papers by Hedvig Szabó in
Current site
Google Scholar
PubMed
A tanulmány célja a mesterséges intelligencia lehetőségeinek vizsgálata a megtévesztés felismerésében, valamint az alkalmazás jogi megfelelőségének értékelése. Ennek érdekében elvégeztük három adatbázis (MU3D, Bag-of-Lies, DOLOS) elemzését gépi tanulási modellek segítségével, majd a felderítésben való alkalmazás jogi értékelését. Az eredmények alapján megállapítható, hogy a kombinált MI-technológiák javítják a csalásdetekció hatékonyságát, de a pontosság még mindig messze van az optimális szinttől. Az MI-alapú megtévesztésfelismerés alkalmazható a magyar jogrendszerben, ha megfelel az AI Act előírásainak. A tanulmány bemutatja az MI-alapú csalásdetekció technológiai és jogi aspektusait, hangsúlyozva a kombinált technológiák hatékonyságát és a jogi megfelelőség fontosságát a felderítésben.
Objective. The aim of the study is to examine the potential of artificial intelligence (AI) in deception detection from an interdisciplinary perspective and to assess the legal compliance of AI-based fraud detection systems. For this purpose, we analyzed three databases (MU3D, Bag-of-Lies, DOLOS) using machine learning models and conducted a legal evaluation of their application in law enforcement based on the Interpol Responsible AI Innovation in Action Workbook.
Methodology. The research involved an extensive analysis of three datasets (MU3D, Bag-of-Lies, DOLOS) using various machine learning models. These datasets included multiple modalities such as video, audio, EEG, and eye-tracking data. The objective was to evaluate how effectively AI can detect deception when these different types of data are combined. Following the technical analysis, a thorough legal assessment was conducted to understand the regulatory framework for deploying AI-based deception detection systems in law enforcement, particularly focusing on compliance with the AI Act.
Findings. The results of the study indicate that combining multiple modalities significantly enhances the accuracy of fraud detection. The use of the PAVF (Plug-in Audio-Visual Fusion) module and multi-task learning approaches showed superior performance compared to single modality methods. Additionally, the implementation of the UT-Adapter proved effective in mitigating the risk of overfitting, which is a common issue in machine learning models. Despite these advancements, the accuracy of the combined AI technologies still falls short of the optimal level required for reliable use in law enforcement. From a legal standpoint, the study found that AI-based deception detection systems can be implemented within the Hungarian legal framework, provided they comply with the requirements set forth in the AI Act. These systems are categorized as high-risk AI systems under the AI Act, necessitating stringent measures to ensure legality, proportionality, and necessity. This includes ensuring human oversight and the protection of fundamental rights. Law enforcement agencies must ensure the lawful handling of data and maintain transparency in the operations of deception detection systems, including the ability to contest decisions and access remedies.
Conclusion. The study provides a comprehensive overview of the technological and legal aspects of AI-based fraud detection. It highlights that while combined AI technologies significantly improve the effectiveness of deception detection, there is still room for improvement in terms of accuracy. The legal analysis emphasizes the importance of compliance with regulatory standards and the protection of fundamental rights in the deployment of these systems. The findings offer valuable guidance on how law enforcement agencies can ethically and legally implement AI-based systems for deception detection, ensuring that they align with both technological advancements and legal requirements.
ProCite
RefWorks
Reference Manager
BibTeX
Zotero
EndNote
-
Abdi, H. (2007) Signal detection theory (SDT). In: Salkind, E. (ed.) Encyclopedia of measurement and statistics. Thousand Oaks, Sage. pp. 886–889.
-
Bond, Ch. F. Jr. & DePaulo, B. M. (2006) Accuracy of deception judgments. Personality and Social Psychology Review, Vol. 10. No. 3. pp. 214–234.
-
Bonet-Solà, D. & Alsina-Pagès, R. M. (2021) A comparative survey of feature extraction and machine learning methods in diverse acoustic environments. Sensors, Vol. 21. No. 4, 1274.
-
Budaházi Á., Fantoly Zs., Kakuszi B., Bitter I. & Czobor P. (2021) A műszeres vallomás-ellenőrzés fejlődési irányai. Budapest, Ludovika Egyetemi Kiadó.
-
DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K. & Cooper, H. (2003) Cues to deception. Psychological Bulletin, Vol. 129. No. 1. pp. 74–118.
-
Docan-Morgan, T. (2007) Training law enforcement officers to detect deception: A critique of previous research and framework for the future. Applied Psychology in Criminal Justice, Vol. 3. No. 2. pp. 143–171.
-
Frank, G. M. & Ekman, P. (1997) The ability to detect deceit generalizes across different types of high-stake lies. Journal of Personality and Social Psychology, Vol. 72. No. 6. pp. 1429–1439.
-
Gupta, V., Agarwal, M., Arora, M., Chakraborty, T., Singh, R. & Vatsa, M. (2019) Bag-of-lies: A multimodal dataset for deception detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA.
-
LaGrandeur, K. (2023) The consequences of AI hype. AI Ethics, Vol. 4. pp. 653–656.
-
Levine, R. T., Blair, J. P. & Clare, D. D. (2014) Diagnostic utility: Experimental demonstrations and replications of powerful question effects in high-stakes deception detection. Human Communication Research, Vol. 40. No. 2. pp. 262–289.
-
Lloyd, E. P., Deska, J. C., Hugenberg, K., McConnell, R. A., Humphrey, B. & Kunstman, W. J. (2019) Miami University deception detection database. Behavior Research Methods, Vol. 51. No. 1. pp. 429–439.
-
O’Sullivan, M. (1991) Who can catch a liar? American Psychologist, Vol. 46. No. 9. pp. 913–920.
-
Pérez-Rosas, V. & Mihalcea, R. (2015) Experiments in Open Domain Deception Detection. In: Màrquez, L., Callison-Burch, Ch. & Su, J. (eds.) Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Lisbon, Portugal, Association for Computational Linguistics. pp. 1120–1125.
-
Qian, C., Vanman, E. J., Dongtao, W., Wenjing, Y., Lei, J. & Qinglin, Z. (2014) Detection of deception based on fMRI activation patterns underlying the production of a deceptive response and receiving feedback about the success of the deception after a mock murder crime. Social Cognitive and Affective Neuroscience, Vol. 9. No. 10. pp. 1472–1480.
-
Russell, S. & Bohannon, J. (2015) Artificial intelligence. Fears of an AI pioneer. Science, Vol. 349. No. 6245.
-
Sarker, I. H. (2021) Machine Learning: Algorithms, Real-World Applications and Research Directions. SN Computer Science, Vol. 2. No. 160.
-
Saxe, L. (1991) Science and the CQT polygraph: A theoretical critique. Integrative Physiological and Behavioral Science, Vol. 26. No. 3. pp. 223–231.
-
Sporer, S. L. (1997) The less travelled road to truth: Verbal cues in deception detection in accounts of fabricated and self-experienced events. Applied Cognitive Psychology, Vol. 11. No. 5. pp. 373–397.
-
Xiaobao, G., Nithish, M. S., Zitong, Y., Adams Wai-Kin, K., Bing-quan, S. & Alex, K. (2023) Audio-Visual Deception Detection: DOLOS Dataset and Parameter-Efficient Crossmodal Learning. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, Paris, France.
-
Zhang, Z., Singh, V., Slowe, T. E., Tulyakov, S. & Govindaraju, V. (2007) Real-time automatic deceit detection from involuntary facial expressions. In: IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN.
Editor-in-Chief:
- József BOKOR (Computer and Automation Research Institute)
Founding Editor-in-Chief:
- Tamás NÉMETH
Managing Editor:
- István SABJANICS (Ministry of Interior, Budapest, Hungary)
Editorial Board:
- Attila ASZÓDI (Budapest University of Technology and Economics)
- Zoltán BIRKNER (University of Pannonia)
- Valéria CSÉPE (Research Centre for Natural Sciences, Brain Imaging Centre)
- Gergely DELI (University of Public Service)
- Tamás DEZSŐ (Migration Research Institute)
- Imre DOBÁK (University of Public Service)
- Marcell Gyula GÁSPÁR (University of Miskolc)
- József HALLER (University of Public Service)
- Charaf HASSAN (Budapest University of Technology and Economics)
- Zoltán GYŐRI (Hungaricum Committee)
- János JÓZSA (Budapest University of Technology and Economics)
- András KOLTAY (National Media and Infocommunications Authority)
- Gábor KOVÁCS (University of Public Service)
- Levente KOVÁCS (Óbuda University)
- Melinda KOVÁCS (Hungarian University of Agriculture and Life Sciences (MATE))
- Miklós MARÓTH (Avicenna Institue of Middle Eastern Studies )
- Judit MÓGOR (Ministry of Interior National Directorate General for Disaster Management)
- József PALLO (University of Public Service)
- István SABJANICS (Ministry of Interior)
- Péter SZABÓ (Hungarian University of Agriculture and Life Sciences (MATE))
- Miklós SZÓCSKA (Semmelweis University)
Ministry of Interior
Science Strategy and Coordination Department
Address: H-2090 Remeteszőlős, Nagykovácsi út 3.
Phone: (+36 26) 795 906
E-mail: scietsec@bm.gov.hu
DOAJ
| 2023 | |
|---|---|
| CrossRef Documents | 32 |
| CrossRef Cites | 15 |
| Days from submission to acceptance | 59 |
| Days from acceptance to publication | 104 |
| Acceptance Rate | 81% |
| 2022 | |
|---|---|
| CrossRef Documents | 38 |
| CrossRef Cites | 10 |
| Days from submission to acceptance | 54 |
| Days from acceptance to publication | 78 |
| Acceptance Rate | 84% |
| 2021 | |
|---|---|
| CrossRef Documents | 46 |
| CrossRef Cites | 0 |
| Days from submission to acceptance | 33 |
| Days from acceptance to publication | 85 |
| Acceptance Rate | 93% |
| 2020 | |
|---|---|
| CrossRef Documents | 13 |
| CrossRef Cites | 0 |
| Days from submission to acceptance | 30 |
| Days from acceptance to publication | 62 |
| Acceptance Rate | 93% |
| Publication Model | Gold Open Access |
| Submission Fee | none |
| Article Processing Charge | none |
| Scientia et Securitas | |
|---|---|
| Language | Hungarian English |
| Size | A4 |
| Year of Foundation |
2020 |
| Volumes per Year |
1 |
| Issues per Year |
4 |
| Founder | Academic Council of Home Affairs and Association of Hungarian PhD and DLA Candidates |
| Founder's Address |
H-2090 Remeteszőlős, Hungary, Nagykovácsi út 3. H-1055 Budapest, Hungary Falk Miksa utca 1. |
| Publisher | Akadémiai Kiadó |
| Publisher's Address |
H-1117 Budapest, Hungary 1516 Budapest, PO Box 245. |
| Responsible Publisher |
Chief Executive Officer, Akadémiai Kiadó |
| Applied Licenses |
CC-BY 4.0 CC-BY-NC 4.0 |
| ISSN | ISSN 2732-2688 (online), 3057-9759 (print) |
Monthly Content Usage
| Abstract Views | Full Text Views | PDF Downloads | |
|---|---|---|---|
| Aug 2024 | 0 | 0 | 0 |
| Sep 2024 | 0 | 0 | 0 |
| Oct 2024 | 0 | 0 | 0 |
| Nov 2024 | 0 | 0 | 0 |
| Dec 2024 | 0 | 15616 | 42 |
| Jan 2025 | 0 | 28566 | 29 |
| Feb 2025 | 0 | 0 | 0 |
Author:
Author:
Author:
Copyright Akadémiai Kiadó AKJournals is the trademark of Akadémiai Kiadó's journal publishing business branch.
Powered by PubFactory