Neural Networks in Forensic Expertology and Expert Practice: Problems and Prospects

The article, from the perspective of the theory the forensic activity digitalization as a particular theory of forensic expertise, examines the prospects for the introduction of neural networks in forensic examination and the current problems arising in this case. The author notes changes in the methodology and technologies for developing expert techniques in connection with the introduction of artificial intelligence algorithms — neural networks. The areas neural networks’ application for solving various problems of science and practical activity are outlined. Specific examples demonstrate the possibilities of using supervised learning algorithms for neural networks in forensic practice. A detailed analysis of the reasons why the use of neural networks in forensic science can lead to erroneous conclusions is given. Particular attention is paid to hallucinations of deep learning neural networks on large language models. There is a danger that an expert, relying entirely on a neural network, may give the wrong conclusion, since self-learning generative neural networks do not provide an explanation for why they made a particular decision. To develop expert methods for solving typical expert problems based on neural networks, it is proposed to create databases (Dataset) for various forensic objects for analysis and machine learning. To store the Dataset, it is necessary to organize repositories that can contain data sets on types (kinds) of forensic examinations. Dataset and repositories will provide data quality control and model verification. The article substantiates the need for new competencies of a Data Scientist — a specialist who develops tools for solving forensic problems when introducing neural networks and other artificial intelligence algorithms into forensic science, as well as a machine learning engineer working in contact with him.

1. Аверьянова Т. В. Судебная экспертиза : курс общей теории. — М. : Норма, 2006. — 480 с.

2. Бахтеев Д. В. Особенности распознавания подлога подписи человеком как первичные критерии для разработки системы искусственного интеллекта // Сибирское юридическое обозрение. — 2020. — № 17 (4). — С. 514—522.

3. Бахтизин А. Р. , Брагин А. В. , Макаров В. Л. Большие языковые модели четвертого поколения как новый инструмент в научной работе // Искусственные общества. — 2023. — T. 18. — Вып. 1. — URL: https://artsoc.jes.su/s207751800025046-9-1 (дата обращения: 02.01.2024).

4. Россинская Е. Р. Концепция частной теории цифровизации судебно-экспертной деятельности // Вестник экономической безопасности. — 2022. — № 5. — С. 173—176.

5. Россинская Е. Р., Галяшина Е. И., Зинин А. М. Теория судебной экспертизы (Судебная экспертология) : учебник / под ред. Е. Р. Россинской. — М. : Норма ; Инфра-М, 2020. — 268 с.

6. Теория информационно-компьютерного обеспечения криминалистической деятельности : монография / под ред. Е. Р. Россинской. — М. : Проспект, 2022. — 256 с.

7. Федоренко В. А., Сорокина К. О., Гиверц П. В. Классификация изображений следов бойков по экземплярам оружия с помощью полносвязной нейронной сети // Известия Саратовского университета. — Новая серия. — Серия : Экономика. Управление. Право. — 2022. — Т. 22. — Вып. 2. — С. 184—190.

8. Чеснокова Е. В., Усов А. И., Омельянюк Г. Г., Никулина М. В. Искусственный интеллект в судебной экспертологии // Теория и практика судебной экспертизы. — 2023. — Т. 18. — № 3. — С. 60—77.

9. Giverts P., Sorokina K., Fedorenko V. Examination of the possibility to use Siamese networks for the comparison of firing marks // Journal of Forensic Sciences. — 2022 Nov. — 67(6). — Р. 2416-2424. — DOI: 10.1111/1556-4029.15143.

10. Humberto Ferreira, Pedro Ruivo, Carolina Reisb. How do data scientists and managers influence machine learning value creation // Procedia Computer Science. — 2021. — Vol. 181. — P. 757—764.

11. Minseok Yoon, Seung-Hun Nam, In-Jae Yu, Wonhyuk Ahn, Myung-Joon Kwon, Heung-Kyu Lee. Framerate up-conversion detection based on convolutional neural network for learning spatiotemporal features // Forensic Science International. — Vol. 340. — Nov. 2022. — 111442.

12. Pre-trained models for natural language processing: A survey / Qiu X. [et al.] // Science China Technological Sciences. — 2020. — Vol. 63. — № 10.

13. Samuel Henrique Silva, Mazal Bethany, Alexis Megan Votto, Ian Henry Scarff, Nicole Beebe, Peyman Najafirad. Deepfake forensics analysis: An explainable hierarchical ensemble of weakly supervised models // Forensic Science International: Synergy. — 2022. — Vol. 4. —100217.

14. Sonam Bhardwaj, Mayank Dave. Enhanced neural network-based attack investigation framework for network forensics: Identification, detection, and analysis of the attack // Computers & Security. — Vol. 135. — December 2023. — 103521.

15. Wei Liu, Zhiqiang Dun. D-Fi: Domain adversarial neural network based CSI fingerprint indoor localization // Journal of Information and Intelligence. — Vol. 1. — Is. 2. — July 2023. — P. 104—114.

16. Wuyang Shan, Yaohua Yi, Ronggang Huang, Yong Xie. Robust contrast enhancement forensics based on convolutional neural networks // Signal Processing: Image Communication. — Vol. 71. — February 2019. — P. 138—146.

17. Yuxi Xie, C. T. Wu, Boyuan Li, Xuan Hu, Shaofan Li. Computer Methods in Applied Mechanics and Engineering. A feed-forwarded neural network-based variational Bayesian learning approach for forensic analysis of traffic accident // Computer Methods in Applied Mechanics and Engineering. — Vol. 397. — 1 July 2022. — 115148.

18. Yuxue Zhang, Yunfeng Yan, Guorui Feng. Feature compensation network based on non-uniform quantization of channels for digital image global manipulation forensics // Signal Processing: Image Communication. — Vol. 107. — September 2022. — 116795.