METHODS FOR DETECTING AND COMBATING DISINFORMATION USING ARTIFICIAL INTELLIGENCE TECHNOLOGIES
DOI:
https://doi.org/10.32689/maup.it.2025.2.8Keywords:
disinformation, artificial intelligence, natural language processing, machine learning, fact-checking, manipulative content, deepfake, cybersecurityAbstract
The article examines modern methods for detecting and combating disinformation using artificial intelligence technologies. The primary focus is on automating the analysis of textual, visual, and video information through the use of machine learning algorithms, natural language processing, and computer vision. Tools for monitoring social networks, fact-checking, and identifying manipulations in media content are discussed. Challenges related to ethical aspects, data scarcity, and the development of technologies for creating fake content are also analyzed. The proposed approaches aim to enhance the effectiveness of combating disinformation, ensure information security, and increase societal resilience to manipulation.Purpose. To study modern methods for detecting and combating disinformation using artificial intelligence technologies, particularly machine learning, natural language processing, and computer vision, to enhance societal information security.Methodology. The study employs machine learning algorithms, such as convolutional neural networks (CNN) and transformer models (e.g., BERT), for textual content analysis. Computer vision technologies are applied to detect manipulations in visual content, while graph neural networks (GNN) are used to analyze the dissemination of disinformation in social networks.Scientific novelty. An integrated approach is proposed, combining the analysis of textual, visual, and video content simultaneously. The study investigates the impact of emotional coloring on the perception of fake news and explores the role of multimodal systems in combating disinformation for the first time.Conclusions. Artificial intelligence technologies provide effective disinformation detection by analyzing large volumes of data in real-time. The integration of multimodal approaches enhances the resilience of the information space, though challenges such as the lack of high-quality training data and ethical considerations remain.
References
Фільтруйте недоброчесних блогерів, дипфейки, маніпуляції – користуйтесь інтернетом з розумом. Хмарочос. 2023. 4 грудня. URL: https://hmarochos.kiev.ua/2023/12/04/filtrujte-nedobrochesnyh-blogeriv-dipfejky-manipulyacziyi-korystujtes-internetom-z-rozumom/ (дата звернення: 12.06.2025).
Anggrainingsih R., Hassan G. M., Datta A. Evaluating BERT-based Pre-training Language Models for Detecting Misinformation. 2022. URL: https://arxiv.org/abs/2209.13594 (дата звернення: 12.06.2025).
Battiato S. DeepFake Detection by Analyzing Convolutional Traces. Journal of Imaging. 2020. Vol. 6. No. 7. P. 1–14.
Dolhansky B., Bitton J., Pflaum B., Lu J., Howes R., Wang M., Canton Ferrer C. The DeepFake Detection Challenge. arXiv preprint. 2020. arXiv:2006.07397.
Ge X., Zhang M., Wang X. A., Liu J., Wei B. Emotion-Driven Interpretable Fake News Detection. International Journal of Data Warehousing and Mining. 2022. Vol. 18. No. 3. P. 1–15.
Goldani M. H., Momtazi S., Safabakhsh R. Detecting Fake News with Capsule Neural Networks. Computational Intelligence and Neuroscience. 2020. Vol. 2020. Article ID 2165391.
Hassan N., Arslan F., Li C., Tremayne M. Toward Automated Fact-Checking: Detecting Check-worthy Factual Claims by ClaimBuster. Proceedings of the 23rd ACM International Conference on Information and Knowledge Management (CIKM 2017), Singapore, 6–10 Nov. 2017. P. 2179–2182.
Hoaxy – офіційний сайт додатку. URL: https://hoaxy.osome.iu.edu/ (дата звернення: 12.06.2025).
Mazurczyk W., Lee D., Vlachos A. Disinformation 2.0 in the Age of AI: A Cybersecurity Perspective. IEEE Security & Privacy. 2023. Vol. 21. No. 3. P. 40–49.
Meta AI. Deepfake Detection Challenge Dataset. 2020. URL: https://ai.facebook.com/blog/deepfake-detection-challenge/ (дата звернення: 12.06.2025).
Mohtaj Salar, Ata Nizamoglu, Premtim Sahitaj, Vera Schmitt, Charlott Jakob, Sebastian Möller (2024). NewsPolyML: Multi-lingual European News Fake Assessment Dataset
Rae J., Irving G., Weidinger L. Language Modelling at Scale: Gopher, Ethical Considerations, and Retrieval. DeepMind Blog. 2021. URL: https://deepmind.google/discover/blog/language-modelling-at-scale-gopher-ethical-considerations- and-retrieval/ (дата звернення: 12.06.2025).
Shahi G. K., Nandini D. FakeCovid – A Multilingual Cross-domain Fact Check News Dataset for COVID-19. arXiv preprint. 2020. arXiv:2011.11459.
Wang T., Liao X., Chow K. P., Lin X., Wang Y. Deepfake Detection: A Comprehensive Survey from the Reliability Perspective. Journal of Cybersecurity. 2024. Vol. 12. No. 1. P. 15–36.
Yang Y., Zheng L., Zhang J., Cui Q., Li Z., Yu P. S. TI-CNN: Convolutional Neural Networks for Fake News Detection. arXiv preprint. 2018. arXiv:1806.00749.
Zhang X., Cao J., Li X., Sheng Q., Zhong L., Shu K. Mining Dual Emotion for Fake News Detection. Proceedings of the Web Conference. 2021. P. 3465–3476.
