Artificial intelligence and social networks: approaches to detecting fake information

DOI: 10.31673/2412-4338.2024.044748

Abstract

The article examines modern approaches to detecting fake information on social networks using artificial intelligence (AI) and machine learning (ML). The growing popularity of social networks is accompanied by a global disinformation problem, which has severe implications for society, the economy, politics, and even public health, as was especially evident during the COVID-19 pandemic. Particular attention is given to methods of detecting fake news, including the use of neural networks, user psychological profiling, and text analysis. Additionally, the study explores approaches to analyzing visual content, such as images and videos, to assess their authenticity. The article also analyzes the role of social bots in disseminating disinformation, including their ability to influence public opinion through manipulation, as well as tools for identifying them, such as graph models and heuristic methods. The study characterizes data collection and processing techniques, including social platform APIs, web scraping, user activity monitoring, and interaction analysis. Special attention is devoted to data preprocessing steps, including cleaning, normalization, tokenization, lemmatization, and annotation, which significantly impact the quality of algorithm performance in detecting fake information. The challenges of scalability and system efficiency in the context of large data volumes are discussed, along with issues related to ensuring user privacy. Furthermore, the article highlights the necessity of adapting algorithms to new patterns of disinformation, which evolve in response to technological advancements, and the importance of an interdisciplinary approach that combines achievements in AI, cognitive science, linguistics, and sociology.

Keywords: artificial intelligence, social media, fake news, machine learning, disinformation, neural networks, news detection, data processing.

References

1. Silva, F., Vieira, R., & Garcia, A. Can machines learn to detect fake news? A survey focused on social media. Hawaii international conference on system sciences - HICSS. 2019. P. 1–8. URL: https://doi.org/10.24251/HICSS.2019.332. ISSN 2412-4338 Телекомунікаційні та інформаційні технології. 2024. № 4 (85)
2. Arming the public with artificial intelligence to counter social bots / K. Yang et al. Human behavior and emerging technologies. 2019. Vol. 1, no. 1. P. 48–61. URL: https://doi.org/10.1002/hbe2.115 (date of access: 03.11.2024).
3. The mass, fake news, and cognition security / B. Guo et al. Frontiers of computer science. 2020. Vol. 15, no. 3. URL: https://doi.org/10.1007/s11704-020-9256-0 .
4. Nistor A., Zadobrischi E. The influence of fake news on social media: analysis and verification of web content during the COVID-19 pandemic by advanced machine learning methods and natural language processing. Sustainability. 2022. Vol. 14, no. 17. URL: https://doi.org/10.3390/su141710466
5. Bio-Inspired artificial intelligence with natural language processing based on deceptive content detection in social networking / A. A.Albraikan et al. Biomimetics. 2023. Vol. 8, no. 6. P. 449. URL: https://doi.org/10.3390/biomimetics8060449.
6. Fake news detection on social media / K. Shu et al. ACM SIGKDD explorations newsletter. 2017. Vol. 19, no.1. P. 22–36. URL: https://doi.org/10.1145/3137597.3137600 .
7. Shevchenko O., Bondarchuk A., Polonevych O., Zhurakovskyi B., Korshun N. (2021) Methods of the objects identification and recognition research in the networks with the IoT concept support. Workshop on Cybersecurity Providing in Information and Telecommunication Systems, (pp. 197–209)
8. Moshenchenko, M., Zhurakovskyi, B., Poltorak, V., Bondarchuk, A., Korshun, N. (2021) Optimization Algorithms of Smart City Wireless Sensor Network Control. Workshop on Cybersecurity Providing in Information and Telecommunication Systems, (pp. 32–42)