Methods and software for the detection of false information on the internet based on neural networks

DOI: 10.31673/2412-9070.2024.045257

Abstract

The article contains an overview of modern methods and software based on neural networks for detecting false information on the Internet, as well as an analysis of current problems and possible directions of future research in this field. The results of the work can be used as an information base for continuing research in the direction of using neural networks of various types to prevent the spread of fake information on the Internet.
With the dawn of the digital age and the popularity of online social networks, information is spread faster and easier than ever before. However, it also promotes the spread of poor quality or intentionally fake information, which can have a negative impact on society. Identifying, flagging and refuting disinformation on the Internet as quickly as possible is becoming an increasingly urgent problem.
The article shows that modern methods of using neural networks in detecting false information are highly effective due to their ability to process large volumes of data and detect complex patterns. The use of graph neural networks, behavioral analysis and other innovative technologies provides wide opportunities for adapting detection systems to different requirements and conditions, which allows developing more flexible and effective solutions that can work in different contexts and with different types of data.
An important advantage of neural networks and their software implementation is the possibility of integration of various data sources and contextual information. This allows software information systems not only to analyze the textual content of news, but also to take into account social interactions, the history of publications and other factors that may indicate the falsity of information.
It is shown that a very important aspect of false information on the Internet is images and videos that are presented with a false interpretation or with modification. The ability of the future system to recognize such cases will significantly increase the effectiveness of determining the reliability of information.
But despite the advantages, there are problems such as high computational requirements and difficulties in interpreting the results. This requires further research and improvements, especially in the area of improving algorithms and developing more efficient and scalable solutions and software. In addition, an important direction is the development of methods for early detection of fake news and minimizing their impact on public opinion.

Keywords: social networks; fake news; false information; neural networks; software engineering.

References
1. Andriotis, Panagiotis, Atsuhiro Takasu. Emotional Bots: Content-Based Spammer Detection on Social Media // IEEE Xplore. 1 Dec. 2018. P. 1–8 [Electronic resource]. URL: https://doi.org/10.1109/WIFS.2018.8630760. Accessed 12 Jan. 2024.
2. Fake News Detection: Taxonomy and Comparative Study / Faramarz Farhangian [et al.] // Information Fusion. vol. 103. Elsevier BV. Mar. 2024. P. 102140–40 [Electronic resource]. URL: https://doi.org/10.1016/j.inffus.2023.102140.
3. Fake News Detection through Graph-Based Neural Networks: A Survey / Gong Shuzhi [et al.] // ArXiv (Cornell University), Cornell University, July 2023 [Electronic resource]. URL: https://doi.org/10.48550/arxiv.2307.12639.
4. Kudugunta Sneha, Emilio Ferrara. Deep Neural Networks for Bot Detection // Information Sciences. vol. 467. Oct. 2018. P. 312–22 [Electronic resource]. URL: https://doi.org/10.1016/j.ins.2018.08.019. Accessed 10 Jan. 2024.
5. Fake News Is Not Simply False Information: A Concept Explication and Taxonomy of Online Content / Maria D. Molina [et al.] // American Behavioral Scientist. 2019. vol. 65, no. 2. p. 000276421987822 [Electronic resource]. URL: https://doi.org/10.1177/0002764219878224.
6. Fake News Detection: A Survey of Graph Neural Network Methods / Huyen Trang Phan [et al.] // Applied Soft Computing. 2023. vol. 139. p. 110235 [Electronic resource]. URL: https://doi.org/10.1016/j.asoc.2023.110235.
7. A Taxonomy of Fake News Classification Techniques: Survey and Implementation Aspects / Dhiren Rohera [et al.] // IEEE Access. 2022. vol. 10. P. 30367–94 [Electronic resource]. URL: https://doi.org/10.1109/ACCESS.2022.3159651.
8. Twitter Bots in Cyber-Physical-Social Systems: Detection and Estimation Based on the SEIR Model / Weisha Zhang [et al.] // Security and Communication Networks. Hindawi Publishing Corporation, May 2023. P. 1–9 [Electronic resource]. URL: https://doi.org/10.1155/2023/6234030.
9. USAID-INTERNEWS. Attitude of the population towards mass media and consumption of different types of media in 2019 [Electronic resource]. URL: https://drive.google.com/file/d/1Oi2Edvl5Srk4hS-D2KoxoKkamCarUX7f/view?pli=1. Accessed 8 Jan. 2024.