This article explores the role of social networks as a catalyst for instability in financial markets. Its purpose is to present strategies for mitigating the media’s influence on adverse outcomes, primarily by analyzing the information cascades and related transactions in stock markets. The innovative aspect of this research lies in the development of methods for analyzing specific types of information, particularly media reports, to identify indicators of fake that distort public perception. The focus of this study is on fake news and the analysis of its potential linguistic features for identification purposes. A significant outcome of this research is the establishment of a comprehensive method for conducting preliminary linguistic analysis of text content from economic and political news portals, enabling a reliable assessment of information credibility. I propose a methodology that acts as a preventive tool—a proactive measure to mitigate the impact of social media posts on the financial sector. The correlation between identified fake news and stock market dynamics is 0.5, indicating a noteworthy relationship. Additionally, predictive models were tested, and neural networks proved to be the most effective.
Konstantyn Malyshenko (Thu,) studied this question.