Sentiment analysis in financial markets to predict and manage bank runs using artificial intelligence techniques

Hamed Rafiee Monfared¹

1) M.Sc. Graduate in Business Management, specializing in E-commerce, Islamic Azad University, Shahr-e-Qods Branch, Iran

IOI Code : AAGK-ABBKKE
Publication : 4th International Conference on Researches in Accounting, Economics, Management & Humanities(meaconf.com/4th)
Abstract :
    Bank runs are one of the most destructive phenomena in financial systems, rooted in information asymmetry and collective panic. With the expansion of social networks, the speed of spreading rumors and negative news has increased dramatically, and traditional liquidity management models are no longer able to respond in a timely manner. This research aims to provide a new framework to examine the role of artificial intelligence (AI) and natural language processing (NLP) in real-time monitoring of the psychological atmosphere of the market. The findings of this research show that sentiment analysis algorithms are able to extract crisis precursor signals from unstructured data with high accuracy. This article, while explaining the theoretical foundations, presents a proposed architecture for early warning systems (EWS) and examines the challenges of implementing it in the modern banking system.

Keywords : artificial intelligence bank run sentiment analysis natural language processing liquidity risk management financial stability