A Review of Social Media Sentiment Analysis in Financial Risk Control: Based on Natural Language Processing Technologies

This review examines the application of social media sentiment analysis, driven by Natural Language Processing (NLP) technologies, within financial risk control. It outlines the four-stage evolution of NLP, from traditional methods to pre-trained models. Using credit risk control as an example, it elaborates on its application across the entire loan lifecycle: pre-loan, in-loan, and post-loan. The study identifies limitations such as data scarcity, insufficient model generalization and interpretability, and high implementation costs. It also explores potential breakthroughs, including dataset construction and model optimization, aiming to provide a framework for related research and practice.