The Application of Text Analysis in Financial Research:A Literature Review and Paradigm of Use
In the current era of Big Data and Artificial Intelligence,unstructured data,particularly text,has attracted widespread attention from scholars.The use of textual data is playing an increasingly important role in the social sciences,especially in financial research.Text analysis is creating new paradigms in financial research by studying non-traditional themes such as sentiment,policy uncertainty,and semantic similarity.This paper focuses on the application of text analysis in financial research.By reviewing relevant domestic and international literature,it first summarizes the usage processes and paradigms of text data in financial research,detailing four key processes:methods for obtaining financial texts,ways of processing financial texts,models for representing financial texts,and the construction of financial text indicators.Subsequently,in the context of the prevalence of Large Language Model(LLM),the paper focuses on analyzing their potential applications in financial research and discusses the potential risks and challenges posed by LLM,such as data privacy and model bias.Finally,the paper summarizes the application of text analysis in financial research,reflects on the criticisms from the academic community regarding text analysis,and looks ahead to the future use of LLM in financial research.
Text AnalysisFinancial ResearchUnstructured Big DataDeep LearningLarge Language Model
NETL
NSTL
万方数据
