Credit risk management of scientific and technological enterprises based on text mining

The purpose of this paper is to verify the impact of financial news on corporate credit risk. Web crawler technology is used to obtain the financial news text from Sina Financial News. The text mining technology is utilized to quantify the financial news text. The quantified financial news text combined with financial indicators is used to build the Logistic regression model. The assessment results show that the risk early warning accuracy of the Logistic regression model incorporating both the quantified financial news text and financial indicators is higher than the Logistic regression model only with pure financial indicators.