The main goal of this research is to predict financial distress in terms of key words and phrases using the text mining technique. The statistical population of this research is the audit reports of 50 financially disadvantaged companies listed in Tehran Stock Exchange in 2017. Therefore, these companies were subjected to text mining in order to collect the required data for the research reports from the years 2015 and 2016. Research findings indicate that the term non-observance of regulations and rules with 162 cases, the word conflicts with 123 cases, the absence of the presence of the beneficiary's manager with 122, loss of profit, decrease in the value of goods and assets with 116 items, lack of access to information And documents with 102 cases, adjustments to 101 cases, bonds of directors with 93 cases, board approvals of 86, and ultimately accumulated losses and renewals with 85 views in audit reports. Given the extracted terms and expressions, the method of text mining can be used to predict corporate financial distress. It can also be used as a method to extract useful information from audit reports.