Financial Diagnostic Study With Data Mining

With the information accumulated during the development of a large number of financial data in companies, otherwise the commercial value of these financial data has not been fully tapped and use.Traditional methods of financial diagnostic simply use a small amount of statistical methods for analysis of financial data and evaluate the results, but if handle, tens of millions of information in a industry is more difficult.mass financial data information about the potential depth is not enough depth when using traditional methods.Data mining as one of efficient and sophisticated tools,its strengths are the ability to quickly get from large, incomplete, noisy, fuzzy and random data capture valuable information, but people do not know to get the data behind unique business rules, and summarized these laws as models. With using data mining technology advantages in this respect. In this paper, financial diagnosis is divided into diagnostic data system, the four most of the general and special judgments, financial diagnostic and financial early warning system, financial diagnostic program planning and decision-making system. Data will be obtained through general and special judgments, then data mining methods will be applied to the financial diagnosis and early warning, which the financial diagnostic program planning and decision-making.This paper describes the theory and methods of data mining in the practical application of financial diagnosis.SPSS Clementine12.0as the operating platform, and using the financial diagnostic methods combining with data mining technology tools.The financial statement data of the HN enterprises published on the Internet using XBRL standards. The paper using decision tree、correlation and cluster analysis to establish the financial diagnosis system.Through this model, we found it can increase the efficiency of financial diagnosis with high practical value.