Research On The Visualization Method Of Audit Trail Orientedto Financial Fraud

In recent years,the sharp increase in the number of listed companies in China and the continuous financial fraud means put forward higher requirements for audit work.It is of great practical significance to overcome the limitations of traditional manual identification methods and to seek efficient financial fraud identification methods instead.Under the information environment,it has become an inevitable trend to identify the relationship between financial fraud indicators from massive data.If we can use information technology to find out the key financial data with a large amount of information,Auditors will significantly improve the efficiency and effectiveness of their work in an information-based environment.Secondly,in order to reflect the characteristics of fraudulent enterprises clearly and intuitively,the key financial data are represented by visual method.Help auditors find audit trail,improve the efficiency of audit work.First of all,on the basis of combing and analyzing the related theories of financial fraud and visualization,this paper proposes to select 22 related indicators from the three dimensions of financial quality,liquidity and internal governance.In the course of the study,this paper collected data on Chinese listed companies punished by CSRC,Shanghai Stock Exchange and Shenzhen Stock Exchange from 2011 to 2017,and selected 93 companies with fraud as samples.Then,according to the data of the financial statements of the sample companies,the related index values are calculated after collating and analyzing them.In order to visualize the important financial information,this paper makes a survey of the 22 indicators selected in this paper.Screening.First,the total proportion of the first 15 financial indexes ranked by entropy weight was 96.35%,and the comprehensive correct rate was 85.5% after logistic regression test.Compared with 22 indexes,the accuracy rate decreased by only 1.2 percentage points.Therefore,the number of financial indicators presented in this paper will be reduced from 22 to 15,and the simplified indicators will be used as the data of the radar chart of financial fraud.Secondly,113 unpunished listed companies which are equal to the assets of fraudulent companies are selected in this paper and the relevant financial index data are sorted out as the data of non-fraud radar chart.Finally,this paper respectively according to the system The classification of manufacturing,mining,accommodation and catering,and wholesale and retail industries,the combination of financial fraud radar and non-fraud radar maps,the construction of audit trail maps,and the comparison between the case enterprises and the constructed audit trail maps.Check its applicability.At present,the research on the causes and identification of financial fraud is an important research direction to ensure the orderly operation of the domestic market economy,as an indispensable part of the process of fraud identification.The audit staff acted as a gatekeeper to the financial statements.This paper puts forward a brand-new angle of view on audit work under big data's environment,and puts forward the audit trail chart for four industries.Secondly,in the process of constructing audit trail chart,information entropy and logistic regression verification are introduced to screen out the key financial fraud indicators.Audit trail charts based on key financial indicators are clear and visually reflect fraudulent enterprises and The differences in the financial position of other enterprises help audit staff to find the audit object quickly;Moreover,the similarity will be the audit trail of auditors,which will help auditors to examine and find more valuable audit evidence,and improve the recognition rate of audit work.