Research And Application Of Electronic Data Audit Method Based On Similarity Detection

A number of different audited documents may contain similar content,which may,in some cases,be a breach of the audit issues.For example,assuming that it is not possible to have some similar content in the two audited documents collected from different sources,and then the similar data found out can be an audit clue.This kind of similar data can be found effectively by similarity detection.At present,the research of similarity detection technology at home and abroad is mainly used to find the similar or duplicate data in the data source.Use this technique to clean up similar or duplicate data,so as to improve the quality of the data.However,the research and application of this technology combined with audit practice is not common at present.First of all,the author collected and read the literature related to the title,including the informatization of audit,similarity detection technology and electronic data auditing and other aspects,in order to analyze the significance and effect of application of similarity detection technology in the electronic data auditing.Then,the second chapter introduces the main theories and techniques needed to develop similar detection system,including distance measurement method,Chinese word segmentation method and similarity algorithm.The third chapter of this paper introduces the three main functional modules of the similarity detection system in detail.The first module is data acquisition,which is responsible for transferring the data to the server.The second module is the similarity detection of database data.By using the edit distance algorithm to calculate the similarity between the character data,querying the similarity of the data in the table and matching the similar data between the tables are realized.The third module is the similarity detection of Word documents and this module mainly realizes two functions.The first function is to carry out a simple similarity detection between the two documents,including the calculation of their similarity and display similar content,which uses the Chinese word segmentation method and Jaccard similarity calculation method.The second function is to establish the online library of audit document,so as to facilitate the sharing of documents within the audit group,and achieve the keyword retrieval and document matching.In the fourth chapter,in order to test the practicality of the functions in the similar detection system,three cases of the audit project of the Audit Commission are selected to test the above three functions.In the case,this paper compared the working methods of auditors and functions of system about effect and efficiency,in order to verify the similarity detection system that it can not only improve the auditors' working efficiency,but also contribute to the discovery of audit clues.Finally,this paper summarizes the innovation and deficiency of the research process and results,and puts forward the future research direction.The research of this paper has realized the combination of similarity detection and the method of electronic data auditing in theory,which enriches the method of using computer technology to assist the audit work.At the same time,the research in this paper is also helpful to find the similar or duplicate content in the data source in the audit work,and then auditors may obtain the audit clues.