Research On Audit Evidence Acquisition Based On Association Rules And Outlier Algorithm

In the face of massive electronic audit data,the traditional method of computer aided audit was just to make the process of manual audit computerization,which could not get more valuable information from electronic data.In recent years,Data Mining Technology was tried to introduce into computer audit by academia and industry.Based on the Audit and Supervision Information System of a large military hospital,Association Rules and Outlier algorithm were introduced into the research of audit evidence acquisition in the paper and were used in the system.The main research contents of this paper were as follows:(1)Through a large number of literatures at home and abroad,the current development situation of Data Mining in computer-aided audit was researched,and based on the Audit and Supervision Information System of a large military hospital,Association Rules and Outlier Algorithm were introduced into the research of audit evidence acquisition.Between them,Association Rule Algorithm was used to obtain "general" audit evidence and Outlier Algorithm was used to obtain audit evidence of abnormal behavior.(2)According to the shortcomings of Apriori Algorithm and FP-Growth Algorithm,an improved algorithm of FP-Growth Algorithm was proposed,which saved the construction of conditional FP-tree and used Apriori Algorithm in frequent item-set mining in the FP-tree.The space-time efficiency of the algorithm was significantly improved.The effectiveness of the improved algorithm was verified through experiments.The Association Rule Algorithm was introduced into the Audit System to obtain the relational audit evidence and the desired results were achieved.(3)According to the shortcomings of Cell-based Outlier Algorithm,a dynamic threshold method was proposed,and then the methods to determine the relevant parameters and the algorithm process were given.The improved algorithm was verified by experiments.The Cell-based Outlier Algorithm was introduced into the Audit System to obtain the audit evidence of abnormal behavior and the expected results were obtained.(4)The Audit and Supervision Information System of the hospital was overall designed and the system modules were determined.The modules of the system were designed in detail by UML and the system development work was completed.Through the actual operation of the system,the expected results were achieved by all modules and audit evidence was obtained well through Association Rules and Outlier Algorithm.