Computer Audit Methods Based On Data Mining Research And Realized

Computer audit, also known as computer-assisted audit, with computer system becoming more applied by enterprises and economic management departments ,is the modern computer technology applied for the audit. With many years of auditor's exploration, the computer audit techniques have been widely used in auditor's routine work.In recent years, the magnanimity trend of the data has been already obvious. Computer audit techniques meet new requirements. How to find valuable information from data or database is a new problem. In such cases, auditors try to use data mining techniques to solve this problem. Data mining extracts implicit,previously unknown and usable knowledge from large amounts of data. Using data mining technology advantage in this respect, auditors can find something beyond our experience.It is very useful for computer audit.In this paper I describes definitions, concepts ,other related objects and the development of our computer audit, mainly focuses on the key point of the computer audit subject - computer audit methods, analyze the lack of common computer audit methods. On this basis, I try to use data mining methods in practice to solve the problem for audit and proposed requirements. As a knowledge-discovery process, data mining technology can model. The method of modeling is base on specific objectives, auditors must avoid subjective ideas in this process. With the use of data mining, auditors can gain many unusual questions and ensure the accuracy of their perspective. Then I show the general flow of the current computer audit. The Internationally CRISP-DM is set forth mining methods to solve practical problems. Summarize the process and the framework of data mining technology based on computer-assisted.Another, the feasibility of the application of data mining to computer-assisted audit is proved. Operate Clementine 12,a professional data mining tools, with actual data, to auditing the funds of agricultural relief and auditing the Credit of Rural Credit Cooperatives . From Business understanding, Data understanding to Modeling, using Factor analysis, Two step Cluster, Abnormal, Decision tree to solve practical problems. The presentation of Clementine is also in this process.