A Study On Online Audit

Online auditing was arising in the context of information technology, is a new audit model. Online auditing has many advantages, but also has not inquired about the use of inquiry method, observation method and inventory method, so online auditing has higher requirements about networked computer forensic audit techniques.In online auditing, electronic data generated by the audited units is the major audit information. The use of query and analysis capabilities of database management system, the use of audit software and the use of online analytical processing (OLAP) technology are the current online auditing data analysis techniques. But the use of these data analysis can't find audit trail behind the data and hidden cheating techniques, OLAP analysis can not qualify for the potential information and knowledge hidden in the data.Data mining extracts implicit but useful information and knowledge, that people do not know in advance, from a lot of not complete, noise, fuzzy data. Different from the traditional analysis tools, it is more focused on the use of pattern matching and a variety of algorithms to determine the important link between the data, with a strong knowledge discovery.Data mining technology has been applied in marketing, customer management, credit analysis and other areas. Online auditing meets the four conditions of the application of data mining technology, so this paper attempts to apply data mining techniques to analyze the audit data processing and extract implicit information and knowledge in the audit data for the audit process.In this paper, data mining technology combines with online auditing system from the data sharing. The data by the preliminary data processing of collected raw data stored in data warehouse, and data mining software imported the data to analyze. In this article, the author visited the Audit Office in Hangzhou to research Financial Online auditing. Based on the existing system and theoretical foundation of System Theory, to build online auditing system based on data mining technology, the system was divided into seven module, focuses on the isolated points, association rules, neural networks and decision trees are four methods in the system application of knowledge will be excavated for the audit process, to build the audit analysis model to determine the audit priorities, providing decision support, etc. . Finally, take the example of the departmental budget data of Financial Online auditing. Because of the characteristics of the selected data, to select the isolated points and association rules to analyze the data. The data from 139 departments were found 10 isolated points, these 10 showed large differences in budget execution department, then take further audit procedures to review the existence of irregularities in the 10 sectors. The decision tree extracts classification rules through the remaining data of 129 sectors, a total of eight rules extraction, 3 of which the credibility of the rule over 80% can be considered as an effective rule-based, these three rules for real services in the practice.The innovation of this paper has two points. One is to build online auditing system based on data mining technology, and second is the first attempt to use data mining techniques to carry out excavation analysis of online auditing data.