基于大数据可视化技术的审计线索特征挖掘方法研究

Study on Feature Mining Methods for Audit Clues Based on Big Data Visualization

电子数据审计的研究与应用是近年来审计领域的热点问题。传统的电子数据审计方法一般是查找审计线索,发现被审计单位中存在的问题,没有进一步挖掘审计线索的特征,分析产生相关问题的规律和原因。因此只能是发现被审计单位存在的表面问题,不能通过发现的审计线索分析出更深层次的问题。本文首先分析了研究审计线索特征挖掘方法的重要性、目前常用审计方法及其存在的不足,在分析大数据可视化技术的基础上,提出了基于大数据可视化技术的审计线索特征挖掘方法,并分析了该方法的原理。以某医院审计为例,验证了该方法的有效性。最后,探讨了该方法的优缺点及适用情况。研究结果为今后大数据环境下开展电子数据审计提供了理论基础与技术方法。

The research on and application of electronic data auditing have been a topic of general interest in audit research. Traditional electronic data auditing methods can identify clues to the problems of auditees, and cannot further explore the features of those clues and sum up the patterns and causes of relevant problems after analyses. Therefore, traditional electronic data auditing methods can only detect certain problems of the auditees, but cannot find out deeper problems through audit clues. However, big data visualization technologies can solve these problems. This paper starts by analyzing the importance of feature mining methods for audit clues based on big data visualization and the shortages of commonly used audit methods, and then put forward feature mining methods for audit clues based on big data visualization. The methods were verified through a hospital audit project. Finally, the advantages and disadvantages of the methods are discussed. Research results of this paper can provide reference to future electronic data audits in big data environment