基于文本数据分析的大数据审计方法研究

Research on Big-Data Audit Method on Basis of Text Data Analysis

研究如何对文本数据进行分析对开展大数据审计具有重要的应用价值。本文首先分析了常用审计方法的不足,进而根据目前开展大数据审计的需要提出了基于文本数据分析的大数据审计方法,并分析了其原理;在此基础上,采用Python语言实现了相似度分析和标签云分析两种可行的文本数据审计方法;最后,结合具体案例,分析了基于文本数据分析的大数据审计方法的应用。研究结果为开展大数据审计提供了技术方法和研究方向。

Studies on how to analyze the text data can produce important practical values for conducting big-data audits. This article first analyzes the shortages of common audit methods, proposes a text data analysis-based big-data audit method, and also analyzes its principles. Based on the above, it adopts the python language to introduce two feasible methods of text data audit, namely similarity analysis and cloud analysis of tags; finally, using specific cases, it analyzes how to apply such a big-data audit method on basis of text data analysis. The article provides practicable technical approaches for conducting big-data audit activities and directions for future research.