摘要
为了改进传统的仅仅是把手工审计流程计算机化的计算机辅助审计方法和发现被审计数据中的隐藏信息和更多的审计证据,提出了一种先对海量数据进行数据划分,然后采用改进的孤立点检测技术的审计证据获取方法。该方法首先利用改进粒子群算法对被审计数据进行划分优化,找到高内聚、低耦合的数据划分;然后使用基于距离的改进孤立点检测技术,查找出孤立点数据;最后通过分析发现审计线索。通过相关对比实验表明,该方法易发现海量被审计数据中的隐藏信息,孤立点检测效率也有很大提高,从而提高了审计效率。
In order to improve the traditional computer assisted audit method which only was computerized that the manual audit progress and found hidden information and more audit evidence in the audit data, applied an audit evidence gathering method which first used data partition and then adopted improved outlier detection to deal with the huge audit data. First, divided and optimized audit data by improved particle swarm optimization algorithm, then found the high-cohesion and low-coupling data partition. The second, used the improved outlier detection based on the distance to select outlier data. At last, gathered more audit evidence by analysis. Through the relevant comparative experiments, it shows that with this method the hidden information is easily found in the huge audit data, the efficiency of outlier detection is greatly improved, thereby the efficiency of the audit is enhanced.
出处
《计算机应用研究》
CSCD
北大核心
2009年第7期2495-2498,2501,共5页
Application Research of Computers
基金
国家自然科学基金资助项目(70701018)
关键词
计算机辅助审计
孤立点检测
粒子群算法
数据划分
剪枝技术
computer assisted audit
outlier detection
particle swarm optimization ( PSO )
data partition
pruning technology