基于聚类方法的审计分层抽样算法研究  

RESEARCH ON AUDITING STRATIFIED SAMPLING WITH CLUSTERING

作　　者：夏锋[1] 彭鑫[1] 赵文耘[1] 

机构地区：[1]复旦大学计算机科学与工程系,上海200433

出　　处：《计算机应用与软件》2008年第1期14-16,共3页Computer Applications and Software

基　　金：国家自然科学基金(60473061);国家863计划(2005AA113120);上海市科学技术委员会(04DZ15022)。

摘　　要：针对审计抽样中最复杂的抽样算法—分层抽样,从数据挖掘中"聚类"的角度出发,较好地运用了聚类思想于审计抽样的分层抽样算法之中,为该算法的实现提供了一种新的解决方案。AICPA39没有为分层抽样提供具体的实现方式,国内的学者曾从统计学角度有过实现,将从计算机科学角度实现方法与统计学实现方法进行分析比较,这是对分层抽样算法实现的有益新探索。The emergence of auditing software is booming recently in China. But none of them have effectively realized auditing stratified sampling which is the most complicated auditing sampling method. According to AICPA39, specific implementation of stratified sampling is not provided. A new auditing stratified sampling algorithm based on clustering is presented. Compared with the traditional methods, the proposed method is more effective in implementing the auditing stratified sampling.

关 键 词：数据挖掘 聚类 审计抽样 分层抽样K-means算法 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]