智能+技术背景下基于聚类算法的自动化审计研究

Research on Automatic Audit Based on Clustering Algorithm Under the Background of Intelligence+Technology

审计工作涉及大量的物资采购、财务以及税务数据,在人工模式下存在效率低下、成本较高的缺点,可将算法引入其中,从而实现自动化审计。具体的实施方法为收集存储原始业务数据,开展数据抽取、清洗和转换,利用聚类算法构建审计模型,通过机器学习训练、改进模型,直至其达到成熟状态。研究过程中依托K-means算法实现聚类,建立审计指标,收集企业的实际数据,开展自动化处理。结果显示:该算法模型能够实现指标重要性排序,并输出量化的审计结果。

The audit work involves a large number of material procurement,financial and tax data,in the manual mode,there are low efficiency and high cost shortcomings,the algorithm can be introduced into it,so as to realize the automatic audit.The specific implementation method is to collect and store the original source Business data,carry out data extraction,cleaning and conversion,use the clustering algorithm to build the audit model,and train and improve the model through machine learning,until it reaches the mature state.In the process of research,we rely on K-means algorithm to realize clustering,establish audit indicators,collect the actual data of enterprises,and carry out automatic processing.The results show that the algorithm model can rank the index importance and output the quantitative audit results.