Cluster analysis for anomaly detection in accounting

Cluster Analysis is a useful technique for grouping data points such that points within a single group or cluster are similar, while points in different groups are distinctive. Clustering as an unsupervised learning algorithm is a good candidate for fraud and anomaly detection. The purpose of this study is to examine the possibility of using clustering technology for continuous auditing. Automating fraud filtering can be of great value to preventive continuous audits. In this paper, cluster-based outliers help auditors focus their efforts when evaluating group life insurance claims. Claims with similar characteristics have been grouped together and those clusters with small population have been flagged for further investigations. Some dominant characteristics of those clusters are, for example, having large beneficiary payment, having huge interest amount and having been submitted long time before getting paid. This study examines the application of cluster analysis in accounting domain. The results provide a guideline and evidence for the potential application of this technique in the field of audit.