Fraud Detection Time Series Analysis Forecasting Cluster Analysis
Issue Date:
2007
Publisher:
Institute of Mathematics and Informatics Bulgarian Academy of Sciences
Citation:
Pliska Studia Mathematica Bulgarica, Vol. 18, No 1, (2007), 271p-292p
Abstract:
It is very often the case that the patterns of a fraudulent activity in trade are hidden within existing trade data time series. Furthermore, with the advent of powerful and affordable computing hardware, relatively big amounts of available trade data can be quickly analyzed with a view to assisting antifraud investigations in this field. In this paper, based on the availability of such import/export data series, we present a statistical method for the identification of potential fraud schemes, by extracting and highlighting those cases which lend themselves to further investigation by anti-fraud domain experts. The proposed method consists in applying time series analysis for prediction purposes, calculating the resulting significant deviations, and finally clustering time series with similar patterns together, thus identifying suspect or abnormal cases.