Using association rule mining in forensic accounting

A few years ago I started to use the R programming language more intensive while writing my master thesis. I used the wonderful arules package for mining association rules and frequent item sets from Michael Hahsler and others. I used this package in the field of forensic accounting. Forensic data analysis is a branch of digital forensics. It examines structured data with regard to incidents of financial crime. The aim is to discover and analyze patterns of fraudulent activities (Wikipedia). Find down below an excerpt from my thesis.

A study by Schneider and John from the year 2013 shows that 37% of the surveyed companies in Germany report that they have already become victims of economic crimes in the last twelve months. The literature research of the present master thesis has been shown that a large part of the forensic analysis methods are used to uncover economic crimes on aggregated data (e.g. balance sheet positions). On the basis of various scientific researches, it can also be shown that there are currently only a few publications which use analytical methods to investigate unaggregated transactions of financial accounting directly on economic crimes. A study of Debreceny and Gray from 2013 reveals that the analysis of the company’s internal financial accounting data has great potential for detecting fraud. For these reasons, this master thesis uses the data mining methodology of association analysis to directly apply financial accounting data for the purposes of forensic accounting to investigate economic crimes.

Make sure you check out the code on my Github along with other projects.