Analytical Processing for Forensic Analysis

The current status of Information Technology has prompted several countries to adopt technologies based on internationally accepted standards for the disclosure of their financial statements. In order to facilitate data exchange and increase the transparency of financial information available on the Internet, the XBRL technology has been adopted as standard by many regulators of the financial market. However, the occurrence of financial crimes in big corporations fosters research on fraud detection using computational resources. With such aim, this paper proposes a support tool for forensic financial analysis, based on OLAP query languages for the digital detection of fraud in financial reports represented in XBRL. We propose an extension of the LMDQL language, which supports both queries in XML documents connected by XLink and XML Schema, and also relational databases. For this reason, forensic operators are presented, extending those defined in LMDQL, and its use is demonstrated on a relational repository of multidimensional data, based on XBRL. To review the forensic operators a case study was realized using XBRL documents provided by US- SEC, to which an ETL processing was applied for the loading of the data, making it possible to analyze the occurrence of fraud in large volumes of data presented.