Software to Streamline the Detection of Fraud in Published Financial Reports of Companies

Considerable effort has been devoted to the development of software to support the detection of fraud in published financial statements of companies. Until the present date, however, the applied use of such research has been extremely limited due to the ``black box'' character of the existing solutions and the cumbersome input task they require. The application described in this paper solves both problems while significantly improving performance. It is based on Web-mining and on the use of three Multilayer Perceptron where a modified learning method leads to the formation of meaningful internal representations. Such representations are then input to a features' map where trajectories towards or away from fraud and other financial attributes are identified. The result is a Web-based, self-explanatory, financial statements' fraud detection solution