Financial Fraudulent Statements Detection through a Deep Dense Artificial Neural Network

Abstract

A very important issue in the financial field is to identify and reliably predict Fraudulent Financial Statements (FFS). For this purpose, several Machine Learning models have been developed that identify the issues that are directly related to FFS. In this paper, we present a new predictive model for fraudulent detection through a deep dense artificial neural network. Specifically, a new forecasting model was tested experimentally using data from Greek companies. The obtained results showed that the proposed scheme is robust and promising.