Abstract:

Since investors, financial professionals, and the securities and exchange administrators highly rely on the financial filings for information about companies they are evaluating for investment purposes, it is impractical to solely trust a financial institute’s assessment report or manual approach to identify financial fraud given that there are increasing numbers of companies listed or looking for investment every day. The research and development of automated financial fraud detection system is of great challenge because of the fact that there is a separation between the normal deception detection techniques and the financial statements analysis. Guided by design science research methodology, we propose an analytical framework to detect financial fraud of companies based on an extended social cognitive theory. Business network analysis is introduced as a novel approach to derive latent contextual features. The potential performance of the prototype system will contribute to implementation into both theoretical and practical domains.