10The Framework of Feature Extraction for Financial Fraud Behavior and Applications
George X. Yuan1,2,3, Shanshan Yang1*, Lan Di4†, Yunpeng Zhou3, Wen Chen3 and Yuanlei Luo5‡
1Business School, Chengdu University, Chengdu, China
2College of Science, Chongqing University of Technology, Chongqing, China
3Shanghai Hammer Digital Tech. Co. Ltd. (Hammer), Shanghai, China
4School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China
5Research Center, China Institute of Ocean Engineering, Beijing, China
Abstract
By employing the Gibbs algorithm, we establish a general framework to extract features for the detection of corporate’s financial frauds by using a fintech method related to the big data analysis. In the empirical analysis, based on those proxy “bad” samples of events in describing illegal behaviors by these Chinese A-share listed companies released by China Securities Regulatory Commission (CSRC) due to their (black) behaviors mainly violating the Rules of Disclosures and related fraudulent actions during the time period from the beginning of 2017 to the end of 2018, we conducted risk assessment for those highly related risk factors (features) that could provide related risk information on the exposure of financial fraud events by detecting the difference between their financial fraud actions with normal performances in the practices in the capital market of China.
In this paper, by employing the Gibbs sampling method, we are able to extract eight ...
Get Artificial Intelligence for Risk Mitigation in the Financial Industry now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
