Data Ming-Based Financial Fraud Detection; Current Status and Key Issues
Recent advances in information, detection, data mining, risk and security technologies have given rise to a new era of research, known as data mining based financial fraud detection ( FFD). Several data mining algorithms including regression, neural network, decision tree, Bayesian and Stacking variant methodology, incorporating financial ratio and learning mechanisms, have been developed that allow one to extract relevant knowledge from large amount of data like fraudulent financial statements ( FFS). FFD is a new attempt; thus, several research questions have often being asked. For instance; ( 1) how to finish the detection process for these algorithms? (2) how to understand and classify these algorithms in terms of fraud detecting? ( 3 ) will they require specific data? (4 ) what kind of detection approach are used? (5) which one is the most popular method in use? To help answer these questions, we conduct an extensive review on literature. We present a generic FFD framework and a classification scheme to guide the review process. Frequencies of different techniques/algorithms used are tableau and analyzed. Finally, we share directions for future research.
Fraudulent Financial Statements Financial Fraud Detection Data Mining Management Fraud
Yue Dian-min Wu Xiao-dan Li Yue Chu Chao-hsien
School of Management, Hebei University of Technology, Tianjin P.R.China, 300401 School of Information Science and Technology, Pennsylvania State University
国际会议
天津
英文
2007-10-20（万方平台首次上网日期，不代表论文的发表时间）