The Knowledge Discovery Laboratory (KDL) is a research group in the School of Computer Science. KDL has developed statistical methods that help the National Association of Securities Dealers (NASD) identify securities brokers who are likely to commit fraud. Data from public records are analyzed for social, professional, and organizational attributes and connections that may indicate possible fraudulent activity. These methods helps field examiners rank brokers by their propensity for fraud by using sophisticated statistical models learned directly from data.
Program Impact Region:
Program Start (and End Dates):
August 1, 2005
National Association of Securities Dealers
1735 K Street NW
Washington, DC 20006United States
Program Director's Email Address:
UNIT: Colleges, Schools, Departments, Centers, and Institutes:
College of Natural Sciences/Computer Science
This Project Receives External Funding:
Scholarly Product Type:Report
Scholarly Product Detail:
J. Neville, Ö. Simsek, D. Jensen, J. Komoroske, K. Palmer, and H. Goldberg (2005). Using relational knowledge discovery to prevent securities fraud. Proc. of the 11th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
L. Friedland and D. Jensen. (2007). Finding tribes: Identifying close-knit individuals from employment patterns. Proc. of The 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
A. Fast, L. Friedland, M. Maier, B. Taylor, D. Jensen, H. Goldberg, & J. Komoroske (2007). Relational data pre-processing techniques for improved securities fraud detection. Proc. of The 13th ACM SIGKDD Intern. Conf. on Knowledge Discovery and Data Mining