|
Activity Number:
|
68
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Risk Analysis
|
| Abstract - #305512 |
|
Title:
|
Application of Kernel Methods to Fraud Detection
|
|
Author(s):
|
Ravi Mallela*+
|
|
Companies:
|
Equbits
|
|
Address:
|
1844 Camino Pablo, Moraga, CA, 94556,
|
|
Keywords:
|
SVM ; kernel ; fraud ; financial ; services
|
|
Abstract:
|
Kernel Methods were introduced in the early 1990s and have become a popular method for solving classification and regression modeling problems. Over the last few years, a number of applications of kernel methods have been presented. This paper assesses the applicability of kernel methods to challenging modeling problems in the financial services industry. We will provide an overview of kernel methods and new knowledge-based approaches that allow modelers to extract meaningful business rules from SVMs. An example assessing merchant risk management will be used for illustration.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
