Title:
FRAUDULENT TRANSACTION DETECTION METHOD BASED ON SAMPLE CLUSTERING
Document Type and Number:
WIPO Patent Application WO/2017/143932
Kind Code:
A1
Abstract:
A fraudulent transaction detection method based on sample clustering. The method can identify a fraudulent transaction with greater accuracy. The method comprises: collecting and establishing, on the basis of a sample clustering algorithm, a training sample group and a testing sample group (A1); using the training sample group and the testing sample group to construct and test a plurality of neural network models (A2); and detecting, on the basis of the plurality of trained and tested neural network models, a fraudulent transaction (A3).
Inventors:
ZHAO JINTAO (CN)
ZENG LIBIN (CN)
HOU LING (CN)
QIU XUETAO (CN)
GE MINGMING (CN)
ZENG LIBIN (CN)
HOU LING (CN)
QIU XUETAO (CN)
GE MINGMING (CN)
Application Number:
PCT/CN2017/073579
Publication Date:
August 31, 2017
Filing Date:
February 15, 2017
Export Citation:
Assignee:
CHINA UNIONPAY CO LTD (CN)
International Classes:
G06F17/30
Foreign References:
CN105787743A | 2016-07-20 | |||
CN104850868A | 2015-08-19 | |||
CN104504362A | 2015-04-08 | |||
US20050192824A1 | 2005-09-01 |
Other References:
LIU, ZHAOLING ET AL.: "The Research of Bank Card Fraud Based on BP Neural Network", BUSINESS CHINA, 31 December 2009 (2009-12-31)
Attorney, Agent or Firm:
CHINA PATENT AGENT (HK)LTD. (CN)
Download PDF: