Title:
SYSTEM, METHOD, AND COMPUTER PROGRAM FOR CREDIT ASSESSMENT BY USING MACHINE LEARNING
Document Type and Number:
WIPO Patent Application WO/2021/060593
Kind Code:
A1
Abstract:
Provided is a credit assessment system comprising: a missing data prediction unit using machine learning to receive an input data set including a value of at least a part of items among at least one item related to a financial service application and at least one item relating to an applicant, generate, from an input vector corresponding to the input data set, a hidden vector in the level lower than that of the input vector, correct the hidden vector on the basis of hidden vector distribution, and output, from the corrected hidden vector, a reconstruction vector including a prediction value of a missing value among values of at least one item of the input data set; and a credit assessment processing unit for performing credit assessment for the application on the basis of the reconstruction vector.
Inventors:
AUM SUNG MIN (KR)
Application Number:
PCT/KR2019/012791
Publication Date:
April 01, 2021
Filing Date:
October 01, 2019
Export Citation:
Assignee:
DATA REPUBLIC INC (KR)
International Classes:
G06Q40/02; G06N20/00; G06Q10/06
Foreign References:
| KR20180119674A | 2018-11-02 | |||
| KR20180069452A | 2018-06-25 | |||
| KR20090030144A | 2009-03-24 | |||
| KR20120134849A | 2012-12-12 | |||
| US20140229158A1 | 2014-08-14 |
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: