Title:
DEEP LEARNING-BASED EVALUATION METHOD AND APPARATUS
Document Type and Number:
WIPO Patent Application WO/2020/022639
Kind Code:
A1
Abstract:
Provided is a deep learning-based evaluation method. The evaluation method according to an embodiment of the present invention, performed by an evaluation apparatus, comprises: a step of receiving an input of class information of an evaluation target; a step of generating first evaluation base data associated with the input class information by using a generative adversarial network (GAN)-based deep learning model; and a step of evaluating a value of the evaluation target by applying the first evaluation base data to an evaluation model. Here, the first evaluation base data comprises a virtual data sample.
More Like This:
| WO/2013/032444 | METHODS AND SYSTEMS FOR INTEGRATED CONTROL OF SUBTERRRANEAN OPERATIONS |
| JP2002109095 | SCHEDULE MANAGEMENT SYSTEM |
| JP2005293005 | JOURNALIZING SYSTEM |
Inventors:
PARK HYUN WOO (KR)
JUN SEUNG PYO (KR)
LEE JONG TAIK (KR)
KIM BYUNG HOON (KR)
JUN SEUNG PYO (KR)
LEE JONG TAIK (KR)
KIM BYUNG HOON (KR)
Application Number:
PCT/KR2019/006911
Publication Date:
January 30, 2020
Filing Date:
June 07, 2019
Export Citation:
Assignee:
KOREA INST SCI & TECH INF (KR)
International Classes:
G06Q10/06; G06N3/08
Foreign References:
| KR20150051969A | 2015-05-13 | |||
| KR101843066B1 | 2018-05-15 | |||
| KR20010035256A | 2001-05-07 | |||
| JP2016224821A | 2016-12-28 | |||
| US20170351952A1 | 2017-12-07 |
Attorney, Agent or Firm:
KASAN IP & LAW FIRM (KR)
Download PDF: