Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD AND SYSTEM FOR PREDICTION OF SYSTEM FAILURE
Document Type and Number:
WIPO Patent Application WO/2022/107935
Kind Code:
A1
Abstract:
Disclosed are a method and system for predicting a failure in a system, based on machine learning. A training method for prediction of a system failure disclosed comprises the steps of: training a plurality of failure prediction models by using first training data including system data regarding a target system; determining a retraining failure prediction model candidate group on the basis of performance evaluation values for the failure prediction models; and retraining failure prediction models included in the retraining failure prediction model candidate group, by using second training data.

Inventors:
NO JAE CHUN (KR)
KWON SE HOON (KR)
SEO HYUNG JUN (KR)
PARK SUNG-SOON (KR)
PARK MUN SIK (KR)
KIM KYEUNG PYO (KR)
Application Number:
PCT/KR2020/016514
Publication Date:
May 27, 2022
Filing Date:
November 20, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
GLUESYS CO LTD (KR)
UNIV SEJONG IND ACAD COOP FOUD (KR)
FALINUX CO LTD (KR)
International Classes:
G06Q10/04; G06F11/30; G06N20/00; G06Q10/00; G06Q10/06
Foreign References:
KR20200051343A2020-05-13
KR101984730B12019-06-03
KR20200072588A2020-06-23
US20160132787A12016-05-12
US20150178638A12015-06-25
Attorney, Agent or Firm:
MIN, Young Joon (KR)
Download PDF: