Title:
ANOMALY DETECTION METHOD AND DETECTION PROGRAM
Document Type and Number:
WIPO Patent Application WO/2017/155292
Kind Code:
A1
Abstract:
The present invention relates to a method and a program for detecting anomalies. A method for detecting anomalies according to one embodiment of the present invention comprises: a collection data reception step (S100) for receiving one or more collection data by a computer; a step (S200) for calculating result data for each collection data through a deep neural network in the computer; a pattern information derivation step (S300) for deriving pattern information at the time of calculating the result data for each collection data; a group designation step (S400) for designating one or more collection data having the same pattern information as a specific group; a characteristic assignment step (S500) for assigning a characteristic of each group; and a step (S700) for calculating new result data after receiving new collection data, and extracting a group corresponding to the new result data. According to the present invention, it is possible to prevent a group corresponding to an anomaly from being mistaken as being included in a group corresponding to a normal behavior.
Inventors:
CHOE JEONG YEOL (KR)
Application Number:
PCT/KR2017/002480
Publication Date:
September 14, 2017
Filing Date:
March 08, 2017
Export Citation:
Assignee:
INFINIGRU CORP (KR)
International Classes:
G06F21/55; G06F17/30; G06N3/08; G06Q40/02
Foreign References:
| KR20130126814A | 2013-11-21 | |||
| KR20040092314A | 2004-11-03 | |||
| KR20150091775A | 2015-08-12 | |||
| KR20080084132A | 2008-09-19 |
Other References:
KIM, JONG-HUN ET AL.: "Big Data Analysis Based on Deep Learning for Baseball Game Data", THE JOURNAL OF KOREAN INSTITUTE OF COMMUNICATIONS AND INFORMATION SCIENCES 2015 AUTUMN CONFERENCE, November 2015 (2015-11-01), pages 263 - 266, XP055413948
Attorney, Agent or Firm:
YOO, Cheol Hyun (KR)
Download PDF: