US 11,861,454 B2
Method for detecting anomaly in time series data and computing device for executing the method
Gyeong Moon Park, Gyeonggi-do (KR); and A Hyung Shin, Gyeonggi-do (KR)
Assigned to University-Industry Cooperation Group of Kyung Hee University, Gyeonggi-do (KR)
Filed by University-Industry Cooperation Group of Kyung Hee University, Gyeonggi-do (KR)
Filed on Aug. 1, 2022, as Appl. No. 17/878,181.
Claims priority of application No. 10-2021-0175107 (KR), filed on Dec. 8, 2021.
Prior Publication US 2023/0179496 A1, Jun. 8, 2023
Int. Cl. G06N 3/0895 (2023.01); H04L 43/067 (2022.01); H04L 43/0817 (2022.01); G06F 18/214 (2023.01); G01W 1/10 (2006.01); A61B 5/024 (2006.01)
CPC G06N 3/0895 (2023.01) [A61B 5/02422 (2013.01); G01W 1/10 (2013.01); G06F 18/214 (2023.01); H04L 43/067 (2013.01); H04L 43/0817 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A method of detecting an abnormality in time series data, the method being performed in a computing device including one or more processors and a memory storing one or more programs executed by the one or more processors, the method comprising:
first masking to cover a portion of input time series data with a mask;
generating first-restored time series data in which the time series data is restored by inputting the first-masked time series data to a generator;
calculating a difference between the first-restored time series data and original time series data;
second masking to cover a portion of the time series data with a mask on basis of the calculated difference; and
generating second-restored time series data in which the time series data is restored by inputting the second-masked time series data to the generator.