Title:
DATA ANOMALY DETECTION METHOD AND DEVICE
Document Type and Number:
WIPO Patent Application WO/2021/114831
Kind Code:
A1
Abstract:
A data anomaly detection method and a device, used for servers. Said method comprises: receiving a clinical data set and a test knowledge set, and generating a clinical test data set on the basis of the clinical data set and the test knowledge set (101); implementing training using the clinical test data set as the training data of a pre-set similarity sorting model, so as to obtain a trained target similarity sorting model (102); receiving data to be examined and inputting the data to be examined into the trained target similarity sorting model, so as to obtain multiple pieces of test knowledge and determining, on the basis of the multiple pieces of test knowledge, whether the clinical test conclusion corresponding to the data to be examined is in an anomalous state (103).
Inventors:
LI YANXUAN (CN)
LIAO XIYANG (CN)
SUN XINGZHI (CN)
LIAO XIYANG (CN)
SUN XINGZHI (CN)
Application Number:
PCT/CN2020/118419
Publication Date:
June 17, 2021
Filing Date:
September 28, 2020
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G16H10/20
Domestic Patent References:
| WO2004093822A2 | 2004-11-04 |
Foreign References:
| CN110808095A | 2020-02-18 | |||
| CN105260588A | 2016-01-20 | |||
| CN105787232A | 2016-07-20 |
Attorney, Agent or Firm:
SCIHEAD IP LAW FIRM (CN)
Download PDF: