Title:
DATA CLASSIFICATION METHOD AND SYSTEM BASED ON MACHINE LEARNING
Document Type and Number:
WIPO Patent Application WO/2019/105258
Kind Code:
A1
Abstract:
A data classification method based on machine learning, comprising: forming a first batch of multiple classification models corresponding to the first batch of multiple machine learning algorithms (S10); respectively performing classification calculation on a first data feature by using each classification model (S11); establishing an Nth batch of multiple iterative models, each iterative model performing learning training on the basis of a second data feature (S12); respectively performing classification calculation on the second data feature by using each iterative model (S13); measuring the proximity of the second classification result to the expected classification result (S14); if the proximity satisfies the first condition (S15), the method ends; otherwise, iteratively performing the steps set by the iterative mode. The method can achieve the effect of "class deep learning", which significantly improves the universality and accuracy of data classification. Meanwhile, the data classification method has high credibility and low level of manual intervention.
Inventors:
HUANG ZILI (CN)
YANG YANG (CN)
CHEN ZHOU (CN)
ZHU HAORAN (CN)
YANG YANG (CN)
CHEN ZHOU (CN)
ZHU HAORAN (CN)
Application Number:
PCT/CN2018/116370
Publication Date:
June 06, 2019
Filing Date:
November 20, 2018
Export Citation:
Assignee:
CHINA UNIONPAY CO LTD (CN)
International Classes:
G06K9/62
Foreign References:
| CN108021940A | 2018-05-11 | |||
| CN107301221A | 2017-10-27 | |||
| CN102324046A | 2012-01-18 | |||
| CN105260746A | 2016-01-20 | |||
| US7362892B2 | 2008-04-22 |
Attorney, Agent or Firm:
CHINA PATENT AGENT (HK) LTD. (CN)
Download PDF:
Previous Patent: SEMI-CIRCULAR ZIPPER AND SLIDERS USED COOPERATIVELY
Next Patent: LENS SYSTEM, IMAGING APPARATUS, AND MOVING OBJECT
Next Patent: LENS SYSTEM, IMAGING APPARATUS, AND MOVING OBJECT