Title:
METHOD FOR TARGET OBJECT BEHAVIOR PREDICTION OF DATA OFFSET AND RELATED DEVICE THEREOF
Document Type and Number:
WIPO Patent Application WO/2022/126961
Kind Code:
A1
Abstract:
The present invention relates to the field of artificial intelligence, and relates to a method for target object behavior prediction of data offset and a related device thereof. The method comprises: obtaining feature variables, and performing preprocessing and feature screening to generate a first feature variable set and a second feature variable set; performing weight assignment on the first feature variable set, obtaining a third feature variable set according to the second feature variable set, and performing weight assignment on the third feature variable set; performing model training according to the first feature variable set and the corresponding weights and outputting the training result, and performing model training according to the first feature variable set, the third feature variable set and the corresponding weights to obtain the training result; and when the comparison result of the two training results meets a preset condition, outputting a second training model for behavior prediction. The present invention further relates to a blockchain technology, and the behavior prediction result can be stored in a blockchain. According to the present invention, the model risk can be reduced while increasing the model prediction precision.
Inventors:
ZHANG QIAOLI (CN)
LIN RONGJI (CN)
LIN RONGJI (CN)
Application Number:
PCT/CN2021/090162
Publication Date:
June 23, 2022
Filing Date:
April 27, 2021
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06K9/62
Foreign References:
| CN112508118A | 2021-03-16 | |||
| CN111178639A | 2020-05-19 | |||
| CN109784373A | 2019-05-21 | |||
| CN112990583A | 2021-06-18 | |||
| US20180348717A1 | 2018-12-06 |
Attorney, Agent or Firm:
SL INTELLECTUAL PROPERTY CO., LTD. (CN)
Download PDF: