Intelligent official document recommendation method based on federal learning

The invention provides an intelligent official document recommendation method based on federal learning. The method comprises the following steps: determining an official document type of a target official document text and an official document type corresponding to each official document sub-databa...

Full description

Saved in:
Bibliographic Details
Main Authors CHENG JIANRUN, HUANG HAIFENG, CHEN LUN, HUANG TIECHUN, ZHOU WEI, XIAO YI, HAN GUOQUAN
Format Patent
LanguageChinese
English
Published 25.10.2022
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
HANDLING RECORD CARRIERS
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
Online AccessGet full text

Cover

More Information
Summary:The invention provides an intelligent official document recommendation method based on federal learning. The method comprises the following steps: determining an official document type of a target official document text and an official document type corresponding to each official document sub-database; determining a federal learning architecture corresponding to each official document sub-database; broadcasting the federated learning architecture to each official document sub-database by a federated learning aggregation server; and after federal learning training, outputting an official document recommendation result corresponding to the target official document text. According to the intelligent official document recommendation method provided by the invention, the official document recommendation model is obtained by adopting federal learning architecture training, so that the problem of training corpus missing caused by non-intercommunication of data of each official document sub-database is solved; meanwh
Bibliography:Application Number: CN202211154292