Title:
CLASSIFICATION MODEL TRAINING, SEMANTIC CLASSIFICATION METHOD, DEVICE, EQUIPMENT, AND MEDIUM
Document Type and Number:
Japanese Patent JP2023160746
Kind Code:
A
Abstract:
To provide a classification model training method, a semantic classification method, a device, equipment and a medium for improving the versatility of a model and coping with a problem in which samples are unbalanced.SOLUTION: A classification model training method includes: a step S101 of obtaining a sample query template and a label category of a prediction waiting category in the sample query template, where the sample query template is constructed on the basis of a sample query phrase and the number of categories waiting prediction; a step S102 of inputting the sample query template into a pre-constructed semantic classification model to obtain a sample semantic category of the prediction waiting category; and a step S103 of training the semantic classification model according to the sample semantic category and the label category of the prediction waiting category.SELECTED DRAWING: Figure 1A
Inventors:
LIN ZENAN
QIN HUAPENG
ZHAO MIN
ZHANG GUOXIN
LV YAJUAN
QIN HUAPENG
ZHAO MIN
ZHANG GUOXIN
LV YAJUAN
Application Number:
JP2023050552A
Publication Date:
November 02, 2023
Filing Date:
March 27, 2023
Export Citation:
Assignee:
BEIJING BAIDU NETCOM SCIENCE & TECNOLOGY CO LTD
International Classes:
G06F40/216; G06F16/35; G06F40/247; G06F40/30
Attorney, Agent or Firm:
Hideki Miyata
Hiroaki Sakai
Hiroaki Sakai