Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method and device for optimizing language semantic comprehension model, equipment and storage medium

A semantic understanding and language technology, applied in natural language data processing, semantic analysis, electronic digital data processing, etc., can solve problems such as few language semantic understanding models, complex training of language semantic understanding models, and low robustness of language semantic understanding models, etc. question

Pending Publication Date: 2021-01-08
BEIJING SINOVOICE TECH CO LTD
View PDF0 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above objective facts lead to fewer training samples for training the language semantic understanding model in the intelligent customer service system, resulting in complex training of the language semantic understanding model and the low robustness of the trained language semantic understanding model.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for optimizing language semantic comprehension model, equipment and storage medium
  • Method and device for optimizing language semantic comprehension model, equipment and storage medium
  • Method and device for optimizing language semantic comprehension model, equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

[0071] FAQ task (Frequently Asked Questions) is used in the business scenario of intelligent customer service to automatically answer business knowledge questions frequently asked by users. It is the core user scenario of intelligent customer service, which can significantly reduce the number and cost of manual customer service. For example, in 10086's online intelligent customer service, if a user asks "how to check the phone bill", the system can automatically give a cor...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a method and device for optimizing a language semantic comprehension model, equipment and a storage medium, and aims to improve the robustness of the language semantic comprehension model. The method comprises the steps of obtaining a first loss value of a language semantic comprehension model in target training, wherein the target training is training of finely adjusting the language semantic comprehension model by utilizing question sample data in a target field; adding disturbance to an input layer of the language semantic comprehension model according to the first loss value; obtaining a second loss value of the language semantic comprehension model added with the disturbance in the target training; and updating parameters of the language semantic comprehension model according to the second loss value to obtain an optimized language semantic comprehension model.

Description

technical field [0001] The present application relates to the technical field of natural language understanding, in particular to a method, device, equipment and storage medium for optimizing a language semantic understanding model. Background technique [0002] The language semantic understanding model is often used in the FAQ (Frequently Asked Questions) task, that is, in the intelligent customer service business. It is used to convert the questions entered by the user into semantic vectors, so as to further determine the classification of the questions entered by the user according to the semantic vectors, and then according to the classification, Answers to user-entered questions are determined in a knowledge base. [0003] However, in the application scenario of the intelligent customer service business, the domain of the knowledge base is vertically closed, depending on the size of the business scope that the company or individual that develops the intelligent customer...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/332G06F16/35G06F40/211G06F40/30
CPCG06F16/3329G06F16/35G06F40/30G06F40/211
Inventor 李健李天阳武卫东陈明
Owner BEIJING SINOVOICE TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com