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

Method and device for deep text matching based on word migration learning

A deep and text-based technology, applied in the field of deep text matching methods and devices based on word migration learning, can solve problems affecting model matching effects and achieve the effect of promoting matching accuracy

Active Publication Date: 2021-07-02
ULTRAPOWER SOFTWARE
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] This application provides a deep text matching method and device based on word migration learning to solve the problem that the existing deep matching model parameters are random initialization parameters, which affect the matching effect of the trained 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 deep text matching based on word migration learning
  • Method and device for deep text matching based on word migration learning
  • Method and device for deep text matching based on word migration learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0063] Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with aspects of the invention as recited in the appended claims.

[0064] Aiming at the problem that the existing model parameters are random initialization parameters, which affect the matching effect of the model, this embodiment provides a basic flowchart of a deep text matching method based on word migration learning, wherein the method can be applied to various Deep matching model.

[0065] figure 1 It is a schematic flowchart of a deep text matching met...

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

This application provides a deep text matching method and device based on word migration learning. When training the deep matching model, the BERT model is fused and pre-trained; then, using the pre-trained BERT model, the input sentence is paired with The sentences in are represented by the initial word vectors, and then the sentences in the sentences represented by the initial word vectors are similarly weighted to obtain the weighted sentence vectors; finally, according to the loss value corresponding to the similarity value of the sentence vectors, adjust Model parameters, use the parameters to adjust the final deep matching model, and perform text matching on the input sentences. Since the parameters of the pre-trained BERT model are no longer randomly initialized parameters, and part-of-speech prediction is added to the pre-trained BERT model, the semantic information of word vectors is enriched. Therefore, using the trained BERT model to express the semantics of the sentences in the sentence pair with word vectors is more accurate, which promotes the improvement of the matching accuracy of the trained model.

Description

technical field [0001] The present application relates to the technical field of natural language processing, in particular to a deep text matching method and device based on word migration learning. Background technique [0002] Text matching is an important basic problem in natural language processing, and many tasks in natural language processing can be abstracted as text matching tasks. For example, webpage search can be abstracted as a correlation matching problem between webpage and user search query, automatic question answering can be abstracted as a matching problem between candidate answers and questions, and text deduplication can be abstracted as a text-to-text similarity matching problem. [0003] Traditional text matching techniques (such as the vector space model algorithm in information retrieval) mainly solve the matching problem at the vocabulary level. In fact, the matching algorithm based on lexical coincidence has great limitations and cannot solve many...

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 Patents(China)
IPC IPC(8): G06F16/33G06F40/289G06F40/30
CPCG06F40/289G06F40/30
Inventor 李健铨刘小康晋耀红
Owner ULTRAPOWER SOFTWARE
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