Text semantic similarity calculation method, device and user terminal

A technology of semantic similarity and calculation method, which is applied in the field of devices and user terminals, and text semantic similarity calculation method, which can solve the problems of word vector burden and large capacity of word vector, and achieve the effect of reducing the difficulty of storage

Active Publication Date: 2022-08-05
ALIBABA (CHINA) CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Existing text semantic similarity matching is based on word vectors of words in the text for similarity calculation. However, due to the thousands of combinations of words, the storage of word vectors takes up a huge capacity, and the words are constantly expanding. The storage of word vectors brings additional burden

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
  • Text semantic similarity calculation method, device and user terminal
  • Text semantic similarity calculation method, device and user terminal
  • Text semantic similarity calculation method, device and user terminal

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0031] like figure 2 A flowchart of a method for calculating text semantic similarity provided by an embodiment of the present invention is shown. See figure 2 , the method includes:

[0032] Step S110: Establish a first character vector matrix according to the first character vector corresponding to each character in the first text, and establish a second character vector matrix according to the second character vector corresponding to each character in the second text.

[0033] For the two texts whose semantic similarity needs to be calculated, they are the first text and the second text, respectively. First, a first character vector matrix corresponding to the first text is established according to the character vector corresponding to each character of the first text, and a character vector matrix corresponding to the second text is established according to the character vector corresponding to each character of the second text.

[0034] Specifically, in this step, as...

no. 2 example

[0079] This embodiment provides an apparatus 200 for calculating text semantic similarity, please refer to Image 6 , the device 200 includes:

[0080] The character vector establishment module 210 is used for establishing the first character vector matrix according to the first character vector corresponding to each character in the first text, and establishing the second character vector matrix according to the second character vector corresponding to each character in the second text; The optimization module 220 is configured to use an artificial neural network algorithm to optimize the first character vector in the first character vector matrix and the second character vector in the second character vector matrix. During the optimization process, the first character vector matrix and all The parameters are shared between the second character vector matrices; the similarity calculation module 230 is used to calculate the semantic similarity between the first text and the se...

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 present invention provides a method, device, and user terminal of text synonymous similarity, involving the computer field.The method includes: Create the first character vector according to the first character vector corresponding to each character in the first text, and establish the second character vector matrix according to the second character vector corresponding to each character in the second text; use artificial nerves to use artificial nervesNetwork algorithm, the first character vector in the first character lines shared by the parameter sharing, and the second character vector in the second character lines matrix; according to the optimized first character vector and the optimized second character vector, calculate the first firstSemantic similarity between a text and the second text.This method is calculated by similarity through character vectors. Compared with the word vector, due to the limited number of characters, it not only reduces the difficulty of storing the preset character vector list, but also exempts the needs of the word vector pre -training for additional large statements.Essence

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular, to a method, device and user terminal for calculating text semantic similarity. Background technique [0002] With the rapid development of Internet technology, language processing tasks such as information retrieval, automatic question answering, and machine translation are increasingly being applied. These language processing tasks can usually be abstracted as the problem of text semantic similarity matching. [0003] The existing text semantic similarity matching is based on the word vector of the words in the text to calculate the similarity, but due to the thousands of word combinations, the storage of the word vector occupies a huge capacity, and the words are constantly expanding. The storage of word vectors brings additional burden. SUMMARY OF THE INVENTION [0004] In view of this, the embodiments of the present invention provide a method and apparatus fo...

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): G06F40/30G06F40/289
CPCG06F40/289G06F40/30
Inventor 孟令勋王嘉勋
Owner ALIBABA (CHINA) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap