Text semantic similarity calculation method and device and user terminal

A technology of semantic similarity and calculation method, which is applied in the field of device and user terminal, text semantic similarity calculation method, can solve the problem of single similarity information and low accuracy, and achieve the effect of rich information and high accuracy

Active Publication Date: 2018-09-07
ALIBABA (CHINA) CO LTD
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Most of the existing text semantic similarity matching focuses on the calculation of the single

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

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0032] Such as figure 2 It shows a flowchart of a method for calculating text semantic similarity provided by an embodiment of the present invention. See figure 2 , The method includes:

[0033] Step S110: Establish a first word vector matrix corresponding to the first text and a second word vector matrix corresponding to the second text. The first word vector matrix is ​​composed of the first word vector, and the second word vector matrix is ​​composed of the second word vector matrix. Word vector composition.

[0034] For the two texts that need to calculate the semantic similarity, they are the first text and the second text. First, establish a first word vector matrix corresponding to the first text according to the word vector of the first text, and establish a word vector matrix corresponding to the second text according to the word vector of the second text.

[0035] Further, such as image 3 As shown, this step S110 may include:

[0036] Step S111: Perform word segmentatio...

no. 2 example

[0080] This embodiment provides a text semantic similarity calculation device 200, please refer to Image 6 , The device 200 includes: a word vector matrix establishment module 210, configured to establish a first word vector matrix corresponding to the first text and a second word vector matrix corresponding to the second text, the first word vector matrix being composed of the first word vector The second word vector matrix is ​​composed of a second word vector; the similarity matrix establishment module 220 is used to establish the first text and the second word vector matrix according to the first word vector matrix, the second word vector matrix and a predetermined number of modalities. A multi-modal similarity matrix between the second texts, the multi-modal similarity matrix includes similarity measurement parameters; an optimization module 230, configured to iteratively optimize the multi-modal similarity matrix using an artificial neural network algorithm The similarity...

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 text semantic similarity calculation method and device and a user terminal and relates to the field of computers. The method comprises the steps that a first word vector matrix corresponding to a first text and a second word vector matrix corresponding to a second text are established; according to the first word vector matrix, the second word vector matrix and a predetermined modal number, a multi-modal similarity matrix of the first text and the second text is established; an artificial neural network algorithm is utilized to iteratively optimize similarity measurement parameters in the multi-modal similarity matrix, first word vectors, second word vectors and introduced artificial neural network parameters; and when a predetermined iteration ending condition ismet, an iteration output result is used as the semantic similarity between the first text and the second text. The multi-modal similarity matrix calculated through the method can be used for realizingtext semantic matching under a one-word-multiple-meaning condition.

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 used. These language processing tasks can usually be abstracted as the problem of text semantic similarity matching. [0003] Most of the existing text semantic similarity matching focuses on the calculation of the single semantic similarity of the text pair. The similarity information is single and the accuracy is not high. Summary of the invention [0004] In view of this, the embodiments of the present invention provide a method and device for calculating text semantic similarity, which builds a multimodal similarity matrix according to multiple semantics of words to measure the simi...

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
IPC IPC(8): G06F17/27G06F17/30
CPCG06F40/30
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