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

A Fusion Method and Device for Sentence Vectors

A fusion method and sentence vector technology, applied in the field of sentence vector fusion method and device, can solve the problems of destroying the semantics of multiple word vectors, affecting the expressive ability of text features, etc., and achieve the effect of improving expressive ability

Inactive Publication Date: 2019-03-22
HANGZHOU JIUYAN TECH CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present invention provides a sentence vector fusion method and device to solve the problem that the existing sentence vector fusion technology will destroy the semantics of multiple word vectors and affect the ability to express text features

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
  • A Fusion Method and Device for Sentence Vectors
  • A Fusion Method and Device for Sentence Vectors
  • A Fusion Method and Device for Sentence Vectors

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0020] figure 1 It is a flow chart of a sentence vector fusion method provided by Embodiment 1 of the present invention. The method of this embodiment can be specifically applied to the situation where sentence vectors are fused in a terminal or a server to extract text sentence vectors from the text to be processed, and it is also applicable to the identification of target information in the text. The method of this embodiment can be executed by a sentence vector fusion device, which can be independently configured in a terminal or a server, or can be distributed in a terminal and a server, and both cooperate to implement the method of this embodiment.

[0021] The method of this embodiment includes:

[0022] S110. Extracting text word vectors included in the text to be processed;

[0023] Generally speaking, the most simple and direct representation of text features is a single word, but because text data contains many words, and some words appear frequently, they are not ...

Embodiment 2

[0058] figure 2 It is a schematic structural diagram of a sentence-vector fusion device provided in Embodiment 2 of the present invention. Such as figure 2 As shown, the device includes:

[0059] Text word vector extraction module 210, for extracting the text word vector included in the text to be processed;

[0060] The second word vector generation module 220 is used to search the text word vector in the set corpus to generate the second word vector;

[0061] The text sentence vector generating module 230 is configured to determine a text sentence vector corresponding to the text word vector according to the spatial similarity between the text word vector and the second word vector.

[0062] The technical solution provided by the embodiment of the present invention can effectively avoid destroying the intrinsic semantic information of individual word vectors by merging multiple sets of word vectors in the text into sentence vectors according to the spatial similarity be...

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 discloses a sentence vector fusion method and apparatus. The method comprises: extracting a text word vector comprised in a to-be-processed text; searching a set corpus for the text word vector, to generate a second word vector; and according to a space similarity between the text word vector and the second word vector, determining a text sentence vector corresponding to the text word vector. According to the technical scheme provided by embodiments of the invention, multiple word vectors in the text are fused into a sentence vector according to the space similarity between the text word vector and the second word vector; destruction of intrinsic semantic information of the individual word vector is effectively avoided; and sentence vector fusion is performed in conjunction with semantics of preceding and following sentences according to a specific application scenario of the text, thereby improving the capability of the sentence vector in expressing the to-be-processed text.

Description

technical field [0001] The invention relates to the technical field of network security, in particular to a sentence vector fusion method and device. Background technique [0002] With the rapid development of the Internet and mobile networks, more and more users choose to communicate with others and share information through the Internet platform, such as through websites or terminal application software. Accompanying it will also produce a lot of content that does not conform to the safe use environment of the Internet, or even violates national laws and regulations, such as politically sensitive, obscene and pornographic content, etc., resulting in the risk of safe operation of related websites; / The exposure of products will promote their own products in various Internet environments, making the user experience of the website or application software extremely poor, and even fraudulent use of advertising information, etc., which has brought great harm to Internet securit...

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): G06F17/24G06F17/27
CPCG06F40/166G06F40/211
Inventor 吕志高邹国平
Owner HANGZHOU JIUYAN 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