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

Technical document term interpretation generation method and device based on GAN network

A technology for generating technical documents and generating devices, applied in the fields of word processing, deep learning and neural networks, it can solve the problems that technical documents do not have professional terminology explanations and the quality of terminology explanations is not high, and achieve the effect of filtering and evaluating professional terms.

Active Publication Date: 2019-09-17
INSPUR GROUP CO LTD
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In recent years, with the rapid development of the Internet, companies have shared the relevant technical documents of their products through the Internet to provide users with better technical services. The explanation of technical terms is a very important part of this. However, the actual technical documents Many have no terminology explanation section or the quality of terminology explanation is not high

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
  • Technical document term interpretation generation method and device based on GAN network
  • Technical document term interpretation generation method and device based on GAN network
  • Technical document term interpretation generation method and device based on GAN network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] The technical document term explanation generation method based on the GAN network of the present invention, the method is to use the GAN network to train the term explanation generator G and the term explanation discriminator D, add the term discriminator Dz, and judge whether the term can be explained in this type of technical document , to achieve better filtering and evaluation of technical terms, and to classify the generated terminology explanations through the terminology explanation classifier C, so that the input professional terminology vector Vi can better meet the user's personalized requirements for professional terminology explanations, and generate High-quality technical terminology explanation; the specific steps are as follows:

[0047] S1, generate the training of the term network model, as attached figure 2 As shown, the specific steps are as follows:

[0048] S101. Collect technical document training data sets, perform data preprocessing and data l...

Embodiment 2

[0065] as attached figure 1 As shown, the technical document term interpretation generation device based on GAN network of the present invention includes a term discriminator Dz, a term vector generator Gv, a term interpretation generator G, a term interpretation discriminator D and a term interpretation classifier C; term The discriminator Dz makes judgments based on the professional term corpus and rules produced by using technical documents, and feeds the results back to the expensive term vector generator Gv; the term vector generator Gv forms a professional Term vector Vi, denoted as Vi(Si, Ri); among them, Si represents a technical term; Ri represents the requirement for term interpretation; the term interpretation generator G generates a professional term explanation according to the term vector Vi; the term interpretation discriminator D judges the production Whether the term in time is true, reasonable, accurate, and effective, and at the same time judge whether the t...

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 technical document term interpretation generation method and device based on a GAN network, and belongs to the technical field of character processing, deep learning and neural networks. The technical problem to be solved by the invention is how to efficiently and accurately generate professional term interpretation for technical documents by effectively utilizing the GAN network technology. The adopted technical scheme is as follows: the method comprises the steps that a GAN network is used for training a term interpretation generator G and a term interpretation discriminator D, a term discriminator Dz is added, whether a term can be interpreted in a technical document of the type or not is judged, a professional term is better filtered and evaluated, and generated term interpretation is classified through a term interpretation classifier C; the method comprises the following specific steps: S1, training a term network model; and S2, generating term interpretation by using the network model. The device comprises a term discriminator Dz, a term vector generator Gv, a term interpretation generator G, a term interpretation discriminator D and a term interpretation classifier C.

Description

technical field [0001] The present invention relates to the technical fields of word processing, deep learning and neural network, in particular to a method and device for generating technical document terminology explanations based on GAN network. Background technique [0002] Generative Adversarial Networks (GAN, Generative Adversarial Networks) is a deep learning model, originally proposed by Ian Goodfellow, and is one of the most important methods for unsupervised learning on complex distributions in recent years. The model produces high-quality output through the mutual game learning of the two modules in the framework, the Generator (Generator) and the Discriminator (Discriminator), and trains the Discriminator and the Generator in turn to make them fight against each other and sample from complex probability distributions. Finally, the training of the neural network is completed. At present, GAN networks have been widely used in the field of computer vision. [0003...

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): G06K9/62G06F17/27
CPCG06F40/30G06F18/24G06F18/214
Inventor 孙善宝张桂红徐驰于玲
Owner INSPUR GROUP 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