Chinese grammar error correction method based on generative adversarial network

A grammatical error, network technology, applied in the field of information processing, can solve problems such as sentences that do not conform to Chinese habits

Inactive Publication Date: 2019-10-08
BEIJING UNIV OF POSTS & TELECOMM
View PDF7 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, most of the work only uses the corpus, and does not solve the problem that the corrected sentences of the model do not conform to Chinese habits.

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
  • Chinese grammar error correction method based on generative adversarial network
  • Chinese grammar error correction method based on generative adversarial network
  • Chinese grammar error correction method based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] Next, embodiments of the present invention will be described in more detail.

[0013] figure 1 It is a schematic flow chart of the steps of the Chinese grammatical error correction method based on the generative confrontation network provided by the present invention, including:

[0014] Step S1: generate a network to generate a correction sentence;

[0015] Step S2: optimize the generated network;

[0016] Step S3: the discriminative network discriminates the source of the sentence correction;

[0017] Step S4: Optimizing the discriminant network;

[0018] Step S5: iterative optimization;

[0019] Each step will be described in detail below:

[0020] Step S1: Generate a network to generate correction sentences. The present invention first establishes a mapping dictionary from words to word vector numbers in the network encoding layer and decoding layer, and maps each word in the text to a corresponding word number. Create a word vector table for the network enco...

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 Chinese grammar error correction method based on a generative adversarial network, and belongs to the field of information processing. The method is characterized by comprising the following steps of: generating a correction statement by utilizing a generation network; calculating a loss function by utilizing the discriminant network, and optimizing the generation network; judging a sentence correction source by utilizing a judgment network; optimizing the discrimination network; and the generation network and the discrimination network are iteratively optimized continuously. Through the generative adversarial network, the Chinese grammar error correction effect is improved, and the method has great use value.

Description

technical field [0001] The invention relates to the field of information processing, in particular to a neural network-based Chinese grammatical error correction method. Background technique [0002] Chinese grammatical error correction is a relatively new task in Chinese natural language processing. The purpose is to judge whether the sentences written by non-Chinese native speakers contain grammatical errors, and propose corrections for the places that contain errors. [0003] At present, there are two most common methods of correcting Chinese grammatical errors. One is to use the Chinese grammatical error detection model to detect errors first, and then use the N-Gram dictionary to calculate the co-occurrence frequency of words to obtain corrected sentences. The other is to build an end-to-end Chinese grammatical error correction model using a sequence-to-sequence model. This model treats the grammatical error correction task as a translation task, translating sentences...

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): G06F17/27G06N3/04
CPCG06F40/253G06N3/045
Inventor 赵建博李思孙忆南梁景贵朱勇杰吕游伟
Owner BEIJING UNIV OF POSTS & TELECOMM
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