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

Neural machine translation NMT model creation method and system

A machine translation and model technology, applied in the field of neural machine translation NMT model creation, can solve problems such as errors, missing proper nouns or rare terms, and achieve the effect of meeting translation needs and improving accuracy and quality

Active Publication Date: 2019-07-02
HENAN UNIVERSITY
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the large differences between professional field texts and general field texts in terms of sentence morphology, syntax, word order, and thesaurus size, it is not uncommon to make mistakes or omit translations of proper nouns or rare terms in professional field texts

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
  • Neural machine translation NMT model creation method and system
  • Neural machine translation NMT model creation method and system
  • Neural machine translation NMT model creation method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] see figure 1 , the method for creating a neural machine translation NMT model provided by an embodiment of the present invention includes the following steps:

[0034] S101, using the crawler technology, obtain a certain number of common Chinese and English sentences from network resources, and generate a common corpus.

[0035] S102: Using the crawler technology, obtain a certain number of Chinese-English texts in the subject category from multiple sets of Chinese-English e-books on information subjects, serialize the Chinese-English texts, and perform the Chinese-English comparison sentences of different lengths in turn. Adjust, translate the adjusted sequence, match the obtained translation result with the corresponding Chinese or English, and set the sentences whose similarity is greater than the set threshold as a professional corp...

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 neural machine translation NMT model creation method and system. The method comprises steps of by using the crawler technology, obtaining a certain number of general Chineseand English comparison sentences from the network resources; generating a generic corpus, utilizing crawler technology, obtaining a certain number of subject class Chinese and English contrast texts from a plurality of sets of Chinese and English contrast information subject class electronic books and serializing the Chinese and English books; adjusting the subject class Chinese and English control statements with different lengths in sequence; translating the adjusted sequence; performing similarity matching on the obtained translation result and the corresponding Chinese or English; settingsentences of which the similarity is greater than a set threshold value as a professional corpus so as to generate a professional corpus, the sequence-to-sequence Seq2Seq model is trained by the aid of the universal corpus and the professional corpus, an NMT model is established, text in the professional field can be translated accurately, translation quality is improved, and translation requirements of people on the text in the professional field are met.

Description

technical field [0001] The invention relates to the technical field of text processing, in particular to a method and system for creating a neural machine translation NMT model. Background technique [0002] Existing neural machine translation systems (such as Google, Baidu, Sogou, etc.) are highly versatile, and the translation quality and speed meet people's translation needs for general-purpose texts to a certain extent. However, due to the fact that the texts in the specialized field are quite different from the general field texts in terms of sentence morphology, syntax, word order, and thesaurus size, it is not uncommon to see errors or omissions of proper nouns or rare terms in the translation of specialized field texts. SUMMARY OF THE INVENTION [0003] In order to solve the deficiencies of the prior art, the embodiments of the present invention provide a method and system for creating a neural machine translation NMT model. [0004] In a first aspect, the method ...

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): G06F17/28G06F16/36G06F16/951G06F16/953G06F16/9535
CPCG06F40/49
Inventor 李涵张东生韩昊天刘纯燕
Owner HENAN UNIVERSITY
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com