Method for improving digital translation quality based on post-processing technology

A translation quality and digital technology, applied in the field of machine translation, can solve problems such as unsatisfactory digital translation effects, and achieve the effect of small computer burden, fast operation speed and transparent structure

Active Publication Date: 2019-06-11
沈阳雅译网络技术有限公司
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In order to solve the above-mentioned technical problems, the object of the present invention is to provide a method for improving the quality of digital translation based on post-processing technology, so as to solve the problems caused by the inaccurate translation of complex numbers that existing machine translation systems face when performing digital translation. Unsatisfactory translation effect

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
  • Method for improving digital translation quality based on post-processing technology
  • Method for improving digital translation quality based on post-processing technology
  • Method for improving digital translation quality based on post-processing technology

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0049] Example 1: The result of a Chinese sentence after splitting using the subword segmentation method is as follows:

[0050] Before the split: the total number of processors reached 164

[0051] After splitting: the total number of processing @@ machines reaches 16@@4

[0052] As shown in Example 1, "handling machine" is an unregistered word in Chinese. Through the subword segmentation method, "handling machine" is split into "handling" and "machine", and the two "handling" and "machine" The part is precisely in the vocabulary, which can be translated accurately, so as to obtain the correct translation result of the "processor".

[0053] However, for the problem of digital translation, most numbers are also unregistered words, and complex numbers will be split into multiple subword units through subword segmentation, which cannot solve the problem of digital translation errors.

example 2

[0054] Example 2: The result of splitting a complex number using the subword segmentation method is as follows:

[0055] Before the split: 37.0687 million

[0056] After splitting: 3,@@70@@6.@@8@@7 million

[0057] Translation: 3@@7.@@0@@67million

[0058] Since complex numbers are split into multiple sub-word units, some sub-word units are often mistranslated during machine translation, resulting in low-quality digital translation.

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

A method for improving digital translation quality based on post-processing technology is provided and comprises the steps of replacing complex numbers in a sentence to be translated with simple numbers, and original numbers are recorded; Performing word segmentation processing and sub-word segmentation processing on the replaced sentence to be translated; Translating the sentences after each sub-word is segmented; Obtaining the attention alignment information of the sentence to be translated and the translated sentence, and obtaining a simple digital translation corresponding to the simple digital; Calculating a correct number translated text through the original number, the simple number and the simple number translated text; Replacing the simple digital translated text with the correctdigital translated text according to the corresponding relationship between the simple digital translated text and the simple digital translated text to obtain a correct translated sentence; And unitnormalization operation is carried out on the digital part and the corresponding unit in the correct translated sentence. According to the method, the digital translation problem is solved through apost-processing method, complex numbers are replaced with simple numbers, then translation reduction is conducted, the operation speed is high, and the burden on a computer is small.

Description

technical field [0001] The invention belongs to the technical field of machine translation, and in particular relates to a method for improving the quality of digital translation based on post-processing technology. Background technique [0002] Digital translation is a common translation problem in machine translation. Digital translation specifically refers to mapping the number part in the source language (content to be translated) to the number in the target language (content to be translated), where the units are different between different languages. Numbers are also represented differently. When the numerical part of a language is translated into the target language, there are often cases such as unit changes. An example of digital translation is shown below: [0003] Source: The daily demand for crude oil this year is 98.85 million barrels [0004] Target language: Demand for crude oil this year is 98.85 million barrels a day [0005] When users use machine trans...

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): G06F17/28G06F17/27
Inventor 王强张哲旸肖桐朱靖波
Owner 沈阳雅译网络技术有限公司
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