Machine translation method based on blocking mechanism

A machine translation and mechanism technology, applied in the field of machine translation, can solve problems such as readability defects and long sentences, and achieve the effect of improving readability, accuracy and translation effect

Inactive Publication Date: 2018-11-06
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is: the present invention provides a machine translation method based on the block mechanism, which makes up for the existing machine translation based o...

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
  • Machine translation method based on blocking mechanism
  • Machine translation method based on blocking mechanism
  • Machine translation method based on blocking mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] A machine translation method based on a block mechanism, the specific steps are as follows:

[0046]1: Block Encoder encoding, first dynamically segment each sentence to be translated according to the segment code encoded by the block Encoder, then apply the reply neural network to convert the segmented clause blocks into vectors, and finally generate the corresponding sentences to be translated Wherein, the separator is a custom symbol applied in the segmentation step of the block Encoder encoding, and the block Encoder encoding converts the sentence to be translated into a corresponding vector as follows:

[0047] 1.1: Dynamically segment each sentence to be translated according to the separator. After segmentation, each sentence corresponds to multiple clause blocks; the specific steps for dynamically segmenting each sentence to be translated according to the separator are as follows:

[0048] 1.1.1: Specify the delimiter as a sign to divide the clause block;

[004...

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 machine translation method based on a blocking mechanism and belongs to the technical field of machine translation. The method comprises the following steps of S1, convertinga to-be-translated statement into a vector corresponding to the to-be-translated statement through utilization of a blocking encoder; and S2, decoding the vector corresponding to the to-be-translatedstatement through utilization of a decoder, and generating a machine translation result corresponding to the to-be-translated statement through utilization of a Beam Search algorithm based on a language model. According to the method, the problem that long sentence translation defect, readability defect and translation omission defect exist in the translation result due to the fact that the existing machine translation based on a reply type neural network is relatively long in statement, independent in word and expression and misaligned in rigid word is solved, and a machine translation effect is improved.

Description

technical field [0001] The invention relates to the field of machine translation, in particular to a machine translation method based on a block mechanism. Background technique [0002] Translation is the process of converting source language sentences into target language sentences. It is an important means to break through the barriers between different languages, and it is very important in the communication between different cultures. With the development of society, people's demand for translation has increased significantly, human translation is not enough to meet the demand, and the demand for machine translation has increased and its importance has become prominent. [0003] Neural network-based machine translation methods directly use neural networks to model translation in an end-to-end manner, including reply-based neural networks, convolutional neural networks (Convolutional Neural Network, CNN) and feedforward neural networks (Feedforward Neural Network), etc. ...

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
IPC IPC(8): G06F17/28G06N3/04
CPCG06F40/58G06N3/045
Inventor 屈鸿黄鹂王淼张晓敏刘洋军张亦洲程绍欢
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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