Optimizing parameters for machine translation

A technology based on users and times, applied in natural language translation, program control using stored programs, instruments, etc., can solve time-consuming and costly problems, achieve efficient running time, increase flexibility, and increase the number of effects

Inactive Publication Date: 2011-08-10
GOOGLE LLC
View PDF8 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Human translation of text by human 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
  • Optimizing parameters for machine translation
  • Optimizing parameters for machine translation
  • Optimizing parameters for machine translation

Examples

Experimental program
Comparison scheme
Effect test

Example

[0125] Other implementations are possible. In particular, additional refinements can be performed to improve the performance of MERT (for lattices). For example, to prevent linear optimization techniques from getting stuck at poor local optima, MERT can explore additional starting points chosen randomly by sampling the parameter space. As another example, the range of weights for some or all feature functions can be limited by defining weight constraints. In particular, for the characteristic function h m The weight constraints for can be specified as the interval

[0126] R m =[l m , r m ], l m , r m ∈R∪{-∞,+∞}, which defines the feature function weight λ m The tolerance region from which to choose. If linear optimization is performed subject to weight constraints, γ is chosen such that:

[0127] l 1 M ≤ λ 1 M + γ · d 1...

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

Methods, systems, and apparatus, including computer program products, for language translation are disclosed. In one implementation, a method is provided. The method includes accessing a hypothesis space, where the hypothesis space represents a plurality of candidate translations; performing decoding on the hypothesis space to obtain a translation hypothesis that minimizes an expected error in classification calculated relative to an evidence space; and providing the obtained translation hypothesis for use by a user as a suggested translation in a target translation.

Description

technical field [0001] This specification deals with statistical machine translation. Background technique [0002] Manual translation of text by human operators can be time-consuming and expensive. One goal of machine translation is to automatically translate text in a source language into an equivalent text in a target language. There are several different approaches to machine translation, including example-based machine translation and statistical machine translation. Statistical machine translation attempts to identify the most likely translation in a target language for a particular input in a source language. For example, when translating a sentence from French to English, statistical machine translation identifies the most likely English sentence to the French sentence. The maximum possible translation can be expressed as: [0003] arg max e P ( e | ...

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): G06F17/00G06F17/20G06F9/06G06F11/00G06F40/00
CPCG06F17/2818G06F40/44G06F17/10
Inventor 沃尔夫冈·马赫赖尚卡尔·库马尔罗伊·W·特朗布利弗朗茨·约瑟夫·欧池伊格纳西奥·E·塞耶雅各布·乌兹科瑞特
Owner GOOGLE LLC
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