统计机器翻译解码器 [英] Statistical machine translator decoder

问题描述

有人可以给我一个统计机器翻译解码器的分步教程吗.在其中，它是在翻译模型中使用基于短语的翻译来实现的，以便我了解其基本知识和步骤.

解析过程，启发式搜索可能的候选翻译等.

Can somebody give me a step by step tutorial of a statistical machine translator decoder.. in which it is implemented using phrase-based translation in the translation model.. so that i can understand its basic and steps

The parsing process,heuristic searching of possible candidate translations etc.

推荐答案

http://cseweb.ucsd.edu/~dkauchak/mt-tutorial/ [ ^ ]

看看第一个基本链接.

http://www.cs.jhu.edu/~jason/465/PowerPoint/lect32b-mt-decoding.pdf [

http://cseweb.ucsd.edu/~dkauchak/mt-tutorial/[^]

Take a look at the first link for the basics.

http://www.cs.jhu.edu/~jason/465/PowerPoint/lect32b-mt-decoding.pdf[^]

Take a look at this for an example

谢谢您，先生！您知道有没有专门集中在解码的教程，因为似乎在该教程中没有讨论对短语进行标记化的过程..它是二元论吗?一次两个字?我的意思是解码器将如何知道该短语是否为包含两个要翻译的单词的组?

而且，启发式搜索在基本方式上也没有得到很好的讨论..

thank you sir! is there any tutorials you know that has concentrates in decoding, because it seems the process of tokenizing of the phrase is not discussed in the tutorials.. is it bigram? two words at a time? I mean how will the decoder know if the phrase is is group with two words to be translated?

Also, the heuristic search is not well discussed in a basic way..

这篇关于统计机器翻译解码器的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！