关于英汉平行句子的单词对齐

[okaygan]

预料库的对齐算法大都是句子级别的。
小弟琢磨着怎么实现单词级别的对齐算法，大家不妨一起讨论一下啊，众人的力量才是强大的~~呵呵
我先抛出一个问题啊。。
she is a beautiful girl in the beautiful world!
她是这个美丽世界里的一个漂亮女孩。

在对齐中，怎么利用位置信息把“美丽”对齐到第二个beautiful，而把“漂亮”对齐到第一个beautiful？

 

[xiaoz]

回复: 关于英汉平行句子的单词对齐

Contextual infromation such as collocations in source and target languages might be of help in such cases.

 

[okaygan]

回复: 关于英汉平行句子的单词对齐

Contextual infromation such as collocations in source and target languages might be of help in such cases.

利用上下文信息，比如“美丽”修饰“世界”。第二个“beautiful”修饰 “world”
只要先找把“世界”对齐到“world” ，就能把“美丽”对齐到“beautiful” ，isnot it？
但是问题远不是如此简单。
例：
这个美丽的女孩比那个美丽的女孩聪明。
this beautiful girl is clever than that beautiful girl.
后面的上下文信息也比较相似，如何处理呢？

 

[xiaoz]

回复: 关于英汉平行句子的单词对齐

Word alignment is of course much more complicated than that. The following paper might serve as a good starting point for research in this area:

Piao, Scott Songlin (2002) Word Alignment in English–Chinese Parallel Corpora. Literary and Linguistic Computing 17(2):207-230.

 

[xialiang]

回复: 关于英汉平行句子的单词对齐

Thanks a lot, am going to read that too.

 

[greatsong]

回复: 关于英汉平行句子的单词对齐

Word alignment is of course much more complicated than that. The following paper might serve as a good starting point for research in this area:

Piao, Scott Songlin (2002) Word Alignment in English–Chinese Parallel Corpora. Literary and Linguistic Computing 17(2):207-230.

可否麻烦Xiaoz上传此文？谢谢！

 

[xiaoz]

回复: 关于英汉平行句子的单词对齐

http://llc.oxfordjournals.org/cgi/content/abstract/17/2/207

 

[xujiajin]

Here you go.

 

[greatsong]

回复: 关于英汉平行句子的单词对齐

Thanks a lot, Xiaoz and Jiajin!

 

[armstrong]

Thanks, Dr.Xiao and Dr.Xu.