请教corpus size对 keyword comparison结果的影响

[greatlion]

我最近用wordsmith 4.o 在做keyword list的时候发现结果中keyness一栏会受到两个corpora规模大小的影响。我想问问，一般在做keyword list的时候，两个corpora在规模方面差距在多大范围内比较理想？ 比如1.5倍？ 还是多少？

 

[armstrong]

回复: 请教corpus size对 keyword comparison结果的影响

我最近用wordsmith 4.o 在做keyword list的时候发现结果中keyness一栏会受到两个corpora规模大小的影响。我想问问，一般在做keyword list的时候，两个corpora在规模方面差距在多大范围内比较理想？ 比如1.5倍？ 还是多少？

巴西的专家做过研究，一般认为reference语料库是你要研究的语料库的5倍以上是比较好的。

 

[ArthurW]

回复: 请教corpus size对 keyword comparison结果的影响

巴西的专家做过研究，一般认为reference语料库是你要研究的语料库的5倍以上是比较好的。

interesting. could you please show me which research it is, and where to read it? thanks

 

[armstrong]

回复: 请教corpus size对 keyword comparison结果的影响

interesting. could you please show me which research it is, and where to read it? thanks

several years ago, I read a paper about the size of the reference, but Now I can not find it.

 

[laohong]

回复: 请教corpus size对 keyword comparison结果的影响

这类文章有不少，有些已经在本坛贴过几次：

Whar are the requirements for the reference corpus?（该贴中包含有巴西学者Tony BERBER-SARDINHA的那篇文章Comparing corpora with WordSmith Tools: How large must the reference corpus be?）

如何判断一篇文章和一个语料库之间的相关性？ （该贴中包含有巴西学者Tony BERBER-SARDINHA的那篇文章Comparing corpora with WordSmith Tools: How large must the reference corpus be?）

G. Leech: The Importance of Reference Corpora

Methodological Considerations in the Determination of Corpus Size for the Study of Frequent Multi-Word Units (MWUs) in Spoken Language

The Effects of Corpus Size and Homogeneity on Language Model Quality


The Effect of Corpus Size on Case Frame Acquisition for Discourse Analysis

....

 

[ArthurW]

回复: 请教corpus size对 keyword comparison结果的影响

Many thanks, laohong!