Who knows the accuracy rate of the GoTagger?

[singer]

I am eager to know the answer.Thanks for your reply.

 

[laohong]

why not ask the developer directly?
http://uluru.lang.osaka-u.ac.jp/~k-goto/main-mail.html

 

[singer]

yeah.It is a good idea.Thanks ,Laohong.

 

[laohong]

It'd be appreciated if you could post the answer to your inquiry here later.

 

[singer]

No reply until now.

 

[andrewcjc]

I am aslo eager to know it.

 

[laohong]

When you got the answer, do you really trust it? How accurate should be accurate enough?

 

[清风出袖]

Accuracy is forever a relative matter. In my mind, it is only a rough indicator.

 

[singer]

The compiler replied :

Dear sir,

Thank you for your interest in my corpus-tool.
The accuracy of my tagger is almost the same as one of Brill's tagger which was
developed by Eric Brill. Brill tagger achieved approximately 95% accuracy rate,
so you can estimate the same rate with my tagger. The rate is variable depending
on text genres, though.
Thank you.

Sincerly,

--------------------------------------------------
Osaka University, Japan
GOTO Kazuaki
fortune@myrealbox.com
http://uluru.lang.osaka-u.ac.jp/~k-goto/index.html

 

[laohong]

Thanks for the update. Yes, as always, "the rate is variable depending on text genres"

 

[清风出袖]

太极拳打得太好了！

 

[ginger]

93%

 

[xujiajin]

最新一期《外语教学与研究》上将有梁茂成教授关于GoTagger（内核是Brill）和CLAWS标注正确率的比较的讨论。

 

[armstrong]

a piece of good news!