search

gregdurrett / berkeley-entity

The Berkeley Entity Resolution System jointly solves the problems of named entity recognition, coreference resolution, and entity linking with a feature-rich discriminative model.
GNU General Public License v3.0
185 stars 35 forks source link

Couldn't parse even with backoff parser! #16

Open asterbini opened 5 years ago

asterbini commented 5 years ago

Hi, I am trying retrieving the entities in Austen's Pride and Prejudice and I get the error: Couldn't parse even with backoff parser! on a sentence in Chapter 6:

"No, indeed. I do not wish to avoid the walk. The distance is nothing, when one has a motive; only three miles. I shall be back by dinner."

The error instead does NOT happen if the sentence is unquoted:

No, indeed. I do not wish to avoid the walk. The distance is nothing, when one has a motive; only three miles. I shall be back by dinner.

What could be the problem? How could I fix it?

asterbini commented 5 years ago

For the moment I have solved by replacing all double quote (") with backquote (`)

gregdurrett commented 5 years ago

This is probably because the "input sentence" is actually several sentences. It needs to be split further for the parser to do the right thing.

BTW I don't really maintain this tool anymore, I'm glad you find it useful but AllenNLP is a more up-to-date and better supported framework.

asterbini commented 5 years ago

Thanks for your kind and rapid answer I'll try AllenNLP Andrea

On 8/27/19 5:57 PM, Greg Durrett wrote:

This is probably because the "input sentence" is actually several sentences. It needs to be split further for the parser to do the right thing.

BTW I don't really maintain this tool anymore, I'm glad you find it useful but AllenNLP is a more up-to-date and better supported framework.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/gregdurrett/berkeley-entity/issues/16?email_source=notifications&email_token=AAZZ6MJA4JYDH2GOQLXOQB3QGVFHDA5CNFSM4IPQ3LNKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD5IHPHQ#issuecomment-525367198, or mute the thread https://github.com/notifications/unsubscribe-auth/AAZZ6MJLSSJVMFGSKPIUBKTQGVFHDANCNFSM4IPQ3LNA.