search

clulab / eidos

Machine reading system for World Modelers
Apache License 2.0
37 stars 24 forks source link

GeoNormFinder.find with locations found with back-off rules #698

Open maxaalexeeva opened 4 years ago

maxaalexeeva commented 4 years ago

Becky's update to GeoNormFinder applies it to all the locations that are found with a gazetteer. It does not seem to work with the locations found with the rules here (even if the entities found are renamed to just Location instead of Location-Expand): https://github.com/clulab/eidos/blob/migration_schema-zupon/src/main/resources/org/clulab/wm/eidos/english/grammars/context.yml#L30

For now, it does not matter much bc all the relevant locations have been added to gazetteer, but it would be good to make the finder apply even to the locations not in the gazetteer.

BeckySharp commented 4 years ago

we can def work on it, and if necessary we can also refactor the normalizer code to use it later

On Wed, Oct 23, 2019 at 9:49 PM Maria (Masha) Alexeeva < notifications@github.com> wrote:

Becky's update to GeoNormFinder applies it to all the locations that are found with a gazetteer. It does not seem to work with the locations found with the rules here (even if the entities found are renamed to just Location instead of Location-Expand):

https://github.com/clulab/eidos/blob/migration_schema-zupon/src/main/resources/org/clulab/wm/eidos/english/grammars/context.yml#L30

For now, it does not matter much bc all the relevant locations have been added to gazetteer, but it would be good to make the finder apply even to the locations not in the gazetteer.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/clulab/eidos/issues/698?email_source=notifications&email_token=ABJCPCLJSU3GOLY2GDYOQELQQESN7A5CNFSM4JEOWXO2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HT774AA, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJCPCP44Y7QATBN3GNZP73QQESN7ANCNFSM4JEOWXOQ .

maxaalexeeva commented 4 years ago

sounds good! This doesn't seem to be very crucial right now, but I do want to get back to it at some point, so I figured I'd just start an issue for future reference.

BeckySharp commented 4 years ago

@maxaalexeeva what's the status on this? same?

maxaalexeeva commented 4 years ago

I don't believe we have addressed this.

BeckySharp commented 4 years ago

to make sure I understand -- the rules apply during the main running of the grammars, right? I think (all of?) the finders run ahead of time. If I'm right, then do we just want to use the normalization of the finder on locations found elsewhere? Or have I misunderstood?

maxaalexeeva commented 4 years ago

to make sure I understand -- the rules apply during the main running of the grammars, right? I think so. I think (all of?) the finders run ahead of time. Is this where the order of finders is defined? https://github.com/clulab/eidos/blob/14fcd19528f93d120bdf39f003edad83d33d8331/src/main/resources/eidos.conf#L7 If yes, based on this order, the rule-found locations should also be normalized. Or maybe I misunderstand.

Something we can look at during wed meeting, @BeckySharp?

BeckySharp commented 4 years ago

yes, good plan!

On Monday, January 20, 2020, Maria (Masha) Alexeeva < notifications@github.com> wrote:

to make sure I understand -- the rules apply during the main running of the grammars, right? I think so. I think (all of?) the finders run ahead of time. Is this where the order of finders is defined? https://github.com/clulab/ eidos/blob/14fcd19528f93d120bdf39f003edad83d33d8331/src/main/resources/ eidos.conf#L7 If yes, based on this order, the rule-found locations should also be normalized. Or maybe I misunderstand.

Something we can look at during wed meeting, @BeckySharp https://github.com/BeckySharp?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/clulab/eidos/issues/698?email_source=notifications&email_token=ABJCPCMYAKSAHT6R56D5X6LQ6YZZPA5CNFSM4JEOWXO2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJOB6CI#issuecomment-576462601, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJCPCNFZUVDSXR7B6T6ZG3Q6YZZPANCNFSM4JEOWXOQ .