[Gate-users] Pronominal Coreferencer

[Lesley Dokter]

Hi,

I have the following 4 sentences in a GATE document corpus:

This set of release notes has been prepared by Richard Smith. He is the WLI tech lead.
Richard Smith prepared these release notes. He is the WLI Tech Lead.

I ran the ANNIE pipeline application over this  with the pronominal coreferencer added at the end of the pipeline. I have marked up in red and underlined in the text below to show the resulting Person annotations.

This set of release notes has been prepared by Richard Smith. He is the WLI tech lead.
Richard Smith prepared these release notes. He is the WLI Tech Lead.

Also below are the Person annotations as output by GATE:

Person          277     292     14286   {gender=male, matches=[14286, 14287, 14288, 5632, 14495, 14496], rule=PersonFinal, rule1=PersonFull}
Person          321     336     14287   {gender=male, matches=[14286, 14287, 14288, 5632, 14495, 14496], rule=PersonFinal, rule1=PersonFull}
Person          367     369     14495   {ENTITY_MENTION_TYPE=PRONOUN, antecedent_offset=321, matches=[14286, 14287, 14288, 5632, 14495, 14496]}

My question is this:

Why is the coreference between the person 'Richard Smith and the 'He' pronoun only drawn in second two sentences?
The only difference that I can see is that the first sentence is a passive construction - which might confuse the coreferencer?

Any help is appreciated

Regards
Lesley







________________________________
-------------------------------
This email and any attachments may contain information that is confidential and subject to legal privilege. If you are not the intended recipient, any use, dissemination, distribution or duplication of this email and attachments is prohibited. If you have received this email in error please notify the author immediately and erase all copies of the email and attachments. The Ministry of Social Development accepts no responsibility for changes made to this message or attachments after transmission from the Ministry.
-------------------------------