uima-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Yasen Kiprov <yasenkip...@yahoo.com>
Subject drools for annotation sequences
Date Mon, 19 Nov 2012 09:06:22 GMT
Hello,

Roberto, thank you for mentioning the drools wrapper for UIMA, drools seems to be a very mature
and stable system.

I'm evaluating drools for rule based text analysis but I'm missing a key concept. How do you
model annotation sequences? For instance I want to write a rule: 

If "Mr." is followed by two capitalized tokens and next token isn't "ltd."
then we have a person name.

I'm not quite familiar with drools and I can only think of adding a custom "index" feature
to the tokens and write something like:

when 
Token with text == Mr. and number i
Token with capital letter and number i + 1
...

But it doesn't look right.

Does anyone have any idea how such patterns can be modeled with Drools?

Thank you,
Yasen

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message