The BILOU chunking scheme (Beginning, Inside and Last tokens of multi-token chunks, Unit-length chunks and Outside) outperform BIO:
Lev Ratinov and Dan Roth. 2009. Design challenges and misconceptions in named entity recognition. In Proceedings of the Thirteenth Conference on Computational Natural Language Learning (CoNLL '09). Association for Computational Linguistics, Stroudsburg, PA, USA, 147-155.
We should implement BILOU in the new org.cleartk.classifier.chunking package.
Comment #1
Posted on Jul 24, 2012 by Grumpy Rabbit(No comment was entered for this change.)
Comment #2
Posted on Feb 17, 2013 by Swift Monkey(No comment was entered for this change.)
Comment #3
Posted on Apr 1, 2013 by Swift HippoSince I wanted to experiment with it, here is a patch.
- bilou.patch 9KB
Comment #4
Posted on May 3, 2013 by Grumpy Rabbit(No comment was entered for this change.)
Comment #5
Posted on Oct 24, 2013 by Grumpy Rabbit(No comment was entered for this change.)
Comment #6
Posted on Oct 30, 2013 by Grumpy RabbitThis issue was closed by revision 8a46b95bbe0f.
Status: Fixed
Labels:
Type-Enhancement
Priority-Medium
Component-ml
Milestone-2.0