Difference between revisions of "Brill, CL 1995"

From Cohen Courses
Jump to navigationJump to search
(Created page with '== Citation == Duame, H., Langford, J., and Marcu, D. 2009. Search-based structured prediction. Machine Learning. 75. 3. p297-325 == Online version == [http://www.umiacs.umd.e…')
 
Line 1: Line 1:
 
== Citation ==
 
== Citation ==
  
Duame, H., Langford, J., and Marcu, D. 2009. Search-based structured prediction. Machine Learning. 75. 3. p297-325
+
Brill, E. 1995. Transformation-based error-driven learning and natural language processing: a case study in part-of-speech tagging. Computational Linguistics. 21. 4. p543-565
  
 
== Online version ==
 
== Online version ==
  
[http://www.umiacs.umd.edu/~hal/docs/daume09searn.pdf Search-based structured prediction]
+
[http://www.ldc.upenn.edu/acl/J/J95/J95-4004.pdf Transformation-based error-driven learning and natural language processing: a case study in part-of-speech tagging]
  
 
== Summary ==
 
== Summary ==
  
This journal [[Category::paper]] introduces [[UsesMethod::SEARN]], a meta-algorithm that combines searching and learning to make structured predictions. Note that this is the journal version of the 2006 paper that introduced this method.
+
This journal [[Category::paper]] introduces ...
  
The algorithm is summarized in the following figure, from page 6 of the paper:
+
The algorithm is summarized in the following figure:
  
[[File:searn-algorithm.png]]
+
[[File:temp.png]]
  
 
The key points from the paper are:
 
The key points from the paper are:
* [[UsesMethod::SEARN]] is an algorithm that solves complex structured predictions with minimal assumptions on the structure of the output and loss function
+
* [[UsesMethod::Transformation Based Learning]] is an algorithm that...
* Their experiments show that [[UsesMethod::SEARN]] is competitive with existing standard structured prediction algorithms on sequence labeling tasks.
+
* Their experiments show that ..
* Authors described the use of [[UsesMethod::SEARN]] on [[AddressesProblem::text summarization]], which yielded state-of-the-art performance.
+
* Authors described the use of [[UsesMethod::Transformation Based Learning]] on [[AddressesProblem::POS Tagging]]...
  
 
== Related papers ==
 
== Related papers ==
  
* '''SEARN in Practice''': This unpublished manuscript showcases three example problems where SEARN can be used - [[RelatedPaper::Daume_et_al,_2006]].
+
*

Revision as of 13:40, 31 October 2010

Citation

Brill, E. 1995. Transformation-based error-driven learning and natural language processing: a case study in part-of-speech tagging. Computational Linguistics. 21. 4. p543-565

Online version

Transformation-based error-driven learning and natural language processing: a case study in part-of-speech tagging

Summary

This journal paper introduces ...

The algorithm is summarized in the following figure:

File:Temp.png

The key points from the paper are:

Related papers