ppke-nlpg / SS05

The original SS05 algorithm from Hong Shen and Anoop Sarkar used in the paper 'Voting Between Multiple Data Representations for Text Chunking'

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SS05

The original SS05 algorithm from Hong Shen and Anoop Sarkar used in the paper 'Voting Between Multiple Data Representations for Text Chunking'

We would like to thank professor Anoop Sarkar for his cooperation and providing the original Perl code and letting us to use and distribute it with our own. This helped a lot to understand their algorithm better.

Warning

This code is unmaintained, outdated and contains serveral bugs (see bugtracker).

This method is outperformed by 'Gut, Besser, Chunker – Selecting the best models for text chunking with voting' by Balázs Indig and István Endrédy For a more general and up-to-date implementation, which outperforms this method see https://github.com/ppke-nlpg/gut-besser-chunker

Reference

If you use this tool, please cite the following paper that were published at Springer:

@inproceedings{conf/ai/ShenS05,
  author	= {Hong Shen and Anoop Sarkar},
  editor	= {Bal{\'{a}}zs K{\'{e}}gl and Guy Lapalme},
  title 	= {Voting Between Multiple Data Representations for Text Chunking},
  booktitle = {Advances in Artificial Intelligence, 18th Conference of the Canadian
           	Society for Computational Studies of Intelligence, Canadian {AI} 2005,
           	Victoria, Canada, May 9-11, 2005, Proceedings},
  series	= {Lecture Notes in Computer Science},
  volume	= {3501},
  pages 	= {389--400},
  publisher = {Springer},
  year  	= {2005},
  url   	= {\url{http://dx.doi.org/10.1007/11424918_40}},
  doi   	= {10.1007/11424918_40},
  timestamp = {Thu, 19 May 2005 10:15:37 +0200},
  biburl	= {\url{http://dblp.uni-trier.de/rec/bib/conf/ai/ShenS05}},
}

About

The original SS05 algorithm from Hong Shen and Anoop Sarkar used in the paper 'Voting Between Multiple Data Representations for Text Chunking'

License:GNU Lesser General Public License v3.0


Languages

Language:Perl 56.4%Language:Lex 39.9%Language:Batchfile 3.7%