Error mining in parsing results

Introduction

This program implements error mining in parsing results, as described by Sagot and De la Clergerie, 2006, with extensions proposed by De Kok and Van Noord, 2009. In addition to unigrams and bigrams, larger problematic ngrams can also be mined.

A viewer for comfortably browsing and analyzing the mining results is also included.

Download

You can download the source code using git:

git clone git://github.com/rug-compling/errormining.git

Please refer to the manual for compilation and usage instructions.

Bibliography

• Benoît Sagot and Éric de la Clergerie, Error mining in parsing results, in ACL–44: Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
• Daniël de Kok, Jianqiang Ma and Gertjan van Noord, A generalized method for iterative error mining in parsing results, in: ACL2009 Workshop Grammar Engineering Across Frameworks (GEAF), Singapore, 2009

Disclaimer

The Fadd library that is included with the error miner is written by Jan Daciuk, and licensed under the GNU Lesser General Public License.