HOO 2012 Error Recognition and Correction Shared Task: Cambridge University Submission Report
View / Open Files
Authors
Kochmar, Ekaterina
Andersen, Oeistein
Briscoe, Edward
Publication Date
2012-06-01Journal Title
http://aclweb.org/anthology/W12-2028
Conference Name
Seventh Workshop on Innovative Use of NLP for Building Educational Applications
Publisher
Association for Computational Linguistics
Volume
Proceedings of the Seventh Workshop on Building Educational Applications Using NLP
Issue
Anthology: W12-2028
Pages
242-250
Language
English
Type
Conference Object
This Version
VoR
Metadata
Show full item recordCitation
Kochmar, E., Andersen, O., & Briscoe, E. (2012). HOO 2012 Error Recognition and Correction Shared Task: Cambridge University Submission Report. http://aclweb.org/anthology/W12-2028, Proceedings of the Seventh Workshop on Building Educational Applications Using NLP (Anthology: W12-2028), 242-250. https://doi.org/10.17863/CAM.9671
Abstract
Previous work on automated error recognition and correction of texts written by learners of English as a Second Language has demonstrated experimentally that training classifiers on error-annotated ESL text generally outperforms training on native text alone and that adaptation of error correction models to the native language (L1) of the writer improves performance. Nevertheless, most extant models have poor precision, particularly when attempting error correction, and this limits their usefulness in practical applications requiring feedback. We experiment with various feature types, varying quantities of error-corrected data, and generic versus L1-specific adaptation to typical errors using Naïve Bayes (NB) classifiers and develop one model which maximizes precision. We report and discuss the results for 8 models, 5 trained on the HOO data and 3 (partly) on the full error-coded Cambridge Learner Corpus, from which the HOO data is drawn.
Sponsorship
We thank Cambridge ESOL, a division of Cambridge Assessment for a partial grant to the first author and a research contract with iLexIR Ltd. We also thank them and Cambridge University Press for granting us access to the CLC for research purposes.
Identifiers
This record's DOI: https://doi.org/10.17863/CAM.9671
This record's URL: https://www.repository.cam.ac.uk/handle/1810/264263
Rights
Attribution-NonCommercial-ShareAlike 4.0 International, Attribution-NonCommercial-ShareAlike 4.0 International, Attribution-NonCommercial-ShareAlike 4.0 International