Paper: NAIST at 2013 CoNLL Grammatical Error Correction Shared Task

ACL ID W13-3604
Title NAIST at 2013 CoNLL Grammatical Error Correction Shared Task
Venue International Conference on Computational Natural Language Learning
Session shared task
Year 2013
Authors

This paper describes the Nara Institute of Science and Technology (NAIST) er- ror correction system in the CoNLL 2013 Shared Task. We constructed three sys- tems: a system based on the Treelet Lan- guage Model for verb form and subject- verb agreement errors; a classifier trained on both learner and native corpora for noun number errors; a statistical machine translation (SMT)-based model for prepo- sition and determiner errors. As for subject-verb agreement errors, we show that the Treelet Language Model-based approach can correct errors in which the target verb is distant from its subject. Our system ranked fourth on the official run.