Paper: Improved Correction Detection in Revised ESL Sentences

ACL ID P14-2098
Title Improved Correction Detection in Revised ESL Sentences
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014
Authors

This work explores methods of automat- ically detecting corrections of individual mistakes in sentence revisions for ESL students. We have trained a classifier that specializes in determining whether consecutive basic-edits (word insertions, deletions, substitutions) address the same mistake. Experimental result shows that the proposed system achieves an F 1 -score of 81% on correction detection and 66% for the overall system, out-performing the baseline by a large margin.