Paper: An Explicit Feedback System for Preposition Errors based on Wikipedia Revisions

ACL ID W14-1810
Title An Explicit Feedback System for Preposition Errors based on Wikipedia Revisions
Venue Innovative Use of NLP for Building Educational Applications
Session
Year 2014
Authors

This paper presents a proof-of-concept tool for providing automated explicit feed- back to language learners based on data mined from Wikipedia revisions. The tool takes a sentence with a grammatical er- ror as input and displays a ranked list of corrections for that error along with evi- dence to support each correction choice. We use lexical and part-of-speech con- texts, as well as query expansion with a thesaurus to automatically match the er- ror with evidence from the Wikipedia revi- sions. We demonstrate that the tool works well for the task of preposition selection errors, evaluating against a publicly avail- able corpus.