Paper: Detecting Missing Hyphens in Learner Text

ACL ID W13-1739
Title Detecting Missing Hyphens in Learner Text
Venue Innovative Use of NLP for Building Educational Applications
Year 2013

We present a method for automatically de- tecting missing hyphens in English text. Our method goes beyond a purely dictionary-based approach and also takes context into account. We evaluate our model on artificially gener- ated data as well as naturally occurring learner text. Our best-performing model achieves high precision and reasonable recall, making it suitable for inclusion in a system that gives feedback to language learners.