Paper: Discriminating Non-Native English with 350 Words

ACL ID W13-1713
Title Discriminating Non-Native English with 350 Words
Venue Innovative Use of NLP for Building Educational Applications
Year 2013

This paper describes MITRE?s participation in the native language identification (NLI) task at BEA-8. Our best effort performed at an ac- curacy of 82.6% in the eleven-way NLI task, placing it in a statistical tie with the best per- forming systems. We describe the variety of machine learning approaches that we ex- plored, including Winnow, language model- ing, logistic regression and maximum-entropy models. Our primary features were word and character n-grams. We also describe several ensemble methods that we employed for com- bining these base systems.