Paper: Experimental Results on the Native Language Identification Shared Task

ACL ID W13-1710
Title Experimental Results on the Native Language Identification Shared Task
Venue Innovative Use of NLP for Building Educational Applications
Session
Year 2013
Authors

We present a system for automatically iden- tifying the native language of a writer. We experiment with a large set of features and train them on a corpus of 9,900 essays writ- ten in English by speakers of 11 different lan- guages. our system achieved an accuracy of 43% on the test data, improved to 63% with improved feature normalization. In this paper, we present the features used in our system, de- scribe our experiments and provide an analysis of our results.