Using Classifier Features for Studying the Effect of Native Language on the Choice of Written Second Language Words
Oren Tsur and Ari Rappoport
In: ACL 2007 Workshop on Cognitive Aspects of Computational Language Acquisition, 21 June - 30 June 2007, Prague, Czech Republic.

Abstract

We apply modern statistical NLP techniques to study language transfer, a major issue in the theory of Second Language Acquisition (SLA). Using an SVM for the problem of native language classification, we show that a careful analysis of the effects of various features can lead to substantial scientific insights. In particular, we demonstrate that character bi-grams alone allow classification levels of about 66% for a 5-class task even when content and function word differences are accounted for. We hypothesize that the phonology of a native language has a strong effect on the word choice of people writing in a second language.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

[Edit]