Paper: Data Driven Language Transfer Hypotheses

ACL ID E14-4033
Title Data Driven Language Transfer Hypotheses
Venue Annual Meeting of The European Chapter of The Association of Computational Linguistics
Session Main Conference
Year 2014

Language transfer, the preferential second language behavior caused by similarities to the speaker?s native language, requires considerable expertise to be detected by humans alone. Our goal in this work is to replace expert intervention by data-driven methods wherever possible. We define a computational methodology that produces a concise list of lexicalized syntactic pat- terns that are controlled for redundancy and ranked by relevancy to language trans- fer. We demonstrate the ability of our methodology to detect hundreds of such candidate patterns from currently available data sources, and validate the quality of the proposed patterns through classifica- tion experiments.