Publishing Partner: Cambridge University Press CUP Extra Publisher Login
amazon logo
More Info

New from Oxford University Press!


It's Been Said Before

By Orin Hargraves

It's Been Said Before "examines why certain phrases become clichés and why they should be avoided -- or why they still have life left in them."

New from Cambridge University Press!


Sounds Fascinating

By J. C. Wells

How do you pronounce biopic, synod, and Breughel? - and why? Do our cake and archaic sound the same? Where does the stress go in stalagmite? What's odd about the word epergne? As a finale, the author writes a letter to his 16-year-old self.

Academic Paper

Title: Detecting Errors in English Article Usage by Non-native Speakers
Author: Na-Rae Han
Institution: University of Pennsylvania
Author: Martin Chodorow
Institution: Hunter College
Claudia Leacock
Institution: ETS Technologies
Linguistic Field: Computational Linguistics; Language Acquisition; Syntax
Subject Language: Chinese, Mandarin
Abstract: One of the most difficult challenges faced by non-native speakers of English is mastering the system of English articles. We trained a maximum entropy classifier to select among a/an, the, or zero article for noun phrases (NPs), based on a set of features extracted from the local context of each. When the classifier was trained on 6 million NPs, its performance on published text was about 83% correct. We then used the classifier to detect article errors in the Test of English as a Foreign Language (TOEFL) essays of native speakers of Chinese, Japanese, and Russian. These writers made such errors in about one out of every eight NPs, or almost once in every three sentences. The classifier's agreement with human annotators was 85% (kappa = 0.48) when it selected among a/an, the, or zero article. Agreement was 89% (kappa = 0.56) when it made a binary (yes/no) decision about whether the NP should have an article. Even with these levels of overall agreement, precision and recall in error detection were only 0.52 and 0.80, respectively. However, when the classifier was allowed to skip cases where its confidence was low, precision rose to 0.90, with 0.40 recall. Additional improvements in performance may require features that reflect general knowledge to handle phenomena such as indirect prior reference. In August 2005, the classifier was deployed as a component of Educational Testing Service's Criterion Online Writing Evaluation Service.


This article appears IN Natural Language Engineering Vol. 12, Issue 2.

Return to TOC.

Add a new paper
Return to Academic Papers main page
Return to Directory of Linguists main page