Output details
11 - Computer Science and Informatics
University of Oxford
Output title
A classifier-based approach to preposition and determiner error correction in L2 English
Type
E - Conference contribution
DOI
-
Name of conference/published proceedings
Coling 2008 - 22nd International Conference on Computational Linguistics, Proceedings of the Conference
Volume number
1
Issue number
-
First page of article
169
ISSN of proceedings
-
Year of publication
2008
Number of additional authors
1
Additional information
<22>
This paper is an overview of a system to detect errors in preposition and determiners by people learning English (e.g. `Car depends of petrol for fuel'). These are the most common type of errors. The system detected such errors and suggested corrections, based on a probabilistic model trained on a large amount of syntactically analysed texts. At the time these results bettered the state of the art obtained by larger and better funded groups like Microsoft and ETS, Princeton. Acceptance rate for this conference was 24%.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
15
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
