A Method for Extracting Formulaic Sequences from a Student Corpus
Search this article
Abstract
type:Article
Use of appropriate formulaic sequences can add fluency, accuracyand appropriacy to written English, and one important place thesesequences occur are as sentence starters such as “Needless to say”or “At the same time”. This article describes and provides opensource code for a tool created using Python and the Natural LanguageToolkit which can help identify formulaic sentence-startersin an untagged corpus of student writing, for use in progress measurementand course design. Example results from two corpora arepresented and discussed.
Journal
-
- 神奈川大学言語研究
-
神奈川大学言語研究 34 35-52, 2012-03-10
神奈川大学言語研究センター
- Tweet
Details
-
- CRID
- 1050282677547305088
-
- NII Book ID
- AN1008864X
-
- ISSN
- 09153136
-
- Web Site
- http://hdl.handle.net/10487/10252
-
- Text Lang
- en
-
- Article Type
- departmental bulletin paper
-
- Data Source
-
- IRDB