Welcome to Duane Kindt's Computer Learner Corpus (CLC) webpage.
Here is an preliminary outline of a preliminary corpus project:
We are exploring the use of a corpus of students' spoken English and a native-speaker corpus as tools in discovering what areas of language study student could most benefit from. To do this we are following the procedure listed below:
- After practicing, students record a 5-minute conversation with a partner on the theme: "What do you think..."
- For homework, the conversation is transcribed and emailed to the teacher without any corrections, except spellchecking. Since our classes are 45 minutes, 3 times a week, one class period was used for transcribing.
- As students send their messages, the teacher prints each submission and saves it as a text file.
- After the assignment deadline, the teacher pastes all submissions into one file.
- The file is then checked for spelling and grammatical errors, the teacher making note of common structural errors or other problems.
- MS Word and corpus software are used to compare the students' language and a native-speaker corpus, the results leading to learning materials that focus on those errors.
- In the next class meeting, the learning material is given to students.
- The original transcription is also given and students look for the errors in question.
- Another recording on the same theme is made in the following class.
- Repeat numbers 2 through 8.
As the initial stages in building a computer corpus need piloting, what comes of this first project will lead to the shape and purpose of a larger learner corpus.
For more information about CLC and corpus linguistics, follow the links below:
- Concordances and Corpora Tutorial (Catherine N. Ball )
- Learner Corpus and SLA Research (Yukio Tono)
- Corpus Linguistics: table of contents (Tony McEnery and Andrew Wilson)
- CORPUS RESOURCES
- British National Corpus
- Mike Scott's Web
- Tim Johns Data-driven Learning Page
- Corpus Linguistics (Michael Barlow)
- Improvising Corpora for ELT: Quick and Dirty ... (Christopher Tribble)
- Web Concordancer (Virtual Language Centre)
- W3-Corpora
- The Centre for English Corpus Linguistics (Sylviane Granger)
- Corpus Research at the University of Birmingham
- Cobuild Home Page
For more about the AVCR research group and their work with audio and video, follow these links:
- Recording conversations
- Videoing conversations
Contact:
© Duane Kindt, 2000
Updated: 1/1/03