Compilation of Bilingual Corpus in Japanese and Asian Languages

点击翻译按钮获取中文摘要

The primary purpose of this project is to collect the following four datasets and to compile a database out of them :1)Essays in Japanese composed by learners of Japanese as a second language2)The translation of 1)into the learner's mother tongue(or the language in which s/he is best at writing)3)The revisions of 1)corrected by teachers of Japanese as a second language4)Information on the learners' and the revisers' linguistic background(e. g. their history of Japanese learning/teaching)Based on this corpus, we also aim to organize researches in various domains such as :1)Comparison between Japanese and the languages which appeared in the corpus2)Error analyses of Japanese essays3)Comparison of discourse structures deriving from the learners' mother tongue4)Methods of correcting essaysIn August 2000 we published the database named "Essays and Translation by Learners of Japanese Language and Corrections by NS/NNS Instructors ver. 1(test version)", which contains essays written by 800 learners from seven different Asian courtiers, and we distributed it to researchers inside and outside Japan. In December 2000 we organized at the NLRI an international symposium based on the Test Version. In this meeting were presented studies in such areas as 1)comparison between Japanese and other languages, 2)differences in language behaviors and discourse structures, 3)application to the instruction of essay writing, and so on.In March 2001, we published, the database ver. 2(complete version)", to which contributed 1100 learners from 10 different Asian countries and regions. Now we have plan to compile, on one hand, another essay database including European and American languages, and on the other hand, a speech corpus by learners of Japanese, besides carrying out further researches based on the existing data.