「台灣日語學習者語料庫」（CTLJ）之試行公開

Development and Implementation concerning the Corpus of Taiwanese Learner of Japanese (CTLJ)

本文報告關於「台灣日語學習者語料庫」（CTLJ）的內容與使用方法。本語料庫收集了台灣高等教育機構之日語學習者的作文，並進行數位化以及標誌註記的處理，經過多年的努力，透過跨領域合作方式克服種種技術上的困難，現階段已完成之語料先試行公開，讓日語研究、教學或學習者利用網際網路進行研究或學習。本語料庫計畫建構1500篇作文，截至2009年2月已有1483篇，本次公開2003至2006學年度前期總計677篇作文之數位語料，其餘數位化完成後，陸續開放使用，本語料庫於收錄時已取得作者同意公開之承諾書。

This report is concerned with the content and the instruction of the Corpus of Taiwanese Learner of Japanese (CTLJ) This corpus has collected compositions of Japanese learners from several higher education institutions in Taiwan, and most of the collected data have been digitalized and annotated with corrections Through interdisciplinary cooperation, many technical problems have been overcome after years' efforts. To date, CTLJ has acquired 1,483 compositions, and its ultimate goal is 1,500. The current stage presents 677 digitalized data collected from 2003 to 2006. The remaining data will be soon open after digitalization. In addition, these data are provided under authors' agreement.Since Huang et al. (2007) had a preliminary report about the way of data-collection, the method of digitalization and the techniques of corpus construction. This report not only supplements the content which is not included in the preliminary report, hut also elucidates it s recent advances and development. Moreover, the website, the registration, the instruction and the functions of the corpus are well explained. The public opening of the CTLJ can realize the ideal of s ha ring research resources and facilitate the relative research and practical development for corpus linguistics.