初・中級学習者用8言語例文パラレルコーパスの開発

This paper delineates a project concerning the development of a parallel corpus for language learners, named PARC8. The project is currently being developed by a group of foreign languages teachers at Hiroshima University. In order to contribute to multilingual education, the endeavor aims to build a parallel corpus of example sentences in eight languages — Japanese, English, German, French, Spanish, Russian, Chinese, and Korean — that can be used by learners at the beginner and intermediate levels, corresponding to A1 and B1 in the CEFR (Common European Framework of Reference for Languages). This corpus will be published on the Internet as a learning-support corpus, and will be widely available to foreign language learners and language educators in Japan and abroad. The corpus will consist of about 14,000 sets of example sentences of the same content in each language. The output language of the example sentences can be changed freely in up to eight languages, depending on the combination of languages required by the user. In this paper, I mainly discuss the challenges involved in creating high-quality example sentences in the eight languages that represent the same content in the development project.