Development of a Japanese/English idiom database using large-scale corpus data

本研究的目的是利用大规模语料库数据建立日英习语数据库。为了调查习语列表的使用频率，我们收集了语料库数据。然后，在互联网上建立了一个成语数据库(大约有1500个成语)。对于每个成语，包括以下项目：阅读、罗马化拼写、意思、成语类型、频率信息和例句。由于从语料库数据中提取的例句数量如此之大，编辑过程仍在继续。

The aim of this research was to develop a Japanese/English idiom database using large-scale corpus data. In order to investigate the frequency of a list of idioms, corpus data was gathered. Then, an idiom database(with about 1, 500 idioms) was constructed on the Internet. For each idiom, the following items were included : reading, Romanized spelling, meaning, idiom type, frequency information, and example sentences. Because the amount of example sentences extracted from the corpus data was so huge, the editing process still continues.