Japanese Orthographic Database

CJKI’s Japanese Orthographic Database (JOD) is a comprehensive database of Japanese orthographic variants. The orthographical complexity of Japanese poses a special challenge to the developers of computational linguistic tools, especially in the area of intelligent information retrieval. These difficulties are exacerbated by the lack of standardized orthography and the highly irregular Japanese orthography.

JOD plays a critical role in enhancing the accuracy of information retrieval, machine translation, and morphological analysis applications as it helps identify and disambiguate the numerous Japanese orthographic variants that have identical or related meanings, such neko ‘cat’ written 猫, ねこ or ネコ and kakiarawasu ‘write out, publish’ written 書き著す, 書 著す, 書き著わす or 書著わす.

This comprehensive database has over 126,000 entries and is being used by such companies as Google, Amazon, and Baidu to dramatically improve search recall. Also included are a large variety of katakana orthographic variants for loanwords. For a detailed description, see Jack Halpern’s white paper The Challenges of Intelligent Japanese Searching.