Our products
We offer own products for Japanese natural language processing.
Kuromoji is our Japanese morphological analyzer, and it is open source.
Kuromoji analyses Japanese text and segments it into words (morphemes) and also accurately identifies parts-of-speech, readings for kanji, etc.
Kuromoji also provides the core Japanese language support for Apache Lucene and Apache Solr, but it is also available standalone for NLP applications.
Akahai is our query-suggester and type-ahead engine designed for Japanese.
Akahai received text it is written by Japanese people using IME (Input Method Editor) and quickly matches combinations of hiragana, katakana, kanji and romaji to provide highly relevant and ranked suggestions.
Akahai integrates with any search application and serves millions of users daily.
Keywords is our keyword extractor that identifies the "information carrying" words in Japanese text. Hover to highlight words in the below newspaper article excerpt:
紫煙 くゆらす オヤジ が 占拠 する 碁会所 のイメージが変わりつつある。主役は「 囲碁ガール 」。「 脳 からキレイに」「 女子力アップ 」。 うたい文句 にいざなわれ、女性が続々と碁の世界へ。彼女たちから未来の名人が輩出するかも？
紫煙 くゆらす オヤジ が 占拠 する 碁会所 のイメージが変わりつつある。主役は「 囲碁ガール 」。「 脳 からキレイに」「 女子力アップ 」。 うたい文句 にいざなわれ、女性が続々と碁の世界へ。彼女たちから未来の名人が輩出するかも？
When applied to a collection of documents we can easily understand the gist of all these documents at a high level.
Keywords is used in search applications to visualize search results, improve information navigation, do high-quality similarity searching or query-by-example, and for content based personalization.