Japanese Data Directory

You must be logged in with a valid customer support login (provided by Lexalytics) in order to access product downloads.

The Japanese data directory extends the text analysis capabilities of Salience to extract meaning and information from Japanese content.

The Japanese data directory contains components trained in the part-of-speech tagging and core natural language processing (NLP) functions available in Salience. A named entity extraction model has been trained from annotations of Japanese content, and can be extended through user customizations such as pattern files and CDL files as in other languages supported by Salience. A Concept Matrix™ has been developed for Japanese to provide concept topic functionality in addition to query-based topic functionality. The data directory contains a base sentiment phrase dictionary, as well as support for intensifiers and negators. Theme patterns have also been developed through analysis of Japanese content.

For more information about the other languages that Salience supports, please see the Language support section on our Developer Wiki.