Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.

Natural Language Processing Technologies

We are developing natural language processing technologies for analyzing the semantics of various text data from newspapers, magazines, blogs, in-house documents, e-mail and so on, and effectively discovering valuable information from multiple information sources.
We are currently engaged in the following four technologies to develop various solutions (e.g. marketing, enterprise risk management, and customer relationship management (CRM) that use a customer's voice or the reputation of products and services extracted from a volume of text data)

  1. Language analysis technology that covers various expressions used in spoken language, blogs, etc.
  2. Synonymy and entailment recognition technology for determining whether words or sentences have the same semantics.
  3. Semantic and relation detection technology for detecting the semantics and relations of sentences from multiple documents.
  4. Search and surveillance technologies for using multiple information sources by using semantics and relations.
A method of (2) to recognizing entailment between two sentences that received 1st prize at NIST evaluation workshop 2011. We plan to develop technologies that provide new values extracted from massive data.