NTT Research and Development 2010 Review of Activities

Overview

The technology that automatically analyzes the syntax and semantics of a natural language is a fundamental technique for a variety of text-based services. Recently, massive text data is available as a result of the explosive growth of the Web and electric documents. This technology can effectively utilize these large-scale data and greatly enhance the standard performance level of text-based services, such as sentiment analysis, information retrieval, e-mail spam filter, machine translation, and document summarization, by gaining the performance of natural language analyzers.

Features

• Achieved highest performance on standard benchmark data for natural language processing
• High-speed statistical machine-learning technology for large-scale data
• Possible to reduce the cost of manual data preparation
• Possible to be applied to other fields, such as image processing, bioinformatics, etc.
• Supports many languages

Application scenarios

• Semantic analysis of natural language documents
• Make existing systems such as evaluation analysis, machine translation, document summarization, and information search highly accurate
• Analyze text data in blogs and on the Web
• Profitable data mining of large-scale text data
• All other applications that use the Natural Language Processing Machine