Natural Language Processing and Consultancy

Introduction

For developers working with natural language, there are a number of useful components available from Phrasys for inclusion into applications. All of the components are implemented in 100% Java and are thus fully platform independent. They do not make use of any platform-specific features, so they really do fullfil the Java promise of write once, run everywhere.

The components are implemented using a very easy to use API, where all complexity is hidden from the application programmer. Where possible method calls have been kept analogous, so that the inital learning curve is very shallow. Including the components into your own application is very straightforward, allowing you to process natural language texts with ease.

The software components available at Phrasys solve the basic problems you will face when developing applications that deal with natural language input:

• splitting a text into sentences
• splitting a sentence into tokens
• assigning word class tags to the tokens
• identifying phrasal groups in a sentence
• identifying names in a text
• reducing words to a stem
• reducing words to a lemma
• resolving anaphoric references (eg pronouns)

Some of the modules are not yet ready for release; we take quality control very seriously, and would not want to release any components that might not work properly.

For more information about the development process, see the page on quality control.

Availability

At present components are only available on request, as they are being overhauled, which involves some API adjustments. The pricing structure is also not yet determined. If you want to know more about the current status, please contact us at info@phrasys.net.

 

[Phrasys Home] [Contact] [Software] [Consultancy]