Leading Multi-Lingual Natural Language Processing Software Company now offers a complete line-up of extraction and geospatial stand-alone products and complete suite solutionsHERNDON, Va., May 7, 2014 — (PRNewswire) — Rosoka Software today announced the expansion of its line of Natural Language Processing (NLP) products to include new features and capabilities for customers and partners. Rosoka multilingual NLP tools are all 100% pure Java engines that perform named entity extraction from unstructured or text documents, entity salience determination, relationship/fact extraction, and sentiment analysis on any platform that has a Java Virtual Machine. This robust set of tools is scalable to handle Big Data, but compact enough to fit on a smartphone or tablet and is the only multilingual product that supports more than 230 languages in a single engine.
The key components of the Rosoka product line include an extraction engine, LxBase (language base), language ID engine, document viewer, and name resolver. In addition, the company offers a robust geotagger as an add-on option and a full geospatial product called GeoGravy which can be purchased separately or as part of a bundle.
Extraction Products and Bundles
Rosoka Extraction is the company's core API-driven information extraction engine that supports language independent processing for over 230 languages, automatic word segmentation tokenization, document metadata extraction and document filter capability. Rosoka Extraction Plus adds geotagging of PLACE entities, giving the user the ability to tag locations as a part of the extraction.
Rosoka NLP is a complete product suite designed for companies that require a total solution for multilingual extraction and analysis. The bundle includes use and modification of the LxBase which includes the rules, dictionaries, and configuration files necessary for using a Rosoka Extraction Engine and the Rosoka Document Viewer (RDV). The Rosoka Document Viewer is a development and analysis GUI for examining Rosoka extraction results (not sold separately).
Additional stand-alone products are available for use with other NLP systems. Rosoka Language ID enables an organization to quickly provide language identification of over 230 languages on a string, document portion, or the entire document and is available as a stand-alone API-driven engine for use with other NLP systems. Rosoka Tokenizer is a java library for multilingual text parsing, tokenizing Unicode into individual words regardless of the language, and code blocks.
Rosoka Name Resolver provides two sets of capabilities: Name Compare and Name Expander. The Name Compare capabilities answers the questions, "Are these two names the same name?" and "How closely do the names match?" The Name Expander capability answers the questions, "What are the possible names variations of this name?" This API returns an ordered list of possible variations ranked from most likely to most common name "misspellings".
Rosoka GeoGravy is a stand-alone geospatial tagging service and gazetteer which will provide either a most likely geotag based on the list of names provided based on the context of the list or the complete list of geotags associated with each place name in the provided list. Rosoka GeoGravy can be used with other NLP software.
For companies who want the flexibility and convenience of single instance extraction on demand can currently purchase Rosoka extraction by the hour via the Amazon AWS hosted web service. The company has also announced plans to significantly expand its cloud services offerings by mid-year to provide a broader range of options for customers and partners.
To learn more about Rosoka Software's Natural Language Processing and Geospatial Analysis software solutions, visit us on line at www.rosoka.com.
About Rosoka Software
Rosoka Software solutions provides the power to unlock large volumes of information from any multilingual source, determines the relevance and relationship of the data, and delivers value specific results on any platform, application, or device from 230+ languages. Rosoka Software Inc. is a wholly owned subsidiary of IMT Holdings, Inc., Herndon, VA, USA. www.rosoka.com
Pivotal Communications Group
SOURCE Rosoka Software