CiceroLite

High-Performance Entity Extraction

Language Computer's CiceroLite recognizes hundreds of different types of named entities in English, Arabic, and Chinese texts with nearly 90% precision and recall. It is available as one of many plug-in NLP components which operate within the Cicero On-Demand server.

Introduction

LCC's CiceroLite family of entity extraction systems provide unsurpassed performance - in terms of precision, recall, and processing speed - for English, Arabic, and Chinese texts. Designed to provide state-of-the-art performance for large entity type hierarchies, CiceroLite's robust machine learning framework enables it to be extensible to new languages of interest quickly and easily, given sources of training data.

The most entity types for English: More than 150.

LCC's CiceroLite recognizes more than 150 different named entity types in English natural language texts with greater than 90% precision and recall. Need entity information targeted for your particular domain of interest? Add up to 250 additional domain-specific entity types with LCC's CiceroLite Entity Packs - or use custom entity extraction tools in LCC's CiceroCustom open-domain information extraction system to create your own entity extractors.

Wide-Coverage NER for Modern Standard Arabic. [Screenshot]

The first wide-coverage named entity recognition (NER) system for Modern Standard Arabic, CiceroLite recognizes a total of 65 entity types with nearly 90% precision and recall.

Industry-leading NER for Mandarin Chinese. [Screenshot]

Get state-of-the-art NER for Mandarin Chinese texts. The top-ranked system at the 2007 U.S. Government sponsored ACE evaluations for Chinese entity recognition, CiceroLite recognizes a total of 65 entity types with greater than 85% precision and recall.

Live Demo

Click here to launch the application:

Visit LCC Labs for more information.

Hardware / Software Requirements

LCC's CiceroLite was designed to run on standard hardware. Have questions about whether it'll run in your environment? E-mail us at support@languagecomputer.com.

  • Minimum Specifications: 1.0 GHz processor, 512 MB of available RAM
  • Recommended Specifications: 2.0 GHz (or better) processor, 2 GB of available RAM
  • Supported Operating Systems (Desktop Applications): Microsoft Windows (95, 98, XP, Vista, 7), Apple OS X, Linux (RedHat, CentOS, Ubuntu, etc.), Solaris
  • Supported Web Service Protocols: REST

Supported Languages

CiceroLite is currently available for the following languages.

English
Modern Standard Arabic
Mandarin Chinese
Japanese
Spanish
German
Dutch
French†
Farsi†
Korean†
Russian†
Romanian†

Languages marked with a † are coming soon. Interested in other languages? Contact us at at support@languagecomputer.com for an estimate.

For More Information

For more information about the CiceroLite, call our product support team at (972) 231-0052, or e-mail us at support@languagecomputer.com.