Information Extraction

Unlocking the Value from Unstructured Text

An enormous amount of information exists only in natural language form.

In order to automatically process and analyze this information, it must be structured and placed in databases. Information Extraction (IE) is the process of identifying and retrieving relevant knowledge from unstructured textual documents and organizing it in tables or lists.

For example, if we want to know which airlines have the best fares or who signed contracts over the past six months to deliver natural gas, or which jurisdictions have enacted new restrictions on smoking, IE techniques produce lists of organizations instead of retrieving reams of documents that must be manually searched for the information of interest.

IE systems operate on natural language texts, providing databases and enabling input for a wide range of retrospective analyses on reports, presentations, emails or manuals generated primarily in natural language form. Our Information Extraction systems can be easily integrated with various knowledge management and portal resources to work within different business domains.

Our Information Extraction technology was developed by our team of renowned researchers. For the past years, we won, by wide margins, several prestigious competitions and we have been awarded several research contracts for further technology development.

Types of Information Extraction

Language Computer's custom information extraction system provides a wide range of customizable information extraction components, including:

  • Entity Extraction: The Entity Recognition component identifies names of things with respect to a potentially large number of semantic categories, such as person, organization, or geopolitical location.
  • Attribute Extraction: The Attribute Extraction component identifies features of a given entity, such as the gender of a person.
  • Relationship Extraction: The Relationship Extraction component identifies relationships between entities and events. For instance, we may wish to identify a common relationship such as is located in between an entity and a location.
  • Event Extraction: The Event Extraction component identifies information about events, such as the time an event occurred, where it was located, and who was involved.

Related Products


Open-domain, customizable entity, attribute, relationship, and event extraction for English, Chinese, and Arabic.


High-performance named entity recognition for English, Arabic, Chinese, Farsi, and Korean texts.


Accurate pronominal and nominal coreference resolution for English.


Temporal and spatial awareness for absolute or relative mentions of times, dates, or locations.


For More Information

For more information on how Language Computer can help your organization meet its custom information extraction needs, contact us at (972) 231-0052 or e-mail us today.