Contact Us

01

Back

Named Entity Recognition

Identify people, locations, and organizations within a body of text. Our NER engine has surpassed the best-performing models in the industry by a wide margin. The engine is optimized on social for short informal content including emojis and spelling mistakes. The user has the option to turn on disambiguation for named entity variations.

The entities that these engines create become the backbone of your knowledge graph and analytic workflows. This engine can be tuned to specific verticals (i.e. finance or geopolitical). This engine is also available on Social Data. Available Languages: English, Arabic, Chinese (Simplified).

  • Example 1

    Admiral Stavridis proposed an expansion to Eagle Guardian the NATO’s contingency plan for the reinforcement and defense of Poland.

    Entities[PER]

    people:

    Stavridis

    confidence:

    0.87

    [MISC]

    miscellaneous:

    Eagle Guardian

    confidence:

    0.79

    [ORG]

    organizations:

    NATO

    confidence:

    0.96

    [LOC]

    locations:

    Poland

    confidence:

    1.00

  • Example 2

    Michelle Leung’s Xingtai Capital avoids companies that have yet to turn a profit and doesn’t hold stocks in Chinese internet giants, including Alibaba and Tencent.

    Entities

    people:

    Michelle Leung

    confidence:

    1.00

    needs disambiguation:

    False

    organizations:

    Xingtai Capital

    confidence:

    1.00

    needs disambiguation:

    False

    organizations:

    Alibaba

    confidence:

    0.99

    needs disambiguation:

    False

    organizations:

    Tencent

    confidence:

    1.00

    needs disambiguation:

    False

Available Engines

  • NER, Social Data (Twitter)
  • NER, Arabic
  • NER, Chinese (Simplified)
Request a Demo