Ambiverse Natural Language Understanding API extracts entities from unstructured text, enabling a more precise transformation of texts into actionable, measurable, and easily accessible knowledge. Entities are identified by types such as person, location, organization, or product, and linked to the Wikipedia-derived YAGO knowledge graph. You can query the knowledge graph to obtain further information about these entities, such as Wikipedia links, textual descriptions, images, and lists of relevant categories.

 

Entity Linking Demo

Identify all people, organizations, locations, events, and products in the text you enter.

 
Enter any text in English, German, Spanish, or Chinese.

Settings

0.075

Our technology builds on years of experience in scientific developments by the Max Planck Institute for Informatics, a world leading institution in automatic text understanding research. The API allows your application to understand that in the sentence "When The Who played Tommy in Columbus, Pete was at his best.", "Pete" refers to Pete Townshend, the English rock guitarist of The Who fame, not to Pete Seeger, or any other entity usually referred to as "Pete". This technology enables you to develop a whole range of next-generation applications to better search, write, and analyze texts.

 

Read the Blog Post


Use our API

Ambiverse Natural Language Understanding API combines two essential tasks. Given a text, our Entity Linking Service recognizes names and matches them onto canonical entities (e.g., individuals, organizations, or places) registered in the Wikipedia-derived YAGO knowledge graph. You can then retrieve additional information about these entities using the Knowledge Graph Service. Our Free Plan allows you to process up to 1,000 documents per month for free!
 
 
curl https://api.ambiverse.com/v1/entitylinking/analyze -d '{ "text": "When Who played Tommy in Columbus, Pete was at his best." }'
{
  "matches": [
    {
      "charLength": 3,
      "charOffset": 5,
      "entity": {
        "id": "http://www.wikidata.org/entity/Q93346",
        "url": "http://en.wikipedia.org/wiki/The%20Who",
        "confidence": 0.5197556706744466
      },
      "text": "Who"
    },
    {
      "charLength": 5,
      "charOffset": 16,
      "entity": {
        "id": "http://www.wikidata.org/entity/Q372055",
        "url": "http://en.wikipedia.org/wiki/Tommy%20%28album%29",
        "confidence": 0.08991167067721392
      },
      "text": "Tommy"
    },
    (...)
  ]
}
curl -H "Content-Type: application/json" https://api.ambiverse.com/v1/knowledgegraph/entities -d '[ "http://www.wikidata.org/entity/Q372055" ]'
{
  "entities": [
    {
      "id": "http://www.wikidata.org/entity/Q372055",
      "name": "Tommy (album)",
      "links": [
        {
          "source": "Wikipedia",
          "url": "http://en.wikipedia.org/wiki/Tommy%20%28album%29"
        }
      ],
      "description": "Tommy is the fourth album by English rock band The Who, released by Track Records and Polydor Records in the UK and Decca Records/MCA in the US. A double album telling a loose story about a \"deaf, dumb and blind boy\", Tommy was the first musical work to be billed overtly as a rock opera. Released in 1969, the album was mostly composed by Pete Townshend. In 1998, it was inducted into the Grammy Hall of Fame for \"historical, artistic and significant value\".",
      "imageUrl": "http://upload.wikimedia.org/wikipedia/en/1/19/Tommyalbumcover.jpg",
      "categories": [
        "YAGO3:<wordnet_physical_entity_100001930>",
        "YAGO3:<wordnet_product_104007894>",
        "YAGO3:<wordnet_psychological_feature_100023100>",
        "YAGO3:<wikicat_The_Who_albums>"
      ]
    }
  ]
}
Entity Linking Service - Recognize and link entities to our knowledge graph.
Slider
 
 


Features

  • Natural language understanding as a service
  • Very high entity linking accuracy
  • Built on world-class research by the Max Planck Institute for Informatics
  • English, Spanish, Chinese, and German. More are coming soon.

 

  • Large and precise knowledge graph: 4+ million entities, 600+ thousand categories
  • Fully extensible knowledge graph
  • Easy to integrate thanks to a powerful and clean REST API
  • Scalable infrastructure powered by Microsoft Azure