Putting a face to the name

How do you know who an entity is in real life? Which “Clinton” is an article about, the 42nd President or the 2016 presidential candidate? Rosette connects entities in your unstructured text to the real-world across 11 languages, extracting information that resolves who or what the entities refer to.

Multi-faceted resolution

Rosette uses the linked information to resolve entity ambiguity in 3 ways:

  • For entities with more than one name, groups them together as “synonyms”.
  • For names that reference more than one entity, determines which one is the right one.
  • Connects information about an entity within and across documents.

Linking with your own source

Rosette uses Wikipedia as its default link source but you can replace it with one of your own, for example, a custom database, watchlist, or employee list.

Metadata enrichment

Rosette adds metadata from the linking process to your text, including confidence measures for each of the linking decisions, that can be used to power entity-centric search and notification, track new entities in text streams, and build custom knowledge graphs.

Pre-trained to link to a Wikipedia-derived 2M+ entity database

Tamerlan Tsarnaev

Tamerlin Tsarnaev (TheAtlantic.com)
Tamerlane Tsarnaevy (Mir24.net)


Apple Corps Ltd. (Music)
Apple Inc. (Technology)


Paris, Texas (33°39 N, 95°32 W)
Paris, France (48°51 N, 2°21 E)

Name Matching for Government

What Universal Challenge Do All Governments Face Today?

Read More

Shining A Light On Consumer Feedback

Read More

Supported Languages & Features

Languages (11)

  • Arabic
  • Chinese
  • English
  • Spanish
  • Japanese
  • Korean
  • Pashto
  • Persian (Dari)
  • Persian (Farsi)
  • Russian
  • Urdu

Entity Types (3)

  • Person
  • Location
  • Organization
Entity Linking
Entity Linking

Live Demo:

Link entities in your unstructured text to your knowledge base in 11 languages.