Named entity recognition

Detect the people, organizations, locations, dates and other named entities mentioned in the provided document.

The API call can be used to extract the named entities mentioned in the provided document. The identified entities include people, organizations, locations, dates, currencies and similar. For each entity we return the mentioned text, the character offsets of the mention in the document and the entity type.

Compared to the annotate endpoint, the named entity recognition does not disambiguate the entities to their corresponding Wikipedia pages — it only identifies which parts of the text represent named entities and of which type. On the other hand, it can also identify entities that don't have their own Wikipedia page, such as dates, monetary amounts or less known people and organizations.

The named entity recognition is supported for documents in English, German, Spanish and Chinese language.