Extract Entities From Documents

Select an input document from one of the examples below, or enter your own input, and extract all important entities. Use the menu on the right if you wish to specify which entity types to extract. Learn more!

Extract Entities From Documents

Identifies and extracts all types of useful entities contained in text. Entities can be of the following types: Keywords, Taxonomy Terms, People, Organizations, Locations, Addresses, Phone Numbers, Email Addresses, URLs, Times, Dates, Ages, Money Amounts, and Credit Cards. Address extraction is currently implemented for the following regions: New Zealand, Australia, United Kingdom, United States, Germany. Phone number extraction is currently implemented for New Zealand formats only.

Entity extraction allows the conversion of unstructured text into structured data. From then on the potential use cases are endless: automatically assign metadata to describe a document's content, thereby improving search; classify a collection of texts geographically on a map; generate analytical insight by computing occurrences of entities within large documents.

Entity extraction eliminates the need for manual work – instantly assemble the key facts and players in any document. An automated process ensures accuracy and consistency, making the assignment of metadata both more reliable and more efficient. Large document collections can be organized and searched in a faster and more productive manner.

Extract Taxonomy Terms > Taxonomy Information

We use a number of publicly available taxonomies. We provide these taxonomies by default as part of our API. The table below lists the vertical, language and source of these taxonomies. Customers may also plug in their own taxonomy or taxonomies in either one of the standard formats: OBO or SKOS.