MarkLogic 9 Product Documentation

entity.extract

entity.extract(
   node as Node,
   [dictionaries as cts.entityDictionary[]],
   [options as String[]],
   [map as Object]
) as Sequence

Summary

Extract entities from a document using the provided entity dictionary. The matching entities are returned.

Usage Notes

Mapping can be done piecemeal. The type of an entity is split into segments, delimited by a colon, and then each segment is matched in turn against map until a match is found. If the value of a segment matches a mapping entry whose value is itself a map, then that submap is used to match subsequent segments of the type. If the submap contains no match, then the empty string is used to find a match within the submap.

The default map defines a set of default recommended entity names:

e:credit-card-number

A number representing a credit card number. (IDENTIFIER:CREDIT_CARD_NUM)

e:coordinate

Latitude and longitude coordinates. (IDENTIFIER:LATITUDE_LONGITUDE)

e:date

Date-related. (TEMPORAL:DATE)

e:distance

A distance. (IDENTIFIER:DISTANCE)

e:email

An email address. (IDENTIFIER:EMAIL)

e:gpe

Geo-political entity. Differs from location in that it has a person-made aspect to it. For example, California is a GPE because its boundaries were defined by a government). (GPE)

e:facility

A place used as a facility. (FACILITY)

e:id

A number identifying a social security number or other ID number. (IDENTIFIER)

e:location

A geographic location. For example, Mount Everest. (LOCATION)

e:money

Currency, such as dollars, euros, and so on. (IDENTIFIER:MONEY)

e:nationality

The nationality of someone or something. For example, American. (NATIONALITY)

e:number

A number. (IDENTIFIER:NUMBER)

e:organization

An organization. (ORGANIZATION)

e:person

A person. (PERSON)

e:phone-number

A number identifying a telephone number. (IDENTIFIER:PHONE_NUMBER)

e:religion

A religion. (RELIGION)

e:title

A title or honorific. (TITLE)

e:url

A URL on the world wide web. (IDENTIFIER:URL)

e:utm

A point in the Universal Transverse Mercator (UTM) coordinate system. (IDENTIFIER:UTM)

e:time

Time-related. (TEMPORAL:TIME)

See Also

• Dictionary-Based Entity Extraction in the Search Developer's Guide
• Using an Entity Type Map for Extraction or Enrichment in the Search Developer's Guide
• cts.entityWalk

Example

// Extract entities using the people onotology from the 
// entity.dictionaryLoad example.

'use strict';
const entity = require('/MarkLogic/entity');

const mapping = {
  '' : fn.QName('http://marklogic.com/entity', 'entity:entity'),
  'administrative district': fn.QName('http://marklogic.com/entity', 'entity:gpe'),
  facility: fn.QName('http://marklogic.com/entity', 'entity:facility'),
  person: fn.QName('http://marklogic.com/entity', 'entity:person'),
  land: fn.QName('http://marklogic.com/entity', 'entity:location'),
  location: fn.QName('http://marklogic.com/entity', 'entity:location'),
  organization: fn.QName('http://marklogic.com/entity', 'entity:organization'),
  region: fn.QName('http://marklogic.com/entity', 'entity:gpe')
};
const node = new NodeBuilder()
                   .addElement('p', 'Nixon went to Paris')
                   .toNode();
entity.extract(node, cts.entityDictionaryGet('/ontology/people'), null, mapping);

// Returns the extracted entity. For example:
//
// <entity:person xmlns:entity="http://marklogic.com/entity">Nixon</entity:person>
// <entity:gpe xmlns:entity="http://marklogic.com/entity">Paris</entity:gpe>
// <entity:gpe xmlns:entity="http://marklogic.com/entity">Paris</entity:gpe>
// <entity:entity type="imaginary being:mythical being" xmlns:entity="http://marklogic.com/entity">Paris</entity:entity>