Extract Entity

Recognizes and extracts named entities from a text

The Extract Entity Activity takes text as input then recognizes the entities of specified types from the input. The output of the activity is a JSON dictionary where the keys are the selected entity types and the values are a list of entities for ach type.

Input

• Input Text : String Argument Required The input text from which entities will be extracted.

• Entities : Required The type of entities to extract from the document. These types can be extracted :

  • Number

  • Ordinal

  • Percentage

  • Age

  • Currency

  • Dimension

  • Temperature

  • DateTime

  • PhoneNumber

  • Email

  • IpAddress

  • Boolean

• Culture : Required The culture to consider while recognizing the entities from the text. This is important as same data is written differently in different cultures. Available values for culture are :

  • ar-* (Arabic)

  • en-us (English US)

  • en-* (English Others)

  • zh-cn (Chinese)

  • es-es (Spanish)

  • es-mx (Spanish Mexico)

  • pt-br (Portuguese Brazil)

  • fr-fr (French)

  • de-de (German)

  • it-it (Italian)

  • ja-jp (Japanese)

  • nl-nl (Dutch)

  • ko-kr (Korean)

  • sv-se (Swedish)

  • bg-bg (Bulgarian)

  • tr-tr (Turkish)

  • hi-in (Hindi)

Output

• Recognized Entities : String Variable Required The JSON with the recognized entities from the document.

Use Evaluate JSON Path activity to get the required data from the output JSON.