Entity extraction

Today, there is little meaningful interaction with text in apps. Most apps offer very little little beyond the basic cut/copy/paste operations when working with text. Entity extraction is intended to improve the user experience inside your app by understanding text and performing specific actions on it.

The entity extraction API allows you to detect and locate entities from raw text as it is being typed. It identifies different types of entities & allows you to make this text richer, letting the user take action based on their type. Examples include:

  • URLs (google.com)
  • e-mail addresses (info@google.com)
  • phone numbers (1-800-555-5555)
  • addresses (1950 Amphitheatre Parkway)
  • payment card numbers (1234 5678 0000 1111)
  • money (including currency)
  • tracking numbers for packages
  • date-time (Let's meet tomorrow at 6pm)

Entity extraction supports the following languages:

  • Arabic
  • Portuguese
  • English
  • Dutch
  • French
  • German
  • Italian
  • Japanese
  • Korean
  • Polish
  • Russian
  • Chinese
  • Spanish
  • Thai
  • Turkish

Examples

Input textDetected entities
Meet me at 1600 Amphitheatre Parkway, Mountain View, CA, 94043 Let’s organize a meeting to discuss. Entity 1 type: Address
Entity 1 text: "1600 Ampitheatre Parkway, Mountain View, CA 94043"
You can contact the test team tomorrow at info@google.com to determine the best timeline. Entity 1 type: Date-Time
Entity 1 text: = "June 24th, 2020"

Your order has shipped from Google. To follow the progress of your delivery please use this tracking number: 9999-9999-9999-9999 Entity type: Tracking number
Entity text: "9999-9999-9999-9999"
Call the restaurant at 555-555-1234 to pay for dinner. My card number is 1111-1111-1111-1111. Entity 1 type: Phone number
Entity 1 text: "555-555-1234"

Entity 2 type: Payment card
Entity 2 text: "1111-1111-1111-1111"

Sign up

Fill out this form to request early access to the entity extraction API.