Type System

Baleen uses the UIMA concept of a type system to categorize extracted entities. By default, the following entity types are supported. Other types are available to capture structural information, metadata, relationships and events.

Version 2.4 and later

• Buzzword
• Chemical
• CommsIdentifier
• DocumentReference
• Frequency
• Location
  • Coordinate
• MilitaryPlatform
• Money
• Nationality
• Organisation
• Person
• Quantity
• Temporal
• Url
• Vehicle
• Weapon

Version 2.3

• Buzzword
• CommsIdentifier
• DocumentReference
• Frequency
• Location
  • Coordinate
• MilitaryPlatform
• Money
• Nationality
• Organisation
• Person
• Quantity
• Temporal
• Url
• Vehicle
• Weapon

Full documentation of the type system is available in the Baleen Javadoc.

Version 2.2 and earlier

• Buzzword
• CommsIdentifier
• DocumentReference
• Frequency
• Location
  • Coordinate
• MilitaryPlatform
• Money
• Nationality
• Organisation
• Person
• Quantity
• Temporal
  • Date
  • DateTime
  • Time
  • TimeSpan
• Url
• Vehicle