Closed list entity

Closed list entity

What are closed list entities

Closed list entities allow you to provide synonyms different ways of spelling for one or multiple entities so that there is no need to add all utterances that use these entities separately. For example, New York is often referred to as NYC, The Big Apple and Gotham, whereas Los Angeles is often referred to as L.A. or the City of Angels. By creating a Closed list entity named 'City' we can add New York and Los Angeles as canonical forms together with their synonyms in the utterances field. The Natural Language Processing (NLP) engine will only grasp the predefined values of the closed list entity.

NLP Support

Closed list entities are the default type of entities of the Blits platform. Currently, closed list entities are supported by all NLP engines.

Example

Intent: Book Flight

Original utterance: "I'd like to fly from New York to Los Angeles"

Utterance with entity: "I'd like to fly from City to City"

User input: "I'd like to fly from L.A. to NYC"

Recognised Entities:

Entity 1: New York 

Entity 2: Los Angeles 


    • Related Articles

    • Add entity

      How to add an entity To add an entity, you mist click the 'Add Entity +' button. Thereafter, you can toggle between Simple entities and Closed list entities. Add a name that resembles the group of words that the entity will cover. Next, the words can ...

    • Simple entity

      What are simple entities Simple entities can be used to group various words together to remove the need to rewrite utterances that simply use another word. This is often the case for various nouns or names, such as with city names. The NLP trains ...

    • About entities

      On this page entities can be created, exported and managed. What is an Entity? An entity is a collection of similar objects. By training entities your bot can detect specific objects a user asks in the conversation. This enables bots to understand ...

    • Regex entity

      What are regex entities Regex entities allow you to provide general letter, number, or combined sequences an entity. For example, the regex ([1-9])+/g matches all number sequences, Meaning that 1234 is recognized as an entity, but also 54214234. ​ ...

    • Is an entity

      The is an entity condition diverts the conversation if the input is equal to either the entity name, a canonical value or a utterance. Please select the desired values in the drop-down menus.