NlpSaftReferent

GoogleApi.ContentWarehouse.V1.Model.NlpSaftReferent

Table of Contents ▼

Jump to a specific part of the page:

A referent contains information about what a discourse context entity is referring to. It acts like a canonical mention of the entity.

This module has the following attributes (case-insensitive ascending order):

View Attributes

distance (type: float(), default: nil)
- Distance of this referent from the markables in the Document. Smaller values imply that the referent is more accessible to be an antecedent for a markable in the Document. The expectation is that this field would increase with every new Document in which this referent is not mentioned.
explicitness (type: String.t, default: nil)
-
info (type: GoogleApi.ContentWarehouse.V1.Model.Proto2BridgeMessageSet, default: nil)
- Application-specific information about this referent.
phrase (type: GoogleApi.ContentWarehouse.V1.Model.NlpSaftPhrase, default: nil)
- Entity name phrase. The phrase indices are relative to the token array above. The phrase should normally cover all the tokens in the name and the head must be set to be the head token of the name.
prominence (type: float(), default: nil)
- Prominence score for referent. This is roughly equivalent to the number of previous mentions of the referent.
role (type: String.t, default: nil)
-
token (type: list(GoogleApi.ContentWarehouse.V1.Model.NlpSaftToken), default: nil)
- Tokenized representation for the canonical name of the referent entity.

@type t() :: %GoogleApi.ContentWarehouse.V1.Model.NlpSaftReferent{
distance: float() | nil,
explicitness: String.t() | nil,
info: GoogleApi.ContentWarehouse.V1.Model.Proto2BridgeMessageSet.t() | nil,
phrase: GoogleApi.ContentWarehouse.V1.Model.NlpSaftPhrase.t() | nil,
prominence: float() | nil,
role: String.t() | nil,
token: [GoogleApi.ContentWarehouse.V1.Model.NlpSaftToken.t()] | nil
}

@spec decode(struct(), keyword()) :: struct()