View Source GoogleApi.ContentWarehouse.V1.Model.QualitySalientTermsSignalTermData (google_api_content_warehouse v0.4.0)

SignalTermData is signal-specific term-level information. Stores mostly internal data as it is one of the primary data structures used in the populators.

Attributes

  • bigramDiscountTf (type: number(), default: nil) - The deduction of bigram counts from its unigram children. This field is only available in debug mode.
  • bigramness (type: number(), default: nil) - How much we trust this bigram. For bigrams only. Range: [0.0, 1.0] This field is only available in debug mode.
  • centrality (type: number(), default: nil) - Measures how topical this term is to a particular signal. A term like "lincoln" in the Abraham Lincoln's Wikipedia page should have a centrality close to 1.0 while non-central terms like "florida" should have a centrality close to 0.0. Range: [0.0, 1.0] This field is only available in debug mode.
  • correctedTf (type: number(), default: nil) - The final term frequency for a particular term. This field is only available in debug mode.
  • expectedTf (type: number(), default: nil) - The term frequency we were expecting for a term given its IDF. Range: [0, observed_volume] This field is only available in debug mode.
  • globalNpmi (type: number(), default: nil) - Global NPMI. For bigrams only. This is a measure of the quality of bigrams calculated using IDF. Range: [-1.0, 1.0] This field is only available in debug mode.
  • idf (type: number(), default: nil) - The IDF of the label of a particular term. For a canonical term, this is the mean IDF of its originals, weighted by their observed TF. This field is only available in debug mode.
  • isBigram (type: boolean(), default: nil) - Whether or not this term is a bigram. This field is only available in debug mode.
  • label (type: String.t, default: nil) - Raw string that identifies a particular term. This field is only available in debug mode.
  • localNpmi (type: number(), default: nil) - Local NPMI (normalized pointwise mutual information). For bigrams only. This is a measure of the quality of bigrams calculated using observed TF. Range: [-1.0, 1.0] This field is only available in debug mode.
  • observedTf (type: number(), default: nil) - The observed term frequency in a particular signal. This field is only available in debug mode.
  • originalTerm (type: list(GoogleApi.ContentWarehouse.V1.Model.QualitySalientTermsSignalTermData.t), default: nil) - The list of the original terms for a canonical. This is used in the pipeline and it is not present in the final output. This field is only available in debug mode.
  • rawTf (type: number(), default: nil) - The raw term frequency in a particular signal. This field is only available in debug mode.
  • salience (type: number(), default: nil) - The measure of how important this term is in this signal. Range: [0.0, 1.0] This field is only available in debug mode.
  • source (type: String.t, default: nil) - source is the type of the signal of this SignalTermData.

Summary

Functions

Unwrap a decoded JSON object into its complex fields.

Types

@type t() :: %GoogleApi.ContentWarehouse.V1.Model.QualitySalientTermsSignalTermData{
  bigramDiscountTf: number() | nil,
  bigramness: number() | nil,
  centrality: number() | nil,
  correctedTf: number() | nil,
  expectedTf: number() | nil,
  globalNpmi: number() | nil,
  idf: number() | nil,
  isBigram: boolean() | nil,
  label: String.t() | nil,
  localNpmi: number() | nil,
  observedTf: number() | nil,
  originalTerm: [t()] | nil,
  rawTf: number() | nil,
  salience: number() | nil,
  source: String.t() | nil
}

Functions

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

Unwrap a decoded JSON object into its complex fields.