KnowledgeAnswersIntentQueryToken

Description

A token represents an ngram with relevant information about it. If the token is a context phrase, it will have a prior score associated with it. The prior is computed via knowledge/answers/query_generalization/ word_prior/word_prior_from_examples_lib.cc, and ranges between 0 and 1. Stopwords and intent tokens (primary and component) have a score of 1.0.

Attributes List

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

Attributes

1. evalData (type: GoogleApi.ContentWarehouse.V1.Model.NlpSemanticParsingAnnotationEvalData, default: nil)
   - This field is used inside Aqua and outside Aqua for identifying the token indices and/or byte offsets of this Token.
2. ngram (type: String.t, default: nil)
   - |ngram| should be populated with a string from the raw query, not the normalized tokens. E.g. The ngram in the ignored token for the Height intent on the query [Height of barack obama], will be "Height". The ngram in the ignored token for the Videos intent on the query [vidéos] will be "vidéos".
3. parsedDueToExperiment (type: list(String.t), default: nil)
   - Experiments that caused this Token to parse, without which this would not have parsed.
4. prior (type: number(), default: nil)
   -
5. provenance (type: String.t, default: nil)
   -
6. provenanceId (type: list(String.t), default: nil)
   - Unique identifiers for the provenance of this token, for example, NLP Repository Example IDs.
7. provenanceLanguage (type: String.t, default: nil)
   -
8. synonyms (type: list(GoogleApi.ContentWarehouse.V1.Model.KnowledgeAnswersIntentQueryTokenSynonym), default: nil)
   -

Type

Function

Data sourced from HexDocs : GoogleApi.ContentWarehouse.V1.Model.KnowledgeAnswersIntentQueryToken

• ‹ KnowledgeAnswersIntentQueryTeleportArgumentAnnotatorSignals
• KnowledgeAnswersIntentQueryTokenSynonym ›

Possibly related Links