RepositoryWebrefMention

AI Overview😉

The potential purpose of this module is to identify and analyze entity mentions within a document or query. It appears to be a part of Google's natural language processing (NLP) capabilities, aimed at understanding the context and meaning of user searches. The module can identify multiple entities within a document or query, and each entity can be mentioned multiple times in different positions.
This module could impact search results by providing more accurate and relevant results to users. By identifying and analyzing entity mentions, the search engine can better understand the user's intent and provide results that are more closely related to the entities mentioned. This could lead to more personalized and accurate search results, improving the overall user experience.
To be more favorable for this function, a website could focus on providing high-quality, relevant, and informative content that clearly mentions entities and their relationships. This could include using schema markup to provide additional context about entities, as well as using natural language processing techniques to improve the readability and understanding of content. Additionally, websites could focus on providing accurate and consistent metadata, such as titles, descriptions, and keywords, to help the search engine better understand the content and its relevance to user searches.

Interesting Module? Vote 👇

Voting helps other researchers find interesting modules.

Current Votes: 0

GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMention (google_api_content_warehouse v0.4.0)

Multiple entities can be identified on a document or query. Each entity can be mentioned several times in different positions on the document or query. This message describes a single mention of the entity. Note that a mention can be either explicit or implicit mentions. All explicit mentions refer to exact range in the document where the entity occurred, but implicit mentions may or may not have corresponding range. Next available tag number: 40

Attributes

addedByCloseAnswers (type: boolean(), default: nil) - Whether this mention was created by CloseAnswers on Postref. This bit is populated into corresponding intent_query through Aqua.
additionalExplainedRange (type: list(GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionAdditionalExplainedRange.t), default: nil) - If the mention's range is discontinuous, additional ranges that are a part of the mention but are not included in the begin/end range above.
begin (type: integer(), default: nil) - The [begin, end) byte offset, compatible with the Goldmine DocState representation. - For CONTENT, TITLE, META_CONTENT_TAG and IMG_ALT_TAG segments, the offsets are relative to the beginning of the document content. - For ANCHOR, QUERY, URL, IMAGE_QUERY, CONTEXT_ENTITY, CONTEXT_QUERY, SPORE_GRAPH, INSTANT_QUERY and VIDEO_TRANSCRIPT segments, the offsets are relative to the beginning of the corresponding (sub-)segment or text from doc attachment. The (indexing.annotations.goldmine) options are marks for Goldmine AnnotationsFinder to locate begin and end offsets in order to extract them
beginTokenIndex (type: integer(), default: nil) - The [begin end) token offsets in the Goldmine DocState. They follow similar rules as "begin" and "end" above, but for tokens rather than byte offsets. These fields are currently only populated by the query annotator.
compoundMention (type: list(GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionCompoundMention.t), default: nil) - For a compound mention, the references to the entity and mention of the components. Each compound_mention claims one or more 'mrf' fields from the WebrefEntity owning this mention. Use QueryJoinToMeaningStructConverter to expand compound mentions into complete MRF.
confidenceScore (type: number(), default: nil) - A probabilistic score describing how certain the annotator is that this exact range in the document or query refers to the entity. Probability that a mention is correct corresponds to confidence score roughly as follows: 0.3 -> 75% 0.5 -> 87% 0.7 -> 89% 0.9 -> 94% 1.0 -> 98% However if you consider all mentions with a score above 0.3, then most of these will have scores close to 1, so the overall precision of these mentions is around 95%.
debugInfo (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionDebugInfo.t, default: nil) - Debug information.
detailedMentionScores (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefDetailedMentionScores.t, default: nil) - Additional detailed scores about this mention.
end (type: integer(), default: nil) -
endTokenIndex (type: integer(), default: nil) -
evalInfo (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionEvalInfo.t, default: nil) - Information used by the evaluation tools to mark mentions annotations as correct/incorrect. This will never be annotated in production (would be nice if was, though :-)).
interpretationNumber (type: list(integer()), default: nil) - DEPRECATED and not populated anymore.
isImplicit (type: boolean(), default: nil) - True if the entity is mentioned implicitly. Note that a mention can be implicit and have a non-empty range, for example if the entity is Sports, and this is inferred from the mention of "gym".
lexicalAnnotation (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefLexicalAnnotation.t, default: nil) - Provides a Wordgraph lexical signals for the mentioned range so it can be use in LooseParsing. Please don't use this field before consulting wordgraph-team.
lowConfidence (type: boolean(), default: nil) - Qref & Postref only: Set to true for candidates part of qref candidate output for which there was no mention in qref. Internal to qref and LooseParsing. Do not use outside of these systems.
matchingText (type: String.t, default: nil) - Refcon-only: Annotated span in Refcon-normalized textual format, corresponding to the [begin end) offset interval in the CDoc. It will only be populated for CDocs originated from Docjoin extraction.
nameMetadata (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefConceptNameMetadata.t, default: nil) - Metadata attached to the name.
perMentionLightweightToken (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefLightweightTokensPerMentionLightweightToken.t, default: nil) - Provides concised access to all matched MatchedLightweightToken.
personalizationContextOutputs (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefPersonalizationContextOutputs.t, default: nil) - Additional personalization output scores about this mention.
priorProbability (type: number(), default: nil) - The prior probability of the entity for this mention.
referenceScore (type: number(), default: nil) - Whether the mention is a reference (e.g. it could be resolved to an entity coming from personal data) or not, currently only used for personal resolutions. Scores from 0 (not a reference) to 1 (reference). Use at your own risk as they are subject to change, advised to talk to refx-personal first.
resolutionScore (type: number(), default: nil) - Whether the mention is a resolution or not, currently only used for personal resolutions. Scores from 0 (not a resolution) to 1 (resolution). Use at your own risk as they are subject to change, advised to talk to refx-personal first.
stuff (type: GoogleApi.ContentWarehouse.V1.Model.Proto2BridgeMessageSet.t, default: nil) - Assorted things that can be added to a Mention.
subsegmentIndex (type: GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefSubSegmentIndex.t, default: nil) - Identifies the sub-segment where the annotation occurs. See SubSegmentIndex for details. Not present in QRef, also deprecated for URL segment types.
trustedNameConfidence (type: number(), default: nil) - Confidence that this name is a trusted name of the entity. This is set only in case the confidence is higher than an internal threshold (see ConceptProbability).

Summary

Types

t()

Functions

decode(value, options)

Unwrap a decoded JSON object into its complex fields.

Types

Link to this type

t()

@type t() :: %GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMention{
  addedByCloseAnswers: boolean() | nil,
  additionalExplainedRange:
    [
      GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionAdditionalExplainedRange.t()
    ]
    | nil,
  begin: integer() | nil,
  beginTokenIndex: integer() | nil,
  compoundMention:
    [
      GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionCompoundMention.t()
    ]
    | nil,
  confidenceScore: number() | nil,
  debugInfo:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionDebugInfo.t()
    | nil,
  detailedMentionScores:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefDetailedMentionScores.t()
    | nil,
  end: integer() | nil,
  endTokenIndex: integer() | nil,
  evalInfo:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefMentionEvalInfo.t()
    | nil,
  interpretationNumber: [integer()] | nil,
  isImplicit: boolean() | nil,
  lexicalAnnotation:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefLexicalAnnotation.t()
    | nil,
  lowConfidence: boolean() | nil,
  matchingText: String.t() | nil,
  nameMetadata:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefConceptNameMetadata.t()
    | nil,
  perMentionLightweightToken:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefLightweightTokensPerMentionLightweightToken.t()
    | nil,
  personalizationContextOutputs:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefPersonalizationContextOutputs.t()
    | nil,
  priorProbability: number() | nil,
  referenceScore: number() | nil,
  resolutionScore: number() | nil,
  stuff: GoogleApi.ContentWarehouse.V1.Model.Proto2BridgeMessageSet.t() | nil,
  subsegmentIndex:
    GoogleApi.ContentWarehouse.V1.Model.RepositoryWebrefSubSegmentIndex.t()
    | nil,
  trustedNameConfidence: number() | nil
}

Functions

Link to this function

decode(value, options)

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

Unwrap a decoded JSON object into its complex fields.