VideoContentSearchAnchorCommonFeatureSet

AI Overview😉

  • The potential purpose of this module is to analyze and rank video content based on various features, such as the relevance of the video title and description to the anchor text, the descriptiveness and usefulness of the anchor, and the predicted retention probability of the video interval associated with the anchor.
  • This module could impact search results by influencing the ranking of video content in search engine results pages (SERPs). Videos with more relevant and descriptive anchors, higher descriptiveness and usefulness scores, and higher predicted retention probabilities may be ranked higher in search results.
  • To be more favorable for this function, a website may:
    • Optimize video titles and descriptions to be more relevant and descriptive of the video content.
    • Use high-quality and descriptive anchor text in video captions and descriptions.
    • Improve the overall user experience and engagement of the video content to increase predicted retention probabilities.
    • Use natural language processing (NLP) and machine learning (ML) techniques to generate high-quality and relevant video captions and descriptions.

Interesting Module? Vote 👇

Voting helps other researchers find interesting modules.

Current Votes: 0

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

Contains anchor level features that apply to all anchor types. Next id: 22.

Attributes

  • anchorQbstDistance (type: number(), default: nil) - QBST distance between the anchor and the top navboost query of the video if exists, or the video title otherwise.
  • asrAverageBabelSimilarityScore (type: number(), default: nil) - Average of babel similarity between the anchor and all asr sentences.
  • asrMaximumBabelSimilarityScore (type: number(), default: nil) - Maximum babel similarity between the anchor and the asr sentences.
  • bleurtFeatures (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchBleurtFeatures.t, default: nil) - Features needed for Bleurt inference.
  • bleurtScore (type: number(), default: nil) - The Bleurt inference score generated using the bleurt_features.
  • descartesScoreWithTitle (type: number(), default: nil) - Descartes similarity score between video title and anchor label.
  • descriptionAverageBabelSimilarityScore (type: number(), default: nil) - Average of babel similarity between the anchor and all description sentences.
  • descriptionMaximumBabelSimilarityScore (type: number(), default: nil) - Maximum babel similarity between the anchor and the description sentences.
  • dolphinDescriptivenessScore (type: number(), default: nil) - The predicted descriptiveness and usefulness rating scores generated by the Unified Dolphin model. Rating template: experimental/video/video_anchors_oneside_without_thumbnail/template.jhtml
  • dolphinEnsembleScore (type: list(GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDolphinEnsembleScore.t), default: nil) - If the dolphin model is an ensemble model, this contains the scores associated to each individual ensemble model.
  • dolphinFeatures (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDolphinFeatures.t, default: nil) - The features used to generate the Dolphin score.
  • dolphinScore (type: number(), default: nil) - The score generated by the Dolphin callout model.
  • dolphinUsefulnessScore (type: number(), default: nil) -
  • labelPhraseEmbedding (type: list(number()), default: nil) - A phrase embedding for the anchor label. The model used to generate the embedding can be found in VideoAnchorSets: video_score_info.common_features.label_phrase_embedding_model
  • mumDescriptivenessScore (type: number(), default: nil) - The predicted descriptiveness of the anchor using the MUM unified scoring model.
  • mumUsefulnessScore (type: number(), default: nil) - The predicted usefulness of the anchor using the MUM unified scoring model.
  • retentionScore (type: number(), default: nil) - A score that is correlated with retention probability of the interval associated with this anchor (start time to end time). Retention probability of an interval is 1 - (probability the user does not watch the interval all the way through, given they started watching it). This score may be predicted by a model, or calculated from actual retention data.
  • saftDocument (type: GoogleApi.ContentWarehouse.V1.Model.NlpSaftDocument.t, default: nil) - A saft document generated from the anchor label.
  • timedLabelFeatures (type: list(GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchCaptionLabelFeatures.t), default: nil) - For annotating labels and their timing and context info. For example, this is used for anchor labels within a passage.
  • timestamp (type: list(GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchAnchorCommonFeatureSetLabelSpanTimestamp.t), default: nil) -
  • titleAnchorBabelMatchScore (type: number(), default: nil) - Babel similarity between the anchor and the video title.

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.VideoContentSearchAnchorCommonFeatureSet{
    anchorQbstDistance: number() | nil,
    asrAverageBabelSimilarityScore: number() | nil,
    asrMaximumBabelSimilarityScore: number() | nil,
    bleurtFeatures:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchBleurtFeatures.t()
      | nil,
    bleurtScore: number() | nil,
    descartesScoreWithTitle: number() | nil,
    descriptionAverageBabelSimilarityScore: number() | nil,
    descriptionMaximumBabelSimilarityScore: number() | nil,
    dolphinDescriptivenessScore: number() | nil,
    dolphinEnsembleScore:
      [
        GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDolphinEnsembleScore.t()
      ]
      | nil,
    dolphinFeatures:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDolphinFeatures.t()
      | nil,
    dolphinScore: number() | nil,
    dolphinUsefulnessScore: number() | nil,
    labelPhraseEmbedding: [number()] | nil,
    mumDescriptivenessScore: number() | nil,
    mumUsefulnessScore: number() | nil,
    retentionScore: number() | nil,
    saftDocument: GoogleApi.ContentWarehouse.V1.Model.NlpSaftDocument.t() | nil,
    timedLabelFeatures:
      [
        GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchCaptionLabelFeatures.t()
      ]
      | nil,
    timestamp:
      [
        GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchAnchorCommonFeatureSetLabelSpanTimestamp.t()
      ]
      | nil,
    titleAnchorBabelMatchScore: number() | nil
  }

Functions

Link to this function

decode(value, options)

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

Unwrap a decoded JSON object into its complex fields.