US 11,531,925 B2Granted

Optimizing content distribution using a model

AssigneeGoogle Inc.

InventorsScott Tadashi Davies, Kai Chen, Michael Jee-Kai Wang, Wei Jiang, Maryam Tavafi, Peter Zaimis Tipton

Life of this patent

Estimating the term…

Loading patent…

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimizing content presentation. In one aspect, a system includes a training database that stores training data including attribute information about users and corresponding proxy metrics quantifying behavior by the users following content presentation; a content database; a model generator that accesses the training data and trains a model for content distribution; and a content distribution server that receives a content request, uses the model to select content, transmits data identifying the selected content, wherein the model: obtains a set of attributes for a user associated with the request, receives information about a given content, predicts a proxy metric based on the set of attributes and the information about the content, the predicted proxy metric providing information about subject retention or awareness; and identifies the given content for distribution if the predicted proxy metrics meet a threshold.

Claims16

Dependent claims

A system comprising:

a training database that stores training data including attribute information about a plurality of users and corresponding proxy metrics that quantify online behavior by the plurality of users following video presentation;
a content database that stores video content received from various video content providers;
a model generator, including one or more data processors, that accesses the training data in the training database and trains a model for video distribution based on the training data;
and
a content distribution server that receives a request for video content to be presented at a client device, wherein the content distribution server uses the model to select video content to be distributed to the client device in response to the request and transmits data identifying the selected video content to the client device,
wherein the model:
obtains a set of attributes from the request, wherein the set of attributes relate to information about a user associated with the request,
receives information about a given video stored in the content database,
determines, for the given video, whether subject retention information that was obtained directly from users through a feedback mechanism is available in the content database, wherein the subject retention information indicates a likelihood that, when asked at a specified time after presentation of the given video has ended, the user will respond that they remember seeing the given video and be able to describe a subject of the given video;
predicts, for the given video and based on the set of attributes and the information about the given video, a likelihood that, when asked at a specified time after presentation of the given video, the user will respond that they remember seeing the given video, wherein the prediction is determined by using a proxy metric of an amount of time that other content was presented before user input that interrupted presentation of the other content was received when the subject retention information obtained directly from users through the feedback mechanism is currently unavailable to the model;
and
identifies, using the predicted likelihood that the user will respond that they remember seeing the given video and for the content distribution server, the given video for distribution based on the likelihood having a value that is indicative of the user responding that they remember seeing the given video after presentation of the given video has ended.

9 more elements

A computer-implemented method comprising:

storing, in a training database, training data including attribute information about a plurality of users and corresponding proxy metrics that quantify online behavior following video presentation;
storing, in a content database, video content received from various video content providers;
accessing, by a model generator that includes one or more data processors, the training data in the training database and training a model for video distribution based on the training database;
receiving, by a content distribution server, a request for video to be presented at a client device;
selecting, by the content distribution server and using the model, video content to be distributed to the client device in response to the request, wherein the model:
obtains a set of attributes from the request, wherein the set of attributes relate to information about a user associated with the request,
receives information about a given video stored in the content database,
receives information about a given video stored in the content database,
determines, for the given video, whether subject retention information that was obtained directly from users through a feedback mechanism is available in the content database, wherein the subject retention information indicates a likelihood that, when asked at a specified time after presentation of the given video has ended, the user will respond that they remember seeing the given video and be able to describe a subject of the given video;
predicts, for the given video and based on the set of attributes and the information about the given video, a likelihood that, when asked at a specified time after presentation of the given video, the user will respond that they remember seeing the given video, wherein the prediction is determined by using a proxy metric of an amount of time that other content was presented before user input that interrupted presentation of the other content was received when the subject retention information obtained directly from users through the feedback mechanism is currently unavailable to the model;
and
identifies, using the predicted likelihood that the user will respond that they remember seeing the given video and for the content distribution server, the given video for distribution based on the likelihood having a value that is indicative of the user responding that they remember seeing the given video after presentation of the given video has ended;
and
transmitting, by the content distribution server, data identifying the given video to the client device.

11 more elements

One or more non-transitory computer-readable media having instructions stored thereon that, when executed by one or more processors, cause performance of operations comprising:

storing, in a training database, training data including attribute information about a plurality of users and corresponding proxy metrics that quantify online behavior following video presentation;
storing, in a content database, video content received from various video content providers;
accessing the training data in the training database and training a model for video distribution based on the training database;
receiving a request for video to be presented at a client device;
selecting, using the model, video content to be distributed to the client device in response to the request, wherein the model:
obtains a set of attributes from the request, wherein the set of attributes relate to information about a user associated with the request,
receives information about a given video stored in the content database,
receives information about a given video stored in the content database,
determines, for the given video, whether subject retention information that was obtained directly from users through a feedback mechanism is available in the content database, wherein the subject retention information indicates a likelihood that, when asked at a specified time after presentation of the given video has ended, the user will respond that they remember seeing the given video and be able to describe a subject of the given video;
predicts, for the given video and based on the set of attributes and the information about the given video, a likelihood that, when asked at a specified time after presentation of the given video, the user will respond that they remember seeing the given video, wherein the prediction is determined by using a proxy metric of an amount of time that other content was presented before user input that interrupted presentation of the other content was received when the subject retention information obtained directly from users through the feedback mechanism is currently unavailable to the model;
and
identifies, using the predicted likelihood that the user will respond that they remember seeing the given video and for a content distribution server, the given video for distribution based on the likelihood having a value that is indicative of the user responding that they remember seeing the given video after presentation of the given video has ended;
and
transmitting, by the content distribution server, data identifying the given video to the client device.

11 more elements

Dependencies are inferred from the claim text.