Watch-time clustering for improving video searches, selection and provision

Patent metadata

Field	Value
Publication number	US-10003847-B2
Application number	US-201615136558-A
Country	US
Kind code	B2
Filing date	Apr 22, 2016
Priority date	Apr 22, 2016
Publication date	Jun 19, 2018
Grant date	Jun 19, 2018

Abstract

This document describes, among other things, systems, methods, devices, and other techniques for using information about how long various videos were presented at client devices to determine subsequent video recommendations and search results. In some implementations, a computing can include a modeling apparatus, a front-end server, a request manager, one or more video file storage devices, a video selector, or a combination of some or all of these. The video selector can select video content for a particular digitized video among a plurality of digitized videos to serve to a computing device responsive to a request. The selection can be based at least in part on how long the particular digitized video has been presented at client devices associated with users having characteristics that match one or more characteristics of the user that submitted the request for video content, as indicated by the modeling apparatus.

First claim

What is claimed is:

1. A computing system for serving videos over a network, the computing system comprising:
a modeling apparatus that (i) obtains, for various digitized videos that the computing system has presented at client devices over a network, watch time information that specifies an amount of time that the various digitized videos were presented at the client devices, (ii) groups the watch time information into different groups based on characteristics of users associated with the client devices, and (iii) determines, based on the grouping of watch time information and for each particular group of users of a plurality of groups of users, an accumulated watch time that indicates how long the various digitized videos were presented to users within the particular group of users;
a front-end server that receives a request for video content and serves video content identified in response to the request over the network to a computing device that is separate from the computing system for presentation at the computing device;
a request manager, including one or more processors, that analyzes the request for video content and identifies selection criteria in the request, including identifying one or more characteristics of a user that submitted the request for video content;
one or more video file storage devices that store a plurality of digitized videos that have been made available by various parties for distribution over the network; and
a video selector, including one or more processors, that selects, from the video file storage devices, video content of a particular digitized video among the plurality of digitized videos to serve to the computing device in response to the request, wherein the selection is based at least in part on a first accumulated watch time that indicates how long the particular digitized video has been presented at client devices associated with users having characteristics that match the one or more characteristics of the user that submitted the request for video content, as indicated by the modeling apparatus.

2. The computing system of claim 1 , wherein the modeling apparatus:
groups the watch time information into different groups by assigning respective portions of the watch time information to different viewer categories that correspond to different sets of characteristics of the users associated with the client devices at which the various digitized videos were presented; and
identifies, for each respective viewer category, a distribution of watch time of users in the respective viewer category among digitized videos in a group of digitized videos that were at least partially presented at computing devices associated with users in the respective category.

3. The computing system of claim 1 , wherein, for each respective digitized video among the plurality of digitized videos stored on the one or more video file storage devices, the modeling apparatus:
identifies, from the watch time information and for each respective viewer category among a plurality of viewer categories, how long the respective digitized video was presented at client devices associated with users in the respective viewer category; and
assigns to the respective digitized video, for each respective viewer category among the plurality of viewer categories and based on how long the respective digitized video was presented at client devices associated with users in the respective viewer category, a score that indicates a relevance of the respective digitized video to the respective viewer category.

4. The computing system of claim 1 , wherein the one or more characteristics of the user that submitted the request for video content comprise at least one of demographic characteristics of the user or behavioral characteristics of the user.

5. The computing system of claim 4 , wherein the one or more characteristics of the user that submitted the request for video content comprise demographic characteristics that indicate at least an age or gender of the user.

6. The computing system of claim 4 , wherein:
the one or more characteristics of the user that submitted the request for video content comprise behavioral characteristics that indicate one or more queries submitted by the user in a same session in which the user submitted the request for video content; and
the video selector selects to serve the video content of the particular digitized video among the plurality of digitized videos based at least on how long the particular digitized video was presented at client devices associated with other users during sessions in which the other users also submitted the one or more queries or at least one query determined to be similar to the one or more queries.

7. The computing system of claim 4 , wherein:
the one or more characteristics of the user that submitted the request for video content comprise behavioral characteristics that indicate a navigation history of the user associated with the request for video content; and
the video selector selects to serve the video content of the particular digitized video among the plurality of digitized videos based at least on how long the particular digitized video was presented at client devices associated with other users having a same or similar navigation history as the user that submitted the request, in browsing sessions that led the other users to watch the particular digitized video.

8. The computing system of claim 1 , wherein the video selector is further configured to select video content to serve to various computing devices as recommendations to respective users of the various computing devices.

9. The computing system of claim 1 , wherein:
in response to receiving the request for video content, the request manager is configured to identify one or more topics of the requested video content, including identifying the topics from at least one of a user-entered query that is a part of the request or identifying the topics from information associated with the user that is not included in the request; and
the video selector is configured to select video content to serve to the computing device by:
identifying digitized videos that are determined to be relevant to the one or more topics;
ranking the identified digitized videos based at least on how long each of the identified digitized videos has been presented at client devices associated with users having characteristics that match the one or more characteristics of the user that submitted the request for video content; and
based on the ranking, selecting video content of one or more top-ranked digitized videos to serve to the computing device, including the video content of the particular digitized video.

10. The computing system of claim 1 , further comprising an instrumentation manager, including one or more processors, that generates and embeds computer code in various web pages that the computing system serves to various client devices for the presentation of various digitized videos, wherein the computer code is configured, when executed on the various client devices, to cause the various computing devices to periodically report over the network and to the computing system amounts of time that the various digitized videos were presented at the various client devices.

11. The computing system of claim 1 , wherein:
the front-end server is configured to serve the video content of the particular digitized video by streaming the video content to the computing device, and
the computing device is configured to (i) receive the stream of the video content of the particular digitized video from the computing system and (ii) play the stream of the video content.

12. A computer-

Assignees

• Google Llc

Inventors

• Lewis Justin
• Durbin Robert Alexander

Classifications

• G06F16/735 Primary

  Filtering based on additional data, e.g. user or group profiles · CPC title

• H04N21/25891

  being end-user preferences (retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18) · CPC title

• H04N21/4665

  involving classification methods, e.g. Decision trees · CPC title

• H04N21/25883

  being end-user demographical data, e.g. age, family status or address (arrangements for identifying locations of users in broadcast systems H04H60/52) · CPC title

• H04N21/4532

  involving end-user characteristics, e.g. viewer profile, preferences (monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18) · CPC title