Automatic targeting of content by clustering based on user feedback data

• US 10,346,871 B2
• Filed: 04/22/2016
• Issued: 07/09/2019
• Est. Priority Date: 04/22/2016
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Jeon, Sung-eok, Husain, Aliasgar Mumtaz
• Primary Examiner(s): Brown, Luis A
• Application Number:
  US15/136,814
• Publication Number:
  US 20170308924A1
• Time in Patent Office: 1,173 Days
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0261 based on user location
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute

Abstract

An online system automatically and dynamically determines an audience for content by clustering users across various dimensions, and refining targeting criteria for the content. The online system receives content and initial targeting criteria from a content provider. The content is provided to a group of users that meet the initial targeting criteria. The system collects content response data from the group of users that were provided the content, including user responses to the content and dynamic data relating to time and location of the user responses. The content response data is further integrated with user characteristics, content presentation data, and social response data to generate integrated user-content data of the content. Clusters of users are generated based on features of the integrated user-content data, and refined targeting criteria are identified based on the generated clusters that can then be used for more accurate targeting of the content to users.

Claims

1. A method comprising:

receiving, at an online system, a sponsored content item and initial targeting criteria from a content provider, the initial targeting criteria comprising a set of rules for targeting the sponsored content item to users of the online system;

providing, over a period of time, the sponsored content item for display to a first group of users of the online system meeting the initial targeting criteria;

collecting content response data on the sponsored content item from the first group of users, the content response data comprising user responses of the first group of users to the sponsored content item and dynamic data relating to time and location of the user responses, wherein the dynamic data comprises proximity of a user to a physical location related to the sponsored content item at a time of the user response for the user, or a characteristic time of day, week, month, year, or season of the user response;

integrating the content response data with corresponding user characteristic data, presentation conditions of the sponsored content item, and social response data to generate integrated user-content data for the sponsored content item, the user characteristic data comprising features relating to a user profile of and actions of the user in the online system, the presentation conditions comprising features relating to presentation times and locations of the sponsored content item, and the social response data comprising features relating to responses of other users associated with the first group of users to the sponsored content item;

identifying, by the online system for the sponsored content item, one or more clusters of users based on the features of the integrated user-content data, each of the one or more clusters being specific to the sponsored content item and being associated with a type of user response to the sponsored content item;

identifying, by the online system for the sponsored content item, refined targeting criteria associated with a set of features of the integrated user-content data based on the one or more clusters identified for the sponsored content item, wherein identifying the refined targeting criteria further comprises extracting a range of a cluster in the one or more clusters along a feature relating to the proximity of the user to the physical location or the characteristic time; and

targeting the sponsored content item to additional users of the online system meeting the refined targeting criteria for the sponsored content item.

2. The method of claim 1, wherein the sponsored content item is any one of a series of sponsored content items in a campaign.

3. The method of claim 1, wherein the type of user responses comprise active user responses for users that interacted with or converted based on the sponsored content item, and inactive user responses for users that did not interact with or did not convert based on the sponsored content item.

4. The method of claim 3, wherein identifying the refined targeting criteria comprises identifying active targeting criteria from clusters associated with the active user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises providing the sponsored content item to a second group of users meeting the active targeting criteria.

5. The method of claim 3, wherein identifying the refined targeting criteria comprises identifying inactive targeting criteria from clusters associated with the inactive user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises targeting the sponsored content item to a second group of users that do not meet the inactive targeting criteria.

6. The method of claim 1, wherein identifying the refined targeting criteria further comprises extracting another range of another cluster in the one or more clusters along another feature in the set of features.

7. The method of claim 1, wherein the users associated with the first group of users are users connected to the first group of users through edges in a social networking system.

8. The method of claim 7, wherein the social response data comprises a number of users connected to the first group of users having an active response to the sponsored content item.

9. The method of claim 1, wherein the users associated with the first group of users are users identified to have user characteristics similar to the first group of users.

10. The method of claim 1, wherein the one or more clusters are identified using a K-means clustering algorithm or a Gaussian mixture algorithm.

11. A computer program product comprising a non-transitory computer-readable storage medium containing computer program code for:

receiving, at an online system, a sponsored content item and initial targeting criteria from a content provider, the initial targeting criteria comprising a set of rules for targeting the sponsored content item to users of the online system;

providing, over a period of time, the sponsored content item for display to a first group of users of the online system meeting the initial targeting criteria;

collecting content response data on the sponsored content item from the first group of users, the content response data comprising user responses of the first group of users to the sponsored content item and dynamic data relating to time and location of the user responses, wherein the dynamic data comprises proximity of a user to a physical location related to the sponsored content item at a time of the user response for the user, or a characteristic time of day, week, month, year, or season of the user response;

integrating the content response data with corresponding user characteristic data, presentation conditions of the sponsored content item, and social response data to generate integrated user-content data for the sponsored content item, the user characteristic data comprising features relating to a user profile of and actions of the user in the online system, the presentation conditions comprising features relating to presentation times and locations of the sponsored content item, and the social response data comprising features relating to responses of other users associated with the first group of users to the sponsored content item;

identifying, by the online system for the sponsored content item, one or more clusters of users based on the features of the integrated user-content data, each of the one or more clusters being specific to the sponsored content item and being associated with a type of user response to the sponsored content item;

identifying, by the online system for the sponsored content item, refined targeting criteria associated with a set of features of the integrated user-content data based on the one or more clusters identified for the sponsored content item, wherein identifying the refined targeting criteria further comprises extracting a range of a cluster in the one or more clusters along a feature relating to the proximity of the user to the physical location or the characteristic time; and

targeting the sponsored content item to additional users of the online system meeting the refined targeting criteria for the sponsored content item.

12. The computer product program of claim 11, wherein the sponsored content item is any one of a series of sponsored content items in a campaign.

13. The computer product program of claim 11, wherein the type of user responses comprise active user responses for users that interacted with or were converted based on the sponsored content item, and inactive user responses for users that did not interact with or were not converted based on the sponsored content item.

14. The computer product program of claim 13, wherein identifying the refined targeting criteria comprises identifying active targeting criteria from clusters associated with the active user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises providing the sponsored content item to a second group of users meeting the active targeting criteria.

15. The computer product program of claim 13, wherein identifying the refined targeting criteria comprises identifying inactive targeting criteria from clusters associated with the inactive user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises targeting the sponsored content item to a second group of users that do not meet the inactive targeting criteria.

16. The computer product program of claim 11, wherein identifying the refined targeting criteria further comprises extracting another range of another cluster in the one or more clusters along another feature in the set of features.

17. The computer product program of claim 11, wherein the users associated with the first group of users are users connected to the first group of users through edges in a social networking system.

18. The computer product program of claim 17, wherein the social response data comprises a number of users connected to the first group of users having an active response to the sponsored content item.

19. The computer product program of claim 11, wherein the users associated with the first group of users are users identified to have user characteristics similar to the first group of users.

20. The computer product program of claim 11, wherein the one or more clusters are identified using a K-means clustering algorithm, or a Gaussian mixture algorithm.

21. A system comprising:

a processor configured to execute instructions;

a non-transitory computer-readable medium containing instructions for execution on the processor, the instructions causing the processor to perform steps of;

receiving, at an online system, a sponsored content item and initial targeting criteria from a content provider, the initial targeting criteria comprising a set of rules for targeting the sponsored content item to users of the online system;

providing, over a period of time, the sponsored content item for display to a first group of users of the online system meeting the initial targeting criteria;

collecting content response data on the sponsored content item from the first group of users, the content response data comprising user responses of the first group of users to the sponsored content item and dynamic data relating to time and location of the user responses, wherein the dynamic data comprises proximity of a user to a physical location related to the sponsored content item at a time of the user response for the user, or a characteristic time of day, week, month, year, or season of the user response;

integrating the content response data with corresponding user characteristic data, presentation conditions of the sponsored content item, and social response data to generate integrated user content data for the sponsored content item, the user characteristic data comprising features relating to a user profile of and actions of the user in the online system, the presentation conditions comprising features relating to presentation times and locations of the sponsored content item, and the social response data comprising features relating to responses of other users associated with the first group of users to the sponsored content item;

identifying, by the online system for the sponsored content item, one or more clusters of users based on the features of the integrated user content data, each of the one or more clusters being specific to the sponsored content item and being associated with a type of user response to the sponsored content item;

identifying, by the online system for the sponsored content item, refined targeting criteria associated with a set of features of the integrated user-content data based on the one or more clusters identified for the sponsored content item, wherein identifying the refined targeting criteria further comprises extracting a range of a cluster in the one or more clusters along a feature relating to the proximity of the user to the physical location or the characteristic time; and

targeting the sponsored content item to additional users of the online system meeting the refined targeting criteria for the sponsored content item.

22. The system of claim 21, wherein the sponsored content item is any one of a series of sponsored content items in a campaign.

23. The system of claim 22, wherein the type of user responses comprise active user responses for users that interacted with or converted based on the sponsored content item, and inactive user responses for users that did not interact with or did not convert based on the sponsored content item.

24. The system of claim 23, wherein identifying the refined targeting criteria comprises identifying active targeting criteria from clusters associated with the active user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises providing the sponsored content item to a second group of users meeting the active targeting criteria.

25. The system of claim 23, wherein identifying the refined targeting criteria comprises identifying inactive targeting criteria from clusters associated with the inactive user responses, and wherein targeting the sponsored content item based on the refined targeting criteria comprises targeting the sponsored content item to a second group of users that do not meet the inactive targeting criteria.

26. The system of claim 21, wherein identifying the refined targeting criteria further comprises extracting another range of another cluster in the one or more clusters along another feature in the set of features.

27. The system of claim 25, wherein the users associated with the first group of users are users connected to the first group of users through edges in a social networking system.

28. The system of claim 27, wherein the social response data comprises a number of users connected to the first group of users having an active response to the sponsored content item.

29. The system of claim 25, wherein the users associated with the first group of users are users identified to have user characteristics similar to the first group of users.

30. The system of claim 25, wherein the one or more clusters are identified using a K-means clustering algorithm or a Gaussian mixture algorithm.