Systems and methods for ranking ephemeral content associated with a social networking system

Patent metadata

Field	Value
Publication number	US-10990635-B2
Application number	US-201715633643-A
Country	US
Kind code	B2
Filing date	Jun 26, 2017
Priority date	Jun 26, 2017
Publication date	Apr 27, 2021
Grant date	Apr 27, 2021

Abstract

Systems, methods, and non-transitory computer readable media can obtain a plurality of ephemeral content collections that are candidates for an ephemeral content feed of a user, wherein each of the plurality of ephemeral content collections includes one or more ephemeral content items. A score for each of the plurality of ephemeral content collections can be determined based at least in part on a probability of the user selecting the ephemeral content collection. The plurality of ephemeral content collections can be ranked based on the respective scores of the plurality of ephemeral content collections.

First claim

What is claimed is:

1. A computer-implemented method comprising:
obtaining, by a computing system, a plurality of ephemeral content collections that are candidates for an ephemeral content feed of a user, wherein each of the plurality of ephemeral content collections includes one or more ephemeral content items;
training a machine learning model based on a plurality of features including at least one of ephemeral content collection attributes, ephemeral content item attributes, or user attributes;
applying the machine learning model to determine, by the computing system, a score for each of the plurality of ephemeral content collections based at least in part on a first probability of the user selecting the ephemeral content collection and a second probability of the user sending a direct message to an authoring user of at least one ephemeral content item included in the ephemeral content collection, wherein the first probability is based at least in part on a number of times the user has selected past ephemeral content collections that include at least one past ephemeral content item associated with the authoring user and the second probability is based at least in part on a connection between the user and the authoring user; and
ranking, by the computing system, the plurality of ephemeral content collections based on the respective scores of the plurality of ephemeral content collections.

2. The computer-implemented method of claim 1 , wherein the score for each of the plurality of ephemeral content collections is determined based at least in part on a value model including one or more factors.

3. The computer-implemented method of claim 2 , wherein the one or more factors include one or more of: a third probability of the user sending a direct message associated with the ephemeral content collection to another user, a fourth probability of the user spending time on the ephemeral content collection, or a fifth probability of the user abandoning the ephemeral content collection.

4. The computer-implemented method of claim 1 , wherein each of the one or more ephemeral content items included in an ephemeral content collection of the plurality of ephemeral content collections is accessible only for a predetermined time period.

5. The computer-implemented method of claim 4 , wherein an ephemeral content collection of the plurality of ephemeral content collections is accessible only when at least one of the one or more ephemeral content items included in the ephemeral content collection is accessible.

6. The computer-implemented method of claim 1 , wherein each of the plurality of ephemeral content collections is associated with a specific user and includes one or ephemeral content items created by the specific user.

7. The computer-implemented method of claim 1 , wherein each of the plurality of ephemeral content collections is associated with a topic and is not associated with a specific user.

8. The computer-implemented method of claim 1 , further comprising ranking the one or more ephemeral content items included in an ephemeral content collection of the plurality of ephemeral content collections.

9. The computer-implemented method of claim 1 , further comprising providing at least some of the ranked plurality of ephemeral content collections in the ephemeral content feed of the user.

10. A system comprising:
at least one hardware processor; and
a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
obtaining a plurality of ephemeral content collections that are candidates for an ephemeral content feed of a user, wherein each of the plurality of ephemeral content collections includes one or more ephemeral content items;
training a machine learning model based on a plurality of features including at least one of ephemeral content collection attributes, ephemeral content item attributes, or user attributes;
applying the machine learning model to determine a score for each of the plurality of ephemeral content collections based at least in part on a first probability of the user selecting the ephemeral content collection and a second probability of the user sending a direct message to an authoring user of at least one ephemeral content item included in the ephemeral content collection, wherein the first probability is based at least in part on a number of times the user has selected past ephemeral content collections that include at least one past ephemeral content item associated with the authoring user and the second probability is based at least in part on a connection between the user and the authoring user; and
ranking the plurality of ephemeral content collections based on the respective scores of the plurality of ephemeral content collections.

11. The system of claim 10 , wherein the score for each of the plurality of ephemeral content collections is determined based at least in part on a value model including one or more factors.

12. The system of claim 11 , wherein the one or more factors include one or more of: a third probability of the user sending a direct message associated with the ephemeral content collection to another user, a fourth probability of the user spending time on the ephemeral content collection, or a fifth probability of the user abandoning the ephemeral content collection.

13. The system of claim 10 , wherein each of the one or more ephemeral content items included in an ephemeral content collection of the plurality of ephemeral content collections is accessible only for a predetermined time period.

14. The system of claim 10 , wherein the instructions further cause the system to perform ranking the one or more ephemeral content items included in an ephemeral content collection of the plurality of ephemeral content collections.

15. A non-transitory computer readable medium including instructions that, when executed by at least one hardware processor of a computing system, cause the computing system to perform a method comprising:
obtaining a plurality of ephemeral content collections that are candidates for an ephemeral content feed of a user, wherein each of the plurality of ephemeral content collections includes one or more ephemeral content items;
training a machine learning model based on a plurality of features including at least one of ephemeral content collection attributes, ephemeral content item attributes, or user attributes;
applying the machine learning model to determine a score for each of the plurality of ephemeral content collections based at least in part on a first probability of the user selecting the ephemeral content collection and a second probability of the user sending a direct message to an authoring user of at least one ephemeral content item included in the ephemeral content collection, wherein the first probability is based at least in part on a number of times the user has selected past ephemeral content collections that include at least one past ephemeral content item associated with the authoring user and the second probability is based at least in part on a connection between the user and the authoring user; and
ranking the plurality of ephemeral content collections based on the respective scores of the plurality of ephemeral content collections.

16. The non-transitory computer readable medium of claim 15 , wherein the score for each of the plurality of ephemeral content collections is determined based at least in part on a value model including one or more factors.

17. The non-transitory computer readable medium of claim 16 , wherein the one or more factors includ

Assignees

• Facebook Inc

Inventors

• Dimson Thomas Frederick
• Wang Rui

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9536 Primary

  Search customisation based on social or collaborative filtering · CPC title

• H04L67/535

  Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34) · CPC title

• H04L67/306

  User profiles · CPC title