Systems and methods for generating content

Patent metadata

Field	Value
Publication number	US-10496750-B2
Application number	US-201715707339-A
Country	US
Kind code	B2
Filing date	Sep 18, 2017
Priority date	Sep 18, 2017
Publication date	Dec 3, 2019
Grant date	Dec 3, 2019

Abstract

Systems, methods, and non-transitory computer-readable media can project data describing a set of media items from which a representative media item is to be determined. A medoid can be determined from the projected data. A media item corresponding to the medoid can be selected as the representative media item of the set of media items.

First claim

What is claimed is:

1. A computer-implemented method comprising:
projecting, by a computing system, data describing a set of media items from which a representative media item is to be determined, wherein the set of media items correspond to media items included in a multi-author story published through a social networking system, and wherein the set of media items includes one or more ephemeral media items associated with the multi-author story, wherein each of the one or more ephemeral media items expires after a pre-defined period of time;
determining, by the computing system, a medoid from the projected data; and
selecting, by the computing system, a media item corresponding to the medoid as the representative media item of the set of media items, wherein the selected media item is an ephemeral media item.

2. The computer-implemented method of claim 1 , wherein projecting the data describing the set of media items further comprises:
determining, by the computing system, respective semantic feature representations for each of the media items in the set; and
projecting, by the computing system, the semantic feature representations in a semantic space.

3. The computer-implemented method of claim 2 , wherein a semantic feature representation corresponds to a high-dimensional floating point vector, and wherein the high-dimensional floating point vector is projected onto a two-dimensional semantic space.

4. The computer-implemented method of claim 1 , wherein a semantic feature representation of a media item is determined based at least in part on visual features included in subject matter captured in the media item.

5. The computer-implemented method of claim 1 , wherein selecting the media item corresponding to the medoid further comprises:
identifying, by the computing system, a semantic feature representation corresponding to the medoid; and
identifying, by the computing system, the media item from which the semantic feature representation was determined.

6. The computer-implemented method of claim 1 , wherein the representative media item is associated with the multi-author story.

7. The computer-implemented method of claim 6 , wherein the representative media item is used as an icon to represent the multi-author story in an interface.

8. The computer-implemented method of claim 1 , wherein the set of media items correspond to media items included in a multi-author story, the method further comprising:
determining, by the computing system, a new representative media item for the multi-author story in response to at least one trigger condition being satisfied.

9. The computer-implemented method of claim 8 , wherein the trigger condition is satisfied when the media item corresponding to the medoid expires, after a specified time period, or a threshold change in the number of media items included in the set of media items.

10. The computer-implemented method of claim 8 , wherein determining the new representative media item further comprises:
obtaining, by the computing system, an updated set of media items included in the multi-author story;
projecting, by the computing system, data describing the updated set of media items;
determining, by the computing system, a new medoid from the projected data describing the updated set of media items; and
selecting, by the computing system, a new media item corresponding to the new medoid as the new representative media item of the updated set of media items.

11. A system comprising:
at least one processor; and
a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
projecting data describing a set of media items from which a representative media item is to be determined, wherein the set of media items correspond to media items included in a multi-author story published through a social networking system, and wherein the set of media items includes one or more ephemeral media items associated with the multi-author story, wherein each of the one or more ephemeral media items expires after a pre-defined period of time;
determining a medoid from the projected data; and
selecting a media item corresponding to the medoid as the representative media item of the set of media items, wherein the selected media item is an ephemeral media item.

12. The system of claim 11 , wherein projecting the data describing the set of media items further causes the system to perform:
determining respective semantic feature representations for each of the media items in the set; and
projecting the semantic feature representations in a semantic space.

13. The system of claim 12 , wherein a semantic feature representation corresponds to a high-dimensional floating point vector, and wherein the high-dimensional floating point vector is projected onto a two-dimensional semantic space.

14. The system of claim 11 , wherein a semantic feature representation of a media item is determined based at least in part on visual features included in subject matter captured in the media item.

15. The system of claim 11 , wherein selecting the media item corresponding to the medoid further causes the system to perform:
identifying a semantic feature representation corresponding to the medoid; and
identifying the media item from which the semantic feature representation was determined.

16. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:
projecting data describing a set of media items from which a representative media item is to be determined, wherein the set of media items correspond to media items included in a multi-author story published through a social networking system, and wherein the set of media items includes one or more ephemeral media items associated with the multi-author story, wherein each of the one or more ephemeral media items expires after a pre-defined period of time;
determining a medoid from the projected data; and

selecting a media item corresponding to the medoid as the representative media item of the set of media items, wherein the selected media item is an ephemeral media item.

17. The non-transitory computer-readable storage medium of claim 16 , wherein projecting the data describing the set of media items further causes the computing system to perform:
determining respective semantic feature representations for each of the media items in the set; and
projecting the semantic feature representations in a semantic space.

18. The non-transitory computer-readable storage medium of claim 17 , wherein a semantic feature representation corresponds to a high-dimensional floating point vector, and wherein the high-dimensional floating point vector is projected onto a two-dimensional semantic space.

19. The non-transitory computer-readable storage medium of claim 16 , wherein a semantic feature representation of a media item is determined based at least in part on visual features included in subject matter captured in the media item.

20. The non-transitory computer-readable storage medium of claim 16 , wherein selecting the media item corresponding to the medoid further causes the computing system to perform:
identifying a semantic feature representation corresponding to the medoid; and
identifying the media item from which the semantic feature representation was determined.

Assignees

• Facebook Inc

Inventors

• Dimson Thomas Frederick
• Gredoña Clark Martin

Classifications

• G06F3/04842

  Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence) · CPC title

• G06F3/0488

  using a touch-screen or digitiser, e.g. input of commands through traced gestures · CPC title

• G06F16/75

  Clustering; Classification · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F3/04817 Primary

  using icons (graphical or visual programming using iconic symbols G06F8/34) · CPC title