Systems and methods for managing content
Systems, methods, and non-transitory computer readable media are configured to determine a likelihood of a user choosing to reveal a given content item when contents of the content item are obscured. The likelihood can be determined based at least in part on a trained machine learning model. An extent by which to obscure the content item based at least in part on the likelihood can be determined. Subsequently, an obscured version of the content item can be provided for display. The content item can be obscured based at least in part on the determined extent.
1. A computer-implemented method comprising:
generating, by a computing system, an obscured version of a content item, wherein the generating the obscured version of the content item comprises:
determining a likelihood that a user will reveal the obscured version of the content item based at least in part on i) user feature data associated with the user, wherein the user feature data includes at least one of: demographic information or social networking system information, and ii) a machine learning model,
and wherein the generating the obscured version of the content item further comprises at least one of:
applying a color, wherein an opacity of the color is inversely proportional to the likelihood that the user will reveal the obscured version of the content item, or
applying a blur, wherein an intensity of the blur is inversely proportional to the likelihood that the user will reveal the obscured version of the content item; and
providing, by the computing system, the obscured version of the content item for display.
2. The computer-implemented method of claim 1 , further comprising:
providing, by the computing system, to the machine learning model, content item feature data, wherein the content item feature data comprises one or more of a sensitive content category or a sensitive content score.
3. The computer-implemented method of claim 1 , further comprising:
providing, by the computing system, to the machine learning model, the user feature data.
4. The computer-implemented method of claim 1 , wherein the generating the obscured version of the content item further comprises at least one of:
applying the color or applying the blur to a full version of the content item; or
applying the color or applying the blur to a thumbnail version of the content item.
5. The computer-implemented method of claim 1 , further comprising:
applying, by the computing system, a text overlay, wherein the text overlay provides a sensitive content warning.
6. The computer-implemented method of claim 1 , wherein the obscured version of the content item is provided via one or more of a profile, a feed, or a single content item display.
7. The computer-implemented method of claim 1 , further comprising:
receiving, by the computing system, an exposure indication; and
retraining, by the computing system, based on the exposure indication, the machine learning model.
8. The computer-implemented method of claim 7 , further comprising:
complying, by the computing system, with the exposure indication.
9. The computer-implemented method of claim 1 , further comprising:
providing, by a computing system, to a second machine learning model, a content item representation; and
receiving, by the computing system, from the second machine learning model, one or more of a sensitive content category or a sensitive content score.
10. The computer-implemented method of claim 9 , wherein the content item representation comprises one or more concepts.
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:
generating an obscured version of a content item, wherein the generating the obscured version of the content item comprises:
determining a likelihood that a user will reveal the obscured version of the content item based at least in part on i) user feature data associated with the user, wherein the user feature data includes at least one of:
demographic information or social networking system information, and ii) a machine learning model,
and wherein the generating the obscured version of the content item further comprises at least one of:
applying a color, wherein an opacity of the color is inversely proportional to the likelihood that the user will reveal the obscured version of the content item; or
applying a blur, wherein an intensity of the blur is inversely proportional to the likelihood that the user will reveal the obscured version of the content item; and
providing the obscured version of the content item for display.
12. The system of claim 11 , wherein the instructions, when executed by the at least one processor, further cause the system to perform:
providing to the machine learning model, content item feature data, wherein the content item feature data comprises one or more of a sensitive content category or a sensitive content score.
13. The system of claim 11 , wherein the instructions, when executed by the at least one processor, further cause the system to perform:
providing to the machine learning model, the user feature data.
14. The system of claim 11 , wherein the generating the obscured version of the content item further comprises at least one of:
applying the color or applying the blur to a full version of the content item; or
applying the color or applying the blur to a thumbnail version of the content item.
15. The system of claim 11 , wherein the instructions, when executed by the at least one processor, further cause the system to perform:
receiving an exposure indication; and
retraining based on the exposure indication, the machine learning model.
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:
generating an obscured version of a content item, wherein the generating the obscured version of the content item comprises:
determining a likelihood that a user will reveal the obscured version of the content item based at least in part on i) user feature data associated with the user, wherein the user feature data includes at least one of: demographic information or social networking system information, and ii) a machine learning model,
and wherein the generating the obscured version of the content item further comprises at least one of:
applying a color, wherein an opacity of the color is inversely proportional to the likelihood that the user will reveal the obscured version of the content item; or
applying a blur, wherein an intensity of the blur is inversely proportional to the likelihood that the user will reveal the obscured version of the content item; and
providing the obscured version of the content item for display.
17. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions, when executed by the at least one processor of the computing system, further cause the computing system to perform:
providing to the machine learning model, content item feature data, wherein the content item feature data comprises one or more of a sensitive content category or a sensitive content score.
18. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions, when executed by the at least one processor of the computing system, further cause the computing system to perform:
providing to the machine learning model, the user feature data.
19. The non-transitory computer-readable storage medium of claim 16 , wherein the generating the obscured version of the content item further comprises at least one of:
applying the color or applying the blur to a full version of the content item; or
applying the color or applying the blur to a thumbnail version of the content item.
20. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions, when executed by the at least one processor of the computing system, further cause the computing system to perform:
receiving an exposure indication; and
retraining based on the exposure indication, the machine learning model.