Detecting digital content misuse based on digital content usage clusters

1. A method comprising:

gathering, by a digital content delivery system, usage signal data describing digital content usage by a plurality of user accounts of the digital content delivery system;

generating, based on the usage signal data, a set of data points, each data point from the set of data points corresponding to a unique user account of the plurality of user accounts and having been generated based on usage signal data describing digital content usage associated with the respective unique user account corresponding to the respective data point;

clustering, based on a clustering algorithm, the set of data points into a set of digital content usage clusters, each digital content usage cluster from the set of digital content clusters having a determined center;

determining, based on the set of digital content usage clusters, a first distance between a first data point from the set of data points and the respective center of a digital content cluster that is closest to the first data point;

determining that the first distance is greater than a threshold distance; and

in response to determining that the first distance is greater than the threshold distance, placing a restriction on the respective user account corresponding to the first data point, the restriction disabling the respective user account corresponding to the first data point from accessing digital content from the digital content delivery system.

2. The method of claim 1 , wherein the usage signal data describing digital content usage by the plurality of user accounts includes one or more of a number of times digital content was accessed, a frequency at which digital content was accessed, digital content items that were accessed, a geographic location of a client-side computing device associated with the unique user account, a number of known family members residing in a dwelling associated with the unique user account, and a number of mobile computing devices detected during performance of digital content.

3. The method of claim 1 , further comprising:

determining, from the set of digital content usage clusters, a subset of digital content usage clusters that meet a quality threshold level, wherein the digital content usage cluster that is closest to the first data point is selected from the subset of digital content usage clusters.

4. The method of claim 3 , wherein determining the subset of digital content usage clusters that meet a quality threshold level comprises:

determining a density of data point in a first digital content usage cluster; and

comparing the density of data point in the first digital content usage cluster to a threshold density level.

5. The method of claim 1 , wherein the clustering algorithm is a k-means clustering algorithm.

6. The method of claim 1 , further comprising:

scanning social media activity of a user associated with the respective user account corresponding to the first data point for postings made by the user that indicate that the user is allowing unauthorized access to the digital content.

7. The method of claim 1 , further comprising:

causing a client-side computing device associated with the respective user account corresponding to the first data point to suspend performance of digital content.

8. A digital content delivery system comprising:

one or more computer processors; and

one or more computer-readable mediums storing instructions that, when executed by the one or more computer processors, causes the digital content delivery system to perform operations comprising:

gathering usage signal data describing digital content usage by a plurality of user accounts of the digital content delivery system;

generating, based on the usage signal data, a set of data points, each data point from the set of data points corresponding to a unique user account of the plurality of user accounts and having been generated based on usage signal data describing digital content usage associated with the respective unique user account corresponding to the respective data point;

clustering, based on a clustering algorithm, the set of data points into a set of digital content usage clusters, each digital content usage cluster from the set of digital content clusters having a determined center;

determining, based on the set of digital content usage clusters, a first distance between a first data point from the set of data points and the respective center of a digital content duster that is closest to the first data point;

determining that the first distance is greater than a threshold distance; and

in response to determining that the first distance is greater than the threshold distance, placing a restriction on the respective user account corresponding to the first data point, the restriction disabling the respective user account corresponding to the first data point from accessing digital content from the digital content delivery system.

9. The digital content delivery system of claim 8 , wherein the usage signal data describing digital content usage by the plurality of user accounts includes one or more of a number of times digital content was accessed, a frequency at which digital content was accessed, digital content items that were accessed, a geographic location of a client-side computing device associated with the unique user account, a number of known family members residing in a dwelling associated with the unique user account, and a number of mobile computing devices detected during performance of digital content.

10. The digital content delivery system of claim 8 , the operations further comprising:

determining, from the set of digital content usage clusters, a subset of digital content usage clusters that meet a quality threshold level, wherein the digital content usage cluster that is closest to the first data point is selected from the subset of digital content usage clusters.

11. The digital content delivery system of claim 10 , wherein determining the subset of digital content usage clusters that meet a quality threshold level comprises:

determining a density of data point in a first digital content usage cluster; and

comparing the density of data point in the first digital content usage cluster to a threshold density level.

12. The digital content delivery system of claim 8 , wherein the clustering algorithm is a k-means clustering algorithm.

13. The digital content delivery system of claim 8 , the operations further comprising:

scanning social media activity of a user associated with the respective user account corresponding to the first data point for postings made by the user that indicate that the user is allowing unauthorized access to the digital content.

14. The digital content delivery system of claim 8 , the operations further comprising:

causing a client-side computing device associated with the respective user account corresponding to the first data point to suspend performance of digital content.

15. A non-transitory computer-readable medium storing instructions that, when executed by the one or more computer processors of a digital content delivery system, causes the digital content delivery system to perform operations comprising:

gathering usage signal data describing digital content usage by a plurality of user accounts of the digital content delivery system;

generating, based on the usage signal data, a set of data points, each data point from the set of data points corresponding to a unique user account of the plurality of user accounts and having been generated based on usage signal data describing digital content usage associated with the respective unique user account corresponding to the respective data point;

clustering, based on a clustering algorithm, the set of data points into a set of digital content usage clusters, each digital content usage cluster from the set of digital content clusters having a determined center;

determining, based on the set of digital content usage clusters, a first distance between a first data point from the set of data points and the respective center of a digital content cluster that is closest to the first data point;

determining that the first distance is greater than a threshold distance; and

in response to determining that the first distance is greater than the threshold distance, placing a restriction on the respective user account corresponding to the first data point, the restriction disabling the respective user account corresponding to the first data point from accessing digital content from the digital content delivery system.

16. The non-transitory computer-readable medium of claim 15 , wherein the usage signal data describing digital content usage by the plurality of user accounts includes one or more of a number of times digital content was accessed, a frequency at which digital content was accessed, digital content items that were accessed, a geographic location of a client-side computing device associated with the unique user account, a number of known family members residing in a dwelling associated with the unique user account, and a number of mobile computing devices detected during performance of digital content.

17. The non-transitory computer-readable medium of claim 15 , the operations further comprising:

determining, from the set of digital content usage clusters, a subset of digital content usage clusters that meet a quality threshold level, wherein the digital content usage cluster that is closest to the first data point is selected from the subset of digital content usage clusters.

18. The non-transitory computer-readable medium of claim 17 , wherein determining the subset of digital content usage clusters that meet a quality threshold level comprises:

determining a density of data point in a first digital content usage cluster; and

comparing the density of data point in the first digital content usage cluster to a threshold density level.

19. The non-transitory computer-readable medium of claim 15 , wherein the clustering algorithm is a k-means clustering algorithm.

20. The non-transitory computer-readable medium of claim 15 , the operations further comprising:

scanning social media activity of a user associated with the respective user account corresponding to the first data point for postings made by the user that indicate that the user is allowing unauthorized access to the digital content.