Global recommendation systems for overlapping media catalogs

• US 10,552,470 B2
• Filed: 03/24/2016
• Issued: 02/04/2020
• Est. Priority Date: 04/06/2015
• Status: Active Grant

• Current Assignee: Netflix, Inc.
• Original Assignee: Netflix, Inc.
• Inventors: Todd, Michael, Elahi, Ehtsham, Fenton, Aish, Uribe, Carlos Gomez
• Primary Examiner(s): Mofiz, Apu M
• Assistant Examiner(s): Samara, Husam Turki
• Application Number:
  US15/079,767
• Publication Number:
  US 20160292158A1
• Time in Patent Office: 1,412 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: H04L 65/60 Network streaming of media packets
• CPC Class Codes: H04L 65/612 for unicast
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

A computer-implemented method and a system for implementation the method are provided. The method includes receiving user activity data for a first set of users of a streaming media system, in which the first set of users having access to a first catalog. Additional user activity data for a second set of users is received. The second set of users has access to a second catalog, which includes media items also present in the first catalog. The second catalog further includes an exclusive media item that is not included in the first catalog. A probability that a user of the first set of users would play the exclusive media item if presented by the streaming media system with the exclusive media item is determined. Information is provided to a user of the second set of users as a recommendation to play the exclusive media item based on the probability.

Claims

1. A computer-implemented method, the method comprising:

receiving first user activity data for a first set of users of a streaming media system, the first set of users having access to a first media item catalog;

receiving second user activity data for a second set of users of the streaming media system, the second set of users having access to a second media item catalog and not having access to one or more media items in the first media item catalog, wherein the second media item catalog includes overlapping media items, the overlapping media items being included in both the first and second media item catalogs, the second media item catalog further includes an exclusive media item that is not included in the first media item catalog;

determining, by a processing device and based on the first and second user activity data, a probability that a user of the first set of users would play the exclusive media item if recommended by the streaming media system, wherein determining the probability of the user playing the exclusive media item includes;

compensating, in the determination, for one or more biasing effects introduced by recommending the exclusive media item, the compensating being based on take-rate information included in the first and second user activity data indicating which users selected the exclusive media item; and

compensating, in the determination, for one or more biasing effects introduced by size differences in overlapping but non-identical media item catalogs in different geographical jurisdictions by analyzing the take-rate information to determine whether the number of times the exclusive media item was selected is proportionate between the different geographical jurisdictions; and

providing information to be rendered in a user interface by a user device as a recommendation to a user of the second set of users to play the exclusive media item based on the probability.

2. The method of claim 1, further comprising determining an inter-catalog ranking of media items, the inter-catalog ranking comprising a ranking of media items included in the first and second media item catalogs.

3. The method of claim 1, further comprising:

identifying a user cluster with which the user of the second set of users is associated;

determining popular media items from that first and second media item catalogs, the popular media items having a ranking above a threshold value among users in the user cluster, the popular media items comprising the exclusive media item; and

determining that the exclusive media item is accessible to the user of the second set of users.

4. The method of claim 1, wherein the media items in the first and second media item catalogs are ranked according to how often users of the streaming media system play the media items when presented with the media items.

5. The method of claim 1, wherein the first and second user activity data are consolidated in a consolidated user activity data set, and wherein the consolidated user activity data set comprises, for each media item and each user, an indication of whether the media item has been presented to the user and an indication of whether the media item has been selected by the user.

6. The method of claim 5, wherein the media item being presented to the user comprises recommending the media item to a user through the user interface.

7. The method of claim 1, further comprising generating artificial user activity data based on the probability that the user of the first set of users would play the exclusive media item if presented with the exclusive media item, and wherein the information to be rendered as a recommendation to the user of the second set of users is based on the first and second user activity data and the artificial user activity data.

8. The method of claim 1, wherein the first set of users reside in a first legal jurisdiction and the second set of users reside in a second legal jurisdiction and wherein the first catalog is available in the first legal jurisdiction and the second catalog is available in the second legal jurisdiction.

9. A system for providing streaming media items to users, the system comprising:

a data storage device, the data storage device storing;

first user activity data for a first set of users of a streaming media system, the first set of users having access to a first media item catalog; and

second user activity data for a second set of users of the streaming media system, the second set of users having access to a second media item catalog and not having access to one or more media items in the first media item catalog, wherein the second media item catalog includes overlapping media items that are included in both the first and second media item catalogs, the second media item catalog further includes an exclusive media item that is not included in the first media item catalog; and

a processing device in communication with the data storage device to access information about contents of the first and second media item catalogs and the first and second user activity data, wherein the processing device;

determines, from the first and second user activity data, a probability that a user of the first set of users would play the exclusive media item if recommended by the streaming media system, wherein determining the probability of the user playing the exclusive media item includes;

compensating, in the determination, for one or more biasing effects introduced by recommending the exclusive media item, the compensating being based on take-rate information included in the first and second user activity data indicating which users selected the exclusive media item; and

compensating, in the determination, for one or more biasing effects introduced by size differences in overlapping but non-identical media item catalogs in different geographical jurisdictions by analyzing the take-rate information to determine whether the number of times the exclusive media item was selected is proportionate between the different geographical jurisdictions; and

provides information to be rendered in a user interface by a user device as a recommendation to a user of the second set of users to play the exclusive media item based on the probability.

10. The system of claim 9, wherein the first and second user activity data stored in the data storage is combined in a consolidated user activity data set, and wherein the consolidated user activity data set comprises, for each media item and each user, an indication of whether the media item has been presented to the user and an indication of whether the media item has been played by the user.

11. The system of claim 9, wherein the processing device determines an inter-catalog ranking of media items, the inter-catalog ranking comprising a ranking of media items included in the first and second media item catalogs.

12. The system of claim 11, wherein the user of the first set of users and the user of the second set of users are both associated with a first cluster of users, and wherein first set of users is additionally associated with a second cluster of users, such that the first user can be defined by association with the first cluster and second cluster.

13. The system of claim 12, wherein the probability associated with each media item is a Bernoulli random variable.

14. The system of claim 9, wherein the first and second media item catalogs are not separated by a pay-wall.

15. A system for providing streaming media items to users, the system comprising:

a data storage device storing, the data storage device storing;

user activity data for a set of users of a streaming media system, the user activity data describing interactions of the users with one or more of a set of media items in a media item catalog;

user access data indicating, for each user of the set of users, a subset of media items to which the user has access, wherein a first user of the set of users has access to a first media item to which a second user of the set of users does not have access; and

a processing device in communication with the data storage device to access information about the set of media items, the user activity data, and the user access data wherein the processing device;

determines, from the user activity data, a probability that the second user would play the first media item if recommended by the streaming media system, wherein determining the probability of the user playing the exclusive media item includes;

compensating, in the determination, for one or more biasing effects introduced by recommending the exclusive media item, the compensating being based on take-rate information included in the first and second user activity data indicating which users selected the exclusive media item; and

compensating, in the determination, for one or more biasing effects introduced by size differences in overlapping but non-identical media item catalogs in different geographical jurisdictions by analyzing the take-rate information to determine whether the number of times the exclusive media item was selected is proportionate between the different geographical jurisdictions; and

provides information to be rendered in a user interface by a user device as a recommendation to the first user to play the first media item based on the probability.

16. The system of claim 15, wherein the user access data comprises an indication of which media items in the media item catalog each user of the set of users is able to access.

17. The system of claim 15, further comprising artificial user activity data, wherein the artificial user activity data comprises a generated play record based on the probability that the second user would play the first media item if presented with the first media item.

18. The system of claim 15, wherein a subset of the media items in the media catalog are associated with a cluster of users that includes both the first user and the second user based on user activity data of the first user and user activity data of the second user.

19. The system of claim 18, wherein the recommendation to the first user is further based on an association of the first user with the cluster of users.

20. The system of claim 18, wherein media items are associated with the cluster based on the user activity data of the first user and the second user and based on artificial user activity data that comprises a generated play record based on the probability that the second user would play the first media item if presented with the first media item.