Recommender system for identifying a new set of media items responsive to an input set of media items and knowledge base metrics

• US 9,576,056 B2
• Filed: 11/12/2012
• Issued: 02/21/2017
• Est. Priority Date: 02/03/2005
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Martin, Francisco J., Shur, Jim, Torrens, Marc
• Primary Examiner(s): Fleurantin, Jean B
• Assistant Examiner(s): Sultana, Nargis
• Application Number:
  US13/674,815
• Publication Number:
  US 20130191318A1
• Time in Patent Office: 1,562 Days
• Field of Search: 707/913, 707/767, 707/736
• US Class Current: 1/1
• 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: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/4825 using a list of items to be played back in a given order, e.g. playlists
• CPC Class Codes: H04N 7/163 by receiver means only

Abstract

Systems and methods are disclosed for identifying a new set of media items in response to an input set (or “query set”) of media items and knowledge base metrics. The system uses a knowledge base consisting of a collection of mediasets. Various metrics among media items are considered by analyzing how the media items are grouped to form the mediasets in the knowledge base. Such association or “similarity” metrics are preferably stored in a matrix form that allows the system to efficiently identify a new set of media items that complements the input set of media items.

Claims

1. A method comprising:

storing a plurality of mediasets for multiple different users in a knowledge base of mediasets; and

preprocessing the knowledge base of mediasets to quantify an association between pairs of media items in the knowledge base, comprising;

for each item in a mediaset, identifying every other item in the same mediaset, so as to define all of the pairs of media items in that mediaset;

repeating said identifying step for every other mediaset in the knowledge base, thereby defining every pair of items that appears in any mediaset in the knowledge base; and

for each identified pair of media items, adjusting a co-concurrency metric value responsive to each additional occurrence of the same pair of media items in the knowledge base, said co-concurrency metric value comprising a co-concurrency metric value being defined as a number of times that multiple media items appear together in multiple users'"'"' mediasets;

receiving, from a client device, at least one input media item;

identifying, from the plurality of mediasets for the multiple different users in the knowledge base of mediasets, a plurality of ordered pairs of media items containing the input media item; and

selecting one or more of the plurality of ordered pairs of media items by determining that the co-concurrency metric value is adjusted above a predetermined threshold value.

2. The method of claim 1, wherein adjusting the co-concurrency metric value comprises incrementing the co-concurrency metric value responsive to every additional occurrence of the same pair of media items in the knowledge base of media sets.

3. The method of claim 1, wherein the media items are ordered within the mediasets for multiple different users in the knowledge base of mediasets;

wherein the identifying comprises defining ordered pairs of media items that appear in any mediaset in the knowledge base of mediasets in the same order; and

wherein the co-concurrency metric value is an ordered concurrency metric value, and adjusting the co-concurrency metric value comprises incrementing the ordered concurrency metric value responsive to every additional occurrence of the same ordered pair of media items in the knowledge base of mediasets.

4. The method of claim 1, further comprising:

sending a recommendation output set of media items to the client device, the recommendation output set comprising media items paired with the input media item and having a metric value above the predetermined threshold value.

5. The method of claim 4, further comprising:

receiving, along with the input media item, one or more recommendation preferences from the client device relating to what types of media items a user prefers to receive in the recommendation output set of media items; and

filtering the recommendation output set according to the one or more recommendation preferences.

6. The method of claim 4, further comprising:

filtering the recommendation output set based on metadata of the input media item.

7. The method of claim 4, further comprising:

storing media library collection data for a plurality of user accounts in relation to the knowledge base of mediasets;

upon receiving the input media item, accessing the user account associated with the client device; and

filtering the recommendation output set based on the user of the client device'"'"'s media library.

8. The method of claim 1, wherein each mediaset comprises a user-assembled media playlist including a subset of media items subjectively selected from multiple different users'"'"' collections of media items, wherein the knowledge base of mediasets comprises a collection of user-assembled media playlists, and wherein the co-concurrency metric value reflects the number of times that multiple media items have been subjectively selected to appear together in the user-assembled media playlists contained in the collection of user-assembled media playlists.

9. A system for preprocessing a knowledge base of mediasets for sending media recommendations comprising:

a database for storing a plurality of mediasets for multiple different users in a knowledge base of mediasets; and

a processor for preprocessing the knowledge base of mediasets to quantify an association between pairs of media items in the knowledge base of mediasets, wherein the processor is further configured to;

for each item in a mediaset, identify every other item in the same mediaset, so as to define all of the pairs of media items in that mediaset;

repeat said identifying step for every other mediaset in the knowledge base of mediasets, thereby defining every pair of items that appears in any mediaset in the knowledge base of mediasets; and

for each identified pair of media items, adjust a co-concurrency metric value responsive to each additional occurrence of the same pair of media items in the knowledge base of mediasets, said co-concurrency metric value being defined as a number of times that multiple media items appear together in multiple users'"'"' mediasets;

receive, from a client device, at least one input media item;

identify, from the plurality of mediasets for multiple different users in a knowledge base of mediasets, a plurality of ordered pairs of media items containing the input media item; and

select one or more of the ordered pairs of media items by determining that the co-concurrency metric value is adjusted above a predetermined threshold value.

10. The system of claim 9, wherein the processor is further configured increment the co-concurrency metric value responsive to every additional occurrence of the same pair of media items in the knowledge base of mediasets.

11. The system of claim 9, wherein the media items are ordered within the mediasets for multiple different users in the knowledge base of mediasets, wherein the processor is configured to define ordered pairs of media items that appear in any mediaset in the knowledge base of mediasets in the same order, wherein the co-concurrency metric value is an ordered concurrency metric value, and wherein the processor is further configured to increment the ordered concurrency metric value responsive to every additional occurrence of the same ordered pair of media items in the knowledge base of mediasets.

12. The system of claim 9, wherein the processor is further configured to:

send a recommendation output set of media items to the client device, the recommendation output set comprising media items paired with the input media item and having a metric value above the predetermined threshold value.

13. The system of claim 12, wherein the processor is further configured to:

receive, along with the input media item, one or more recommendation preferences from the client device relating to what types of media items a user prefers to receive in the recommendation output set of media items; and

filter the recommendation output set according to the one or more recommendation preferences.

14. The system of claim 12, wherein the processor is further configured to:

filter the recommendation output set based on metadata of the input media item.

15. The system of claim 12, wherein the processor is further configured to:

store media library collection data for a plurality of user accounts in relation to the knowledge base of mediasets;

upon receiving the input media item, access the user account associated with the client device; and

filter the recommendation output set based on the user of the client device'"'"'s media library.

16. The system of claim 9, wherein each mediaset comprises a user-assembled media playlist including a subset of media items subjectively selected from multiple different users'"'"' collections of media items, wherein the knowledge base of mediasets comprises a collection of user-assembled media playlists, and wherein the co-concurrency metric value reflects the number of times that multiple media items have been subjectively selected to appear together in the user-assembled media playlists contained in the collection of user-assembled media playlists.

17. A non-transitory computer-readable medium comprising:

a medium configured to store computer-readable instructions thereon; and

the computer-readable instructions that, when executed by a processing device cause the processing device to perform a method, comprising;

storing a plurality of mediasets in a knowledge base of mediasets of mediasets; and

preprocessing the knowledge base of mediasets to quantify an association between identified pairs of media items in the knowledge base of mediasets, comprising;

for each of the identified pairs of media items in the knowledge base of mediasets, identifying occurrences of each of the identified pairs of media items within the plurality of mediasets; and

for each identified pair of media items, calculating a co-concurrency metric value responsive to each occurrence of each of the identified pair of media items within the plurality of mediasets in the knowledge base of mediasets, said co-concurrency metric value being defined as a number of times that multiple media items appear together in multiple of the plurality of mediasets;

receiving, from a client device, at least one input media item;

identifying a plurality of ordered pairs of media items containing the input media item; and

selecting one or more of the ordered pairs of media items by determining that the co-currency metric value is calculated above a predetermined threshold value.

18. The non-transitory computer-readable storage medium of claim 17, wherein calculating the co-concurrency metric value comprises incrementing the co-concurrency metric value responsive to every additional occurrence of the same pair of media items within the plurality of mediasets in the knowledge base of mediasets.

19. The non-transitory computer-readable storage medium of claim 17, wherein the media items are ordered within the mediasets in the knowledge base, the instructions further comprising identifying ordered pairs of media items that appear in any mediaset in the knowledge base in the same order, and wherein the concurrency metric value is an ordered concurrency metric value, and incrementing the ordered co-concurrency metric value responsive to every additional occurrence of the same ordered pair of media items within the plurality of mediasets in the knowledge base of mediasets.

20. The non-transitory computer-readable storage medium of claim 17, the instructions further comprising:

sending a recommendation output set of media items to the client device, the recommendation output set comprising media items paired with the input media item and having a metric value above the predetermined threshold value.

21. The non-transitory computer-readable storage medium of claim 20, the instructions further comprising:

receiving, along with the input media item, one or more recommendation preferences from the client device relating to what types of media items a user prefers to receive in the recommendation output set of media items; and

filtering the recommendation output set according to the one or more recommendation preferences.

22. The non-transitory computer-readable storage medium of claim 20, the instructions further comprising:

filtering the recommendation output set based on metadata of the input media item.

23. The non-transitory computer-readable storage medium of claim 20, the instructions further comprising:

storing media library collection data for a plurality of user accounts in relation to the knowledge base of mediasets;

upon receiving the input media item, accessing the user account associated with the client device; and

filtering the recommendation output set based on the user of the client device'"'"'s media library.

24. The non-transitory computer-readable storage medium of claim 17, wherein each mediaset comprises a user-assembled media playlist including a subset of media items subjectively selected from multiple different users'"'"' collection of media items, wherein the knowledge base of mediasets comprises a collection of user-assembled media playlists, and wherein the co-concurrency metric value reflects the number of times that multiple media items have been subjectively selected to appear together in the user-assembled media playlists contained in the collection of user-assembled media playlists.

25. A method comprising:

storing a plurality of mediasets for multiple different users in a knowledge base of mediasets; and

preprocessing the knowledge base of mediasets to quantify an association between pairs of media items in the knowledge base of mediasets, comprising;

for each item in a mediaset, identifying every other item in the same mediaset, so as to define all of the pairs of media items in that mediaset;

repeating said identifying step for every other mediaset in the knowledge base of mediasets, thereby defining every pair of items that appears in any mediaset in the knowledge base of mediasets; and

for each identified pair of media items, adjusting a concurrency metric value responsive to each additional occurrence of the same pair of items in the knowledge base of mediasets, said concurrency metric value comprising a pre-concurrency metric value defined as a number of times a given item precedes an other item in the mediasets of the knowledge base of mediasets;

receiving, from a client device, at least one input media item;

identifying, from the plurality of mediasets for the multiple different users in the knowledge base of mediasets, a plurality of ordered pairs of media items containing the input media item; and

selecting one or more of the plurality of ordered pairs of items by determining that the concurrency metric value is adjusted above a predetermined threshold value.

26. A method comprising:

storing a plurality of mediasets from a community of multiple different users in a knowledge base of mediasets; and

preprocessing the knowledge base of mediasets to quantify an association between pairs of media items in the knowledge base of mediasets, comprising;

for each item in a mediaset, identifying every other item in the same mediaset, so as to define all of the pairs of media items in that mediaset;

repeating said identifying step for every other mediaset in the knowledge base of mediasets, thereby defining every pair of items that appears in any mediaset in the knowledge base of mediasets; and

for each identified pair of media items, adjusting a concurrency metric value responsive to each additional occurrence of the same pair of media items in the knowledge base of mediasets, said concurrency metric value comprising a post-concurrency metric value defined as a number of times a given item follows an other item in the mediasets of the knowledge base of mediasets;

receiving, from a client device, at least one input media item;

identifying, from the plurality of mediasets for the multiple different users in the knowledge base of mediasets, a plurality of ordered pairs of media items containing the input media item; and

selecting one or more of the plurality of ordered pairs of media items by determining that the concurrency metric value is adjusted above a predetermined threshold value.