Media library analyzer

• US 9,383,965 B1
• Filed: 08/16/2013
• Issued: 07/05/2016
• Est. Priority Date: 08/16/2013
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Qureshi, Aneeb Naveed
• Primary Examiner(s): Tank, Andrew
• Application Number:
  US13/968,540
• Time in Patent Office: 1,054 Days
• Field of Search: 707/738, 707/749, 715/716
• US Class Current: 1/1
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 3/04847 Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G11B 27/105 of operating discs
• CPC Class Codes: G11B 27/329 on a disc [VTOC]
• CPC Class Codes: G11B 27/36 Monitoring, i.e. supervising the progress of recording or reproducing for digital recording G11B20/00 and s.gr.; for monitoring, testing or measuring of TV recorders of the type covered by H04N5/76 and subgroups, see H04N17/06

Abstract

Disclosed are various embodiments analyzing a user'"'"'s interaction with his or her music library. The system generates a time series by tracking a plurality of instances of music library interaction between a user and a music library. The system also determines a distribution expressed in the time series, the distribution indicating a trend of playing a set of audio items for a particular period of time, the set of audio items being associated with a class, wherein a timestamp corresponds to an occurrence of the trend. The system associates the distribution with a triggering event and generates a recommendation according to the class in response to the triggering event.

Claims

1. A non-transitory computer-readable medium embodying a program executable in at least one computing device, the program comprising program code that, when executed by the at least one computing device, causes the at least one computing device to:

generate a time series comprising a plurality of instances of an interaction with a music library, wherein individual ones of the plurality of instances are defined in the time series by a period of time;

identify a transition from a first playback trend associated with a first class of music to a second playback trend associated with a second class of music by;

determining a first playback frequency distribution expressed in the time series, the first distribution indicating an affinity towards a first set of audio items associated with a first class of music, the first playback frequency distribution being determined based on a width or a height of the first playback frequency distribution expressed in the time series meeting a predefined threshold indicative of the first playback frequency distribution being statistically significant;

determining a second playback frequency distribution expressed in the time series, the second distribution indicating an affinity towards a second set of audio items associated with the second class of music, the second playback frequency distribution being determined based on a width or a height of the second playback frequency distribution expressed in the time series meeting a predefined threshold indicative of the second playback frequency distribution being statistically significant; and

generating a first timestamp for the first playback frequency distribution for a historical point in time when the affinity towards the first set of audio items associated with the first class of music was expressed;

generating a second timestamp for the second playback frequency distribution for a historical point in time when the affinity towards the second set of audio items associated with the second class of music was expressed;

identifying the transition from the first playback trend to the second playback trend based on a relationship between the first timestamp and the second timestamp;

identify that an audio item associated with the first class of music has been recently played; and

in response to the audio item associated with the first class having been recently played, identify a recommended audio item associated with the second class of music based on the recently played audio item and the relationship.

2. The non-transitory computer-readable medium of claim 1, wherein the first class of music comprises a musical artist, a album, a music genre, or a music tempo.

3. The non-transitory computer-readable medium of claim 1, wherein the first timestamp and the second timestamp are generated as:

a time at which a user began listening to a music item, a time at which the user stopped listening to the music item, a peak time during playback of the music item, a standard deviation, a p-value, or a t-value.

4. A system, comprising:

at least one computing device; and

program instructions executable in the at least one computing device that, when executed, cause the at least one computing device to;

generate a time series by tracking a plurality of instances of media library interaction between a user and a media library;

identify a transition from a first trend associated with a first class to a second trend associated with a second class of by;

determining a first distribution expressed in the time series indicating an affinity towards a first set of media items associated with a first class, the first distribution being defined for a particular period of time and being associated with a first timestamp, the first distribution being determined based on a width or a height of the first distribution expressed in the time series meeting a predefined threshold indicative of the first distribution being statistically significant; and

determining a second distribution expressed in the time series indicating an affinity towards a second set of media items associated with a second class, the second distribution being defined for a particular period of time, the second distribution being associated with a second timestamp, the second distribution being determined based on a width or a height of the second distribution expressed in the time series meeting a predefined threshold indicative of the second distribution being statistically significant; and

identify a recommended media item from the second set of media items associated with the second class based on a comparison of the first timestamp with the second timestamp.

5. The system of claim 4, further comprising program instructions that, when executed, cause the least one computing device to display an identifier of the recommended media item at a predetermined period of time following the first timestamp.

6. The system of claim 4, wherein the first distribution and the second distribution are determined according to a user input that specifies a portion of the time series.

7. The system of claim 4, wherein an instance of media library interaction among the plurality of instances of media library interaction comprises at least one of a skip command, a mute command, an add command, or a remove command.

8. The system of claim 4, further comprising program instructions that, when executed, cause the at least one computing device to periodically update the time series according to a predetermined schedule.

9. The system of claim 4, further comprising program instructions that, when executed, cause the at least one computing device to associate the first timestamp with a triggering event.

10. The system of claim 9, wherein the triggering event comprises the second timestamp associated with the second distribution.

11. The system of claim 9, wherein the recommended media item is included in the media library.

12. The system of claim 4, wherein the recommended media item is identified in response to a media item in the first set of media items associated with the first class having been recently played.

13. A method, comprising:

generating, by at least one computing device comprising at least one hardware processor, a time series by tracking a plurality of instances of music library interaction between a user and a music library;

determining, by the at least one computing device, a first distribution expressed in the time series, the first distribution indicating a first trend of playing a first set of audio items for a first period of time, the first set of audio items being associated with a first class, wherein a first timestamp corresponds to a peak time corresponding to a peak height in the first distribution;

determining, by the at least one computing device, a second distribution expressed in the time series, the second distribution indicating a second trend of playing a second set of audio items for a second period of time, the second set of audio items being associated with a second class, wherein a second timestamp corresponds to a peak time corresponding to a peak height in the second distribution;

associating, by the at least one computing device, the first distribution with a triggering event based on a comparison of the peak time in the first distribution with the peak time in the second distribution; and

generating, by the at least one computing device, a recommendation comprising a recommended audio item associated with the second class in response to a subsequent occurrence of the triggering event.

14. The method of claim 13, wherein the triggering event comprises a playing of an audio item in the first set of audio items.

15. The method of claim 13, wherein the first class or the second class comprise a musical artist, an album, or a music genre.

16. The method of claim 13, wherein the first distribution or the second distribution is determined according to a user input that specifies a duration of time of the time series.

17. The method of claim 13, wherein an instance of music library interaction among the plurality of instances of music library interaction comprises a play command.

18. The method of claim 13, wherein the recommendation comprises an item offered over an electronic commerce system.

19. The method of claim 13, wherein:

the peak time in the first distribution is identified between a time at which the user began the playing of the first set of audio items and a time at which the user stopped the playing of the first set of audio items; and

the peak time in the second distribution is identified between a time at which the user began the playing of the second set of audio items and a time at which the user stopped the playing of the second set of audio items.

20. The method of claim 13, further comprising periodically updating the time series according to a predetermined schedule.