Methods and systems for session clustering based on user experience, behavior, and interactions

Patent metadata

Field	Value
Publication number	US-10334073-B2
Application number	US-201715682391-A
Country	US
Kind code	B2
Filing date	Aug 21, 2017
Priority date	Feb 24, 2017
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

A server system divides respective groups of a plurality of groups of sessions for multiple users of the media-providing service into respective pluralities of clusters. The server system tracks user interaction with a client device during a user session and identifies the user session as belonging to a first cluster of the pluralities of clusters based at least in part on the user interaction. The server system personalizes content for the user session using one or more content criteria associated with the first cluster, in accordance with identifying that the user session belongs to the first cluster.

First claim

What is claimed is:

1. A method, comprising
at a server system of a media-providing service having one or more processors and memory storing instructions for execution by the one or more processors:
for a plurality of groups of sessions for multiple users of the media-providing service, dividing respective groups into respective pluralities of clusters;
tracking user interaction with a client device during a user session;
identifying the user session as belonging to a first cluster of the pluralities of clusters based at least in part on the user interaction; and
in accordance with identifying that the user session belongs to the first cluster, personalizing content for the user session using one or more content criteria associated with the first cluster.

2. The method of claim 1 , wherein:
the method further comprises, at the server system, sorting a plurality of sessions for the multiple users of the media-providing service into the plurality of groups of sessions, the sorting comprising applying one or more sorting rules to one or more session characteristics for the plurality of sessions.

3. The method of claim 2 , wherein the one or more session characteristics include whether a shuffle mode of the media-providing service is enabled.

4. The method of claim 2 , wherein the one or more session characteristics include an account type.

5. The method of claim 2 , wherein the one or more session characteristics include whether content is skipped during a session.

6. The method of claim 2 , wherein the one or more session characteristics include a mode of playback.

7. The method of claim 1 , further comprising, at the server system, identifying, for each cluster in a respective plurality of clusters, a corresponding session characteristic that distinguishes the cluster in the respective plurality of clusters;
wherein identifying the user session as belonging to the first cluster comprises determining that the user interaction matches the corresponding session characteristic that distinguishes the first cluster.

8. The method of claim 7 , further comprising, at the server system, sorting a plurality of sessions for the multiple users of the media-providing service into the plurality of groups of sessions, the sorting comprising applying one or more sorting rules to one or more session characteristics for the plurality of sessions;
wherein the corresponding session characteristic that distinguishes the first cluster is distinct from the one or more session characteristics to which the one or more sorting rules are applied.

9. The method of claim 7 , further comprising, at the server system, sorting a plurality of sessions for the multiple users of the media-providing service into the plurality of groups of sessions, the sorting comprising applying one or more sorting rules to one or more session characteristics for the plurality of sessions;
wherein the corresponding session characteristic that distinguishes the first cluster is a session characteristic of the one or more session characteristics to which the one or more sorting rules are applied.

10. The method of claim 1 , further comprising, at the server system, assigning a first label of a plurality of labels to one or more clusters of the pluralities of clusters, including assigning the first label to the first cluster;
wherein the one or more content criteria associated with the first cluster correspond to the first label.

11. The method of claim 1 , wherein dividing the respective groups into the respective pluralities of clusters is performed using a machine-learning algorithm.

12. The method of claim 11 , wherein the machine-learning algorithm comprises a K-means algorithm.

13. The method of claim 11 , wherein dividing the respective groups using the machine-learning algorithm comprises dividing each group of the plurality of groups into a predefined number of clusters.

14. The method of claim 11 , wherein the machine-learning algorithm comprises a Gaussian Mixture Model.

15. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between providing an advertisement to the client device or forgoing providing the advertisement to the client device.

16. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between providing a tooltip to the client device or forgoing providing the tooltip to the client device.

17. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between providing a notification to the client device or forgoing providing the notification to the client device.

18. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between automatically providing a next media content to the client device or forgoing providing the next media content to the client device.

19. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between enabling voice search or forgoing enabling voice search.

20. The method of claim 1 , wherein personalizing the content for the user session comprises selecting between providing a recommendation for a session type to the client device or forgoing providing the recommendation for the session type to the client device.

21. The method of claim 1 , wherein the user session is a session for a specific user, the method further comprising, at the server system, determining a pattern of use of the media-providing service for the specific user,
wherein identifying the user session as belonging to a first cluster is based at least in part on the pattern.

22. A server system, comprising:
one or more processors; and
memory storing one or more programs for execution by the one or more processors, the one or more programs including instructions for:
for a plurality of groups of sessions for multiple users of the media-providing service, dividing respective groups into respective pluralities of clusters;
tracking user interaction with a client device during a user session;
identifying the user session as belonging to a first cluster of the pluralities of clusters based at least in part on the user interaction; and
in accordance with identifying that the user session belongs to the first cluster, personalizing content for the user session using one or more content criteria associated with the first cluster.

23. A non-transitory computer-readable storage medium, storing one or more programs for execution by one or more processors of a server system, the one or more programs including instructions for:
for a plurality of groups of sessions for multiple users of the media-providing service, dividing respective groups into respective pluralities of clusters;
tracking user interaction with a client device during a user session;
identifying the user session as belonging to a first cluster of the pluralities of clusters based at least in part on the user interaction; and
in accordance with identifying that the user session belongs to the first cluster, personalizing content for the user session using one or more content criteria associated with the first cluster.

Assignees

• Spotify Ab

Inventors

• Gibson Clay
• Shapiro Will
• Gil Santiago
• Anderson Ian
• Mpossi Margreth
• Semerci Oguz
• Wolf Scott

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• H04L67/327 Primary

  Electricity · mapped topic

• H04L67/22

  Electricity · mapped topic

• H04L67/142

  Managing session states for stateless protocols; Signalling session states; State transitions; Keeping-state mechanisms · CPC title

• H04L67/63 Primary

  Routing a service request depending on the request content or context · CPC title