Personalized recommendations based on item usage

• US 8,452,797 B1
• Filed: 03/09/2011
• Issued: 05/28/2013
• Est. Priority Date: 03/09/2011
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Paleja, Ameesh, Okereke, Mekka C., Jones, Matthew A.
• Primary Examiner(s): Alam, Shahid
• Assistant Examiner(s): Ly, Anh
• Application Number:
  US13/044,196
• Time in Patent Office: 811 Days
• Field of Search: 707/767, 707/776, 707/758, 707/780, 707/737, 707/751, 707/732, 707/E17.043, 707/E17.109, 707/E17.017, 707/E17.014, 707/805, 705/26.7, 705/347, 705/14.67, 705/26.5, 705/26.61, 705/26.63, 705/27.2, 705/29, 705/27
• US Class Current: 707/767
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

This disclosure describes systems and associated processes for generating recommendations for users based on usage, among other things. These systems and processes are described in the context of an interactive computing system that enables users to download applications for mobile devices (such as phones) or for other computing devices. Users'"'"' interactions with applications once they are downloaded can be observed and tracked, with such usage data being collected and provided to the interactive computing system. The interactive computing system can include a recommendation system or service that processes the usage data from a plurality of users to detect usage patterns. Using these usage patterns, among possibly other data, the recommendation system can recommend applications to users for download.

Claims

1. A method of recommending personalized computer applications to users, the method comprising:

receiving usage data collected from a computing device of a user, the usage data reflecting a degree of usage of user applications installed on the computing device of the user, the degree of usage reflecting a selected usage pattern that indicates how often the user applications were used;

identifying one or more related applications that are related to the user applications by at least comparing the usage data for the user to usage of the one or more related applications by other users, said identifying comprising identifying an application that was used with a similar usage pattern to the selected usage pattern;

weighting the user applications in the usage data based at least in part on the degree of usage of each application, such that the user applications used more frequently by the user are weighted more heavily than the user applications used less frequently by the user; and

selecting a subset of the related applications as recommendations for the user based at least in part on the weightings, thereby filtering the recommendations based on the degree of usage of the user applications;

wherein at least said identifying is implemented by one or more processors.

2. The method of claim 1, wherein the usage data reflects one or more of the following:

a selected frequency that the one or more user applications were used on the computing device; and

a selected amount of time that the one or more user applications were used on the computing device.

3. The method of claim 2, wherein said identifying the one or more related applications comprises identifying an application that was used a similar amount of time as the selected amount of time.

4. The method of claim 2, wherein said identifying the one or more related applications comprises identifying an application that was used at a similar frequency as the selected frequency.

5. The method of claim 1, wherein the usage data reflects a selected time period in which the one or more used applications were used.

6. The method of claim 5, wherein said identifying the one or more related applications comprises identifying an application that was used at a related time period to the selected time period.

7. The method of claim 1, wherein the usage data reflects a selected location in which the one or more used user applications were used.

8. The method of claim 7, wherein said identifying the one or more related applications comprises identifying an application that was used at a related location to the selected location.

9. The method of claim 1, wherein the usage data reflects a selected device other than the computing device of the user with which the one or more user applications were used.

10. The method of claim 9, wherein said identifying the one or more related applications comprises identifying an application that was used with a related device to the selected device.

11. Non-transitory physical storage comprising computer-executable instructions configured to implement a method of recommending personalized computer applications to users, the method comprising:

receiving usage data associated with a computing system of a user, the usage data reflecting a degree of usage of a plurality of user applications installed on the computing system;

identifying a subset of the user applications that have relatively higher usage by the user than others of the user applications, wherein said identifying the subset of user applications comprises identifying any of the user applications that have more than one usage by the user;

identifying, by one or more computing devices, one or more related applications related to the subset of user applications, wherein said identifying the one or more related applications comprises accessing the one or more related applications from a data set that relates applications based at least partly on an analysis of application selection behavior by a plurality of users; and

weighting the user applications in the usage data based at least in part on the degree of usage of each application, such that the user applications used more frequently by the user are weighted more heavily than the user applications used less frequently by the user;

selecting a subset of the related applications as recommendations for the user based at least in part on the weightings, thereby selecting the recommendations based on the degree of usage of the user applications.

12. The non-transitory physical storage of claim 11, wherein said identifying the one or more related applications comprises comparing the usage data for the subset of user applications to usage of the one or more related applications by other users.

13. The non-transitory physical storage of claim 12, wherein said identifying the one or more related applications comprises identifying usage metadata associated with the one or more related applications and comparing the usage metadata to the usage data for the subset of user applications.

14. A system for recommending personalized computer applications to users, the system comprising:

an application distribution service configured to provide applications for download by a plurality of user computing systems; and

a recommendation service comprising computer hardware, the recommendation service comprising a usage recommender configured to;

receive usage data associated with a selected computing system of a selected user, the usage data reflecting a degree of usage of a plurality of user applications installed on the computing system, the user applications being obtained from the application distribution service;

identify a subset of the user applications that have relatively higher usage by the user than others of the user applications;

identify, by one or more computing devices, one or more related applications related to the subset of user applications by at least comparing the usage data for the subset of user applications to usage of the one or more related applications by other users, said comparing comprising performing one or more of the following;

identifying a first application of the subset of user applications that was used at a similar frequency as a selected one of the one or more user applications and identifying a second application of the subset of user applications that was used at a related location as a selected one of the one or more user applications;

apply weights to the user applications in the usage data based at least in part on the degree of usage of each application, such that the user applications used more frequently by the user are weighted more heavily than the user applications used less frequently by the user; and

select a subset of the related applications as recommendations for the user based at least in part on the weights to thereby select the recommendations based on the degree of usage of the user applications.

15. The system of claim 14, wherein the usage recommender is further configured to identify the one or more related applications by at least identifying an application that was used a similar amount of time as a selected one of the one or more user applications.

16. The system of claim 14, wherein the usage recommender is further configured to identify the one or more related applications by at least identifying an application that was used at a related time period as a selected one of the one or more user applications.