Personalized application recommendations

Patent metadata

Field	Value
Publication number	US-10332184-B2
Application number	US-201414571110-A
Country	US
Kind code	B2
Filing date	Dec 15, 2014
Priority date	Dec 15, 2014
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

An example system includes a recommendation module. The recommendation module may receive a recommendation request including a set of installed application identifiers and identify a set of candidate application groups including at least one matching candidate application identifier. For each candidate application group, the recommendation module may determine a recommendation score based on a number of matching candidate application identifiers included in the candidate application group that match at least one installed application identifier. The recommendation module may select a first candidate application group based on the recommendation scores and select at least one non-matching candidate application identifier from the first candidate application group that does not match any of the installed application identifiers, resulting in a set of recommended application identifiers. The recommendation module may transmit to the user device, recommendation data for at least one recommended application identified by the set of recommended application identifiers.

First claim

The invention claimed is:

1. A method comprising:
maintaining, by a data processing device, a recommendation data store on a storage system, the recommendation data store storing a plurality of application group data records, each application group data record of the plurality of application group data records defining a group of applications represented by application identifiers of applications belonging to the group, each group of applications represented by the application identifiers in each application group data record of the plurality of application group data records being grouped based on one or more group generation rules, the group generation rules comprising instructions for inclusion of applications into application groups;
receiving, by the data processing device, a recommendation request from a user device via a network, the recommendation request including a first set of installed application identifiers and a second set of uninstalled application identifiers, the first set of installed application identifiers being determined based on installed application information collected by the user device according to a command of a user or according to a predetermined schedule, the second set of uninstalled application identifiers being determined based on uninstalled application information collected by the user device according to a command of the user or according to a predetermined schedule, the installed application information identifying a set of applications that are installed on the user device, and the uninstalled application information identifying a set of applications that have been uninstalled from the user device;
identifying, by the data processing device, a set of one or more application group data records from the recommendation data store based on the first set of installed application identifiers, each identified application group data record including at least one application identifier that matches to at least one installed application identifier included in the recommendation request;
for each identified application group data record:
determining, by the data processing device, a number of installed application identifiers from the first set that are also indicated in the identified application group data record,
setting, by the data processing device, a recommendation score of the identified application group data record based at least in part on the number of matching application identifiers, the recommendation score comprising a numerical value,
determining, by the data processing device, a subset of the uninstalled application identifiers in the second set that match one or more of the application identifiers in the application group data record, and
in response to the subset of the uninstalled application identifiers in the second set including one or more application identifiers that match one or more application identifiers in the application group data record, adjusting, by the data processing device, the recommendation score of the identified application group data record to negatively affect the recommendation score;

selecting, by the data processing device, a first application group data record from the set of identified application group data records based on the recommendation score of the first application group data record;
selecting, by the data processing device, at least one non-matching application identifier from the first application group data record for inclusion in a set of recommended application identifiers, wherein the selected non-matching application identifier does not match any of the installed application identifiers of the first set or uninstalled application identifiers of the second set; and
transmitting, by the data processing device, to the user device, recommendation data for at least one recommended application identified by the set of recommended application identifiers.

2. The method of claim 1 , wherein the selecting the first application group data record comprises:
comparing the recommendation score of the first application group data record to the recommendation score of at least a second application group data record of the set of identified application group data records; and
determining that the recommendation score of the first application group data record is greater than the recommendation score of at least the second application group data record.

3. The method of claim 1 ,
wherein determining the number of matching application identifiers in each identified application group data record comprises:
for each application identifier included in the application group data record that matches to at least one of the installed application identifiers received in the recommendation request:
determining a usage value of an application corresponding to the matching application identifier, the usage value representing the usage of the application on the user device,
determining whether the usage value for the corresponding application meets or exceeds a predetermined usage threshold, and
in response to the usage value for the corresponding application meeting or exceeding the predetermined usage threshold, incrementing the number of matching application identifiers, and

wherein, in response to the usage value for the corresponding application being less than the predetermined usage threshold, the matching application identifier is not counted towards the number of matching application identifiers.

4. The method of claim 1 , wherein the set of recommended application identifiers does not include non-matching application identifiers corresponding to applications that have been uninstalled from the user device.

5. The method of claim 1 , wherein selecting the first application group data record comprises:
determining whether a geographic location of the user device is within a predetermined geographic range associated with a candidate application group data record; and
selecting the candidate application group data record as the first application group data record in response to the geographic location of the user device being within the predetermined geographic range.

6. The method of claim 5 , wherein selecting the first application group data record comprises preventing selection of the candidate application group data record as the first application group data record in response to the geographic location of the user device being outside the predetermined geographic range.

7. The method of claim 1 , wherein determining the recommendation score for an application data record group comprises:
determining an application popularity value of at least one application corresponding to a matching application identifier included in the application group data record of a candidate application group; and
determining the recommendation score for the application group data record based on the application popularity value.

8. The method of claim 7 , wherein the recommendation score for the candidate application group is affected positively in response to the application popularity value being below a predetermined threshold popularity value.

9. The method of claim 7 , wherein the recommendation score for the candidate application group is affected negatively in response to the application popularity value being above a predetermined threshold popularity value.

10. A system comprising:
a storage system that stores a recommendation data store, the recommendation data store storing a plurality of application group data records, each application group data record of the plurality of application group records defining a group of applications represented by application identifiers of applications belonging to the group, each group of applicat

Assignees

• Samsung Electronics Co Ltd

Inventors

• Glover Eric

Classifications

• G06Q30/0631 Primary

  Recommending goods or services · CPC title