System and method to recommend a bundle of items based on item/user tagging and co-install graph

Abstract

A system and method of recommending a bundle of content items to a user, including storing a plurality of content items in a computer system, determining a respective co-selection score for each pair of content items among the plurality of content items, the co-selection score indicating a probability that a given pair of content items among the plurality of content items will both be downloaded by a user of the computer system, and outputting, to a first user, a plurality of content items comprising a sub-set of the plurality of content items.

1 . A computer-implemented method, comprising:
 storing a plurality of content items in a computer system;   determining a respective co-selection score for each pair of content items among the plurality of content items, the co-selection score indicating a probability that a given pair of content items among the plurality of content items will both be downloaded by a user of the computer system; and   outputting, to a first user, a plurality of content items comprising a sub-set of the plurality of content items, the sub-set being selected to correspond to a graph having vertices each connected by unique edges, wherein each vertex corresponds to a content item and each unique edge corresponds to a co-selection score greater than a predetermined threshold value.   
     
     
         2 . The method of  claim 1 , further comprising:
 determining respective quality scores for each of the plurality of content items; and   selecting the sub-set to have a graph quality score which is greater than a predetermined threshold value, the graph quality score being based on an average quality score of the sub-set content items and an average co-install score of the sub-set content item pairs.   
     
     
         3 . The method of  claim 2 , wherein the sub-set is further selected to have a graph quality score that is an approximate maximum possible value for a predetermined number of content items. 
     
     
         4 . The method of  claim 2 , wherein the content items correspond to installable applications which may be downloaded by a plurality of users, and wherein the quality scores are determined based on at least one factor selected from the group consisting of: a number of times the application has been installed by a plurality of users, a number of times the application has been uninstalled by the plurality of users, an average user rating of the application, a number of users that have rated the application, and a rating of a developer who created the application. 
     
     
         5 . The method of  claim 1 , further comprising:
 storing installation data corresponding to statistics of content item installations made by a plurality of users of the computer system,   wherein the co-selection scores are further determined based on the installation data.   
     
     
         6 . The method of  claim 5 , wherein the co-selection scores for each pair (A,B) of content items are further determined based on mutual information as follows:
     s   (A,B) =Σ a∈{A,!A} Σ b∈{B,!B}   P ( a,b )log( P ( a, b )/ P ( a ) P ( b )))
   where A means a first content item has been installed by a user, !A means the first content item has not been installed by a user, B means a second content item has been installed by a user, !B means the second content item has not been installed by a user, and P(•) are probabilities approximated based on the installation data.   
     
     
         7 . The method of  claim 1 , further comprising:
 associating one or more data tags with each of the plurality of content items,   wherein the sub-set is further selected such that each of the sub-set content items are associated with one or more data tags of a predetermined set of one or more data tags.   
     
     
         8 . The method of  claim 7 , further comprising:
 associating one or more data tags with user data for each of a plurality of users of the computer system,   wherein the predetermined set of one or more data tags comprises the one or more data tags which are associated with a given user.   
     
     
         9 . The method of  claim 8 , wherein the output is provided to the user when there is a change in the data tags associated with the given user's user data. 
     
     
         10 . The method of  claim 1 , further comprising:
 storing first data tags associated with demographic user data for each of a plurality of users of the computer system,   wherein the sub-set is further selected such that each sub-set content item is associated with a same first data tag.   
     
     
         11 . The method of  claim 1 , wherein the output is provided to the user when the user selects one of the content items through an interface of the computer system. 
     
     
         12 . A system, comprising:
 a storage device; a memory that stores computer executable components; and   a processor that executes the following computer executable components stored in the memory:   a storage component that stores a plurality of content items in the storage device;   a calculating component that determines a respective co-selection score for each pair of content items among the stored plurality of content items, the co-selection score indicating a probability that a given pair of content items among the plurality of content items will both be downloaded by a user of the system;   an output component that outputs, to a first user, a plurality of content items comprising a sub-set of the stored plurality of content items; and   a processing component that selects the sub-set such that the sub-set corresponds to a graph having vertices each connected by unique edges, wherein each vertex corresponds to a content item and each unique edge corresponds to a co-selection score greater than a predetermined threshold value.   
     
     
         13 . The system of  claim 12 , wherein the calculating component further determines respective quality scores for each of the plurality of content items, and
 wherein the processing component further selects the sub-set to have a graph quality score which is greater than a predetermined threshold value, the graph quality score being based on an average quality score of the sub-set content items and an average co-selection score of the sub-set content item pairs.   
     
     
         14 . The system of  claim 13 , wherein the processing component further selects the sub-set to have a graph quality score that is an approximate maximum possible value for a predetermined number of content items. 
     
     
         15 . The system of  claim 13 , wherein the content items correspond to installable applications which may be downloaded from the system by a plurality of users, and
 the calculating component further determines the quality scores based on at least one factor selected from the group consisting of: a number of times an application has been installed by the users, a number of times the application has been uninstalled by the users, an average rating of the application by the users, a number of users that have rated the application, and a rating of a developer who created the application.   
     
     
         16 . The system of  claim 12 , wherein the storage component stores in the storage device download data corresponding to statistics of content items downloaded by a plurality of users, and
 wherein the scoring component further determines the co-selection scores based on the download data.   
     
     
         17 . The system of  claim 12 , further comprising:
 a tagging component to associate one or more data tags with each of the plurality of content items,   wherein the processing component further selects the sub-set such each of the sub-set content items are associated with one or more data tags of a predetermined set of one or more data tags.   
     
     
         18 . The system of  claim 17 , wherein the tagging component associates one or more data tags with user data for each of the plurality of users, and
 wherein the processing component further selects the sub-set such the predetermined set of one or more data tags comprises the one or more data tags which are associated with a given user.   
     
     
         19 . The system of  claim 12 , wherein the storage component stores in the storage device first data tags associated with demographic user data for each of the plurality of users, and
 wherein the processing component further selects the sub-set such that each sub-set content item is associated with a first data tag   
     
     
         20 . The system of  claim 12 , wherein the output is provided to the user when there is a change in the short term user data.