Methods and systems for activity-based recommendations

Abstract

Embodiments of computer-implemented methods and systems for activity-based recommendations are described. One example embodiment includes receiving historical activities of a plurality of users, the historical activities comprising actions associated with the items in the database, the plurality of users including the target user. A group of users is identified from the plurality of users based on historical activities of the group of users and historical activities of the target user. Generally current time activities of the group of users are received, the generally current time activities comprising activities that have occurred within a defined time window. A recommendation for one or more items is generated based on the generally current time activities, and the recommendation is provided for display to the target user.

The invention claimed is: 
     
         1 . A computer-implemented method for providing recommendations for items stored in a database to a target user, the method comprising:
 receiving historical activities of a plurality of users, the historical activities comprising actions associated with the items in the database, the plurality of users including the target user;   identifying a group of users from the plurality of users based on historical activities of the group of users and historical activities of the target user;   receiving generally current time activities of the group of users, the generally current time activities comprising activities that have occurred within a defined time window;   generating a recommendation for one or more items based on the generally current time activities; and   providing the recommendation for display to the target user.   
     
     
         2 . The method of  claim 1 , wherein identifying the group of users comprises:
 generating a plurality of activity-based networks based on the historical activities of the plurality of users;   identifying a first activity-based network that includes at least one historical activity of the target user; and   identifying the group of users from the first activity-based network.   
     
     
         3 . The method of  claim 2 , wherein the plurality of activity-based networks are generated based on activity clusters from the historical activities of the plurality of users. 
     
     
         4 . The method of  claim 3 , wherein a first activity cluster is defined by:
 identifying a first historical activity performed by a first subset of users;   identifying a second historical activity performed by a threshold percentage of users from the first subset of users; and   clustering the first historical activity and the second historical activity in the first activity cluster.   
     
     
         5 . The method of  claim 1 , wherein identifying the group of users comprises:
 generating an activity-based network for the target user based on one or more historical activities of the target user; and   identifying users whose historical activities are included in the activity-based network for the target user.   
     
     
         6 . The method of  claim 1 , wherein the historical activities comprise at least one selected from the following: searches for items from the database, purchase activity associated with items from the database, and selections of items from the database. 
     
     
         7 . The method of  claim 1 , wherein the recommendation for the one or more items is generated based on one or more preferences of the target user. 
     
     
         8 . The method of  claim 1 , wherein the method further comprises:
 receiving updated historical activities for the plurality of users; and   updating the group of users based on the updated historical activities.   
     
     
         9 . One or more computer storage media storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform operations for providing recommendations for items stored in a database to a target user, the operations comprising:
 receiving historical activities of a plurality of users, the historical activities comprising actions associated with the items in the database, the plurality of users including the target user;   identifying a group of users from the plurality of users based on historical activities of the group of users and historical activities of the target user;   receiving generally current time activities of the group of users, the generally current time activities comprising activities that have occurred within a defined time window;   generating a recommendation for one or more items based on the generally current time activities; and   providing the recommendation for display to the target user.   
     
     
         10 . The media of  claim 9 , wherein identifying the group of users comprises:
 generating a plurality of activity-based networks based on the historical activities of the plurality of users;   identifying a first activity-based network that includes at least one historical activity of the target user; and   identifying the group of users from the first activity-based network.   
     
     
         11 . The media of  claim 10 , wherein the plurality of activity-based networks are generated based on activity clusters from the historical activities of the plurality of users. 
     
     
         12 . The media of  claim 11 , wherein a first activity cluster is defined by:
 identifying a first historical activity performed by a first subset of users;   identifying a second historical activity performed by a threshold percentage of users from the first subset of users; and   clustering the first historical activity and the second historical activity in the first activity cluster.   
     
     
         13 . The media of  claim 9 , wherein identifying the group of users comprises:
 generating an activity-based network for the target user based on one or more historical activities of the target user; and   identifying users whose historical activities are included in the activity-based network for the target user.   
     
     
         14 . The media of  claim 9 , wherein the historical activities comprise at least one selected from the following: searches for items from the database, purchase activity associated with items from the database, and selections of items from the database. 
     
     
         15 . The media of  claim 9 , wherein the recommendation for the one or more items is generated based on one or more preferences of the target user. 
     
     
         16 . The media of  claim 9 , wherein the operations further comprise:
 receiving updated historical activities for the plurality of users; and   updating the group of users based on the updated historical activities.   
     
     
         17 . A computer system comprising:
 one or more processors; and   one or more computer storage media storing computer-useable instructions that cause the one or more processors to:   receive historical activities of a plurality of users, the historical activities comprising actions associated with the items in the database, the plurality of users including the target user;   identify a group of users from the plurality of users based on historical activities of the group of users and historical activities of the target user;   receive generally current time activities of the group of users, the generally current time activities comprising activities that have occurred within a defined time window;   generate a recommendation for one or more items based on the generally current time activities; and   provide the recommendation for display to the target user.   
     
     
         18 . The system of  claim 17 , wherein the group of users are identified by:
 generating a plurality of activity-based networks based on activity clusters from the historical activities of the plurality of users;   identifying a first activity-based network that includes at least one historical activity of the target user; and   identifying the group of users from the first activity-based network.   
     
     
         19 . The system of  claim 17 , wherein the group of users are identified by:
 generating an activity-based network for the target user based on one or more historical activities of the target user; and   identifying users whose historical activities are included in the activity-based network for the target user.   
     
     
         20 . The system of  claim 17 , wherein the recommendation for the one or more items is generated based on one or more preferences of the target user.