Systems and methods for contextual recommendations

Abstract

A method and a system for making contextual recommendations to users on a network-based system. For example, activity associated with a user interacting with a network-based system is tracked. Based, at least in part, on the tracked user activity on the network-based system, a recommendation relationship type is selected. The recommendation relationship type can be either a substitute relationship type or a complement relationship type. A recommended object can be selected based at least in part on the recommendation relationship type and a first object accessed by the user interacting with the network-based system. A recommendation can be generated for the recommended object for presentation to the user interacting with the network-based system.

What is claimed is: 
     
         1 . A method comprising:
 tracking, on a server including a processor, user activity associated with a user interacting with a network-based system;   determining, on the server, a recommendation relationship type based at least in part on the tracked user activity, the recommendation relationship type selected from a group of recommendation relationship types including a substitute relationship type and a complement relationship type;   selecting, on the server, a recommended object based at least in part on the recommendation relationship type and a first object accessed by the user interacting with the network-based system;   determining a relationship strength between the recommended object and the first object accessed by the user interacting with the network-based system, the relationship strength based on at least a navigation history between the first object and the recommended object; and   causing display of a relationship strength indicator based on at least the relationship strength between the first object and the recommended object and the recommended relationship type.   
     
     
         2 . The method of  claim 1 , wherein determining the relationship strength includes evaluating a historical transactional relationship between the recommended object and the first object. 
     
     
         3 . The method of  claim 1 , wherein determining the relationship strength includes evaluating a historical navigational relationship between the recommended object and the first object. 
     
     
         4 . The method of  claim 1 , wherein the determining the recommendation relationship type includes determining a purchase lifecycle location based at least in part on the tracked user activity. 
     
     
         5 . The method of  claim 4 , wherein determining the purchase lifecycle location includes evaluating a series of navigation operations performed by the user on the network-based system. 
     
     
         6 . The method of  claim 4 , wherein determining the purchase lifecycle location includes determining from the series of navigation operations whether the user is pre-purchase or post-purchase, and,
 wherein determining the recommendation relationship type includes selecting a complementary relationship type if the user is determined to be post-purchase, and selecting a substitutionary relationship type if the user is determined to be pre-purchase.   
     
     
         7 . The method of  claim 1 , wherein selecting the recommended object includes evaluating a plurality of relationship scores between the recommended object and a plurality of second objects related to the first object according the determined recommendation relationship type. 
     
     
         8 . A computer-implemented recommendation system for use within a network-based system, the system comprising:
 a tracking module to track user activity associated with a user interacting with a network-based system;   a relationship module to calculate a recommendation relationship type based at least in part on the tracked user activity, the recommendation relationship type selected from a group of recommendation relationship types including a substitute relationship type and a complement relationship type;   a recommendation engine to select a recommended object based at least in part on the recommendation relationship type and a first object accessed by the user interacting with the network-based system;   a type module to determining a relationship strength between the recommended object and the first object accessed by the user interacting with the network-based system, the relationship strength based on at least a navigation history between the first object and the recommended object, and cause display of a relationship strength indicator based on at least the relationship strength between the first object and the recommended object and the recommended relationship type.   
     
     
         9 . The computer-implemented recommendation system of  claim 8 , wherein the type module is configured to determine the relationship strength based on an evaluation of a historical transactional relationship between the recommended object and the first object. 
     
     
         10 . The computer-implemented recommendation system of  claim 8 , wherein the type module is configured to determine the relationship strength based on an evaluation of a historical navigational relationship between the recommended object and the first object. 
     
     
         11 . The computer-implemented recommendation system of  claim 8 , wherein the recommendation engine determines the recommended relationship type based on a determination of a purchase lifecycle location based at least in part on the tracked user activity. 
     
     
         12 . The computer-implemented recommendation system of  claim 11 , wherein the recommendation engine determines the purchase lifecycle location based on an evaluation of a series of navigational operations performed by the user on the network-based system. 
     
     
         13 . The computer-implemented recommendation system of  claim 11 , wherein the recommendation engine determines the purchase lifecycle location based on:
 a determination from the tracked user activity of whether the user is pre-purchase or post-purchase, and   wherein the determining the recommendation relationship type includes selecting a complementary relationship type if the user is determined to be post-purchase, and selecting a substitutionary relationship type if the user is determined to be pre-purchase.   
     
     
         14 . The computer-implemented recommendation system of  claim 11 , wherein the recommendation engine selects the recommended object based on an evaluation of a plurality of relationship scores between the recommended object and a plurality of second objects related to the first object according the determined recommendation relationship type. 
     
     
         15 . A machine-readable storage medium embodying instructions which, when executed by a computer-implemented network-based system, cause the network-based system to:
 track, on a server including a processor, user activity associated with a user interacting with a network-based system;   determine, on the server, a recommendation relationship type based at least in part on the tracked user activity, the recommendation relationship type selected from a group of recommendation relationship types including a substitute relationship type and a complement relationship type;   select, on the server, a recommended object based at least in part on the recommendation relationship type and a first object accessed by the user interacting with the network-based system;   determine a relationship strength between the recommended object and the first object accessed by the user interacting with the network-based system, the relationship strength based on at least a navigation history between the first object and the recommended object; and   cause display of a relationship strength indicator based on at least the relationship strength between the first object and the recommended object and the recommended relationship type.   
     
     
         16 . The machine-readable storage medium of  claim 15 , wherein the instructions to determine the relationship strength include instructions to evaluate a historical transactional relationship between the recommended object and the first object. 
     
     
         17 . The machine-readable storage medium of  claim 15 , wherein the instructions to determine the relationship strength include instructions to evaluate a historical navigational relationship between the recommended object and the first object. 
     
     
         18 . The machine-readable storage medium of  claim 15 , wherein the instructions to determine the recommendation relationship type include instructions to determine a purchase lifecycle location based at least in part on the tracked user activity. 
     
     
         19 . The machine-readable storage medium of  claim 18 , wherein the instructions to determine the purchase lifecycle location includes instructions to evaluate a series of navigation operations performed by the user on the network-based system. 
     
     
         20 . The machine-readable storage medium of  claim 18 , wherein the instructions to determining the purchase lifecycle location include instructions to determine from the series of navigation operations whether the user is pre-purchase or post-purchase, and,
 wherein the instructions to determine the recommendation relationship type includes instructions for selecting a complementary relationship type if the user is determined to be post-purchase, and selecting a substitutionary relationship type if the user is determined to be pre-purchase.