Recommending Content Items

Abstract

Recommending content items is disclosed. In some embodiments, determining an item to recommend includes receiving an indication that a first item is of interest and selecting a second item to recommend based at least in part on a degree of similarity of the second item to the first item as determined based at least in part on a number of other instances in which both the first item and the second item have been of interest and a first decayed popularity of the first item. In some embodiments, the degree of similarity of the second item to the first item is determined based in addition on a second decayed popularity of the second item.

1 . (canceled) 
     
     
         2 . A method of determining an item to recommend, comprising:
 receiving at a processor an indication that a first item is of interest to a user;   determining a second item to recommend to the user based at least in part on a degree of similarity of the second item to the first item; and   providing the second item as a recommendation to the user in the event that the second item is not already associated with the user;   wherein the degree of similarity is determined based at least in part on a similarity score that quantifies the degree of similarity of the second item to the first item based at least in part on a number of instances in which both the first item and the second item have been of interest to one or more other users, and a time-decayed popularity of the second item; wherein, when determining the time-decayed popularity of a respective item, a more recent instance in which the respective item has been of interest to another user contributes more to the time-decayed popularity of the respective item than a less recent instance in which the respective item has been of interest to another user.   
     
     
         3 . The method as recited in  claim 2 , wherein the similarity score that quantifies the degree of similarity of the second item to the first item is based at least in part on the number of instances in which both the first item and the second item have been of interest to one or more other users, a time-decayed popularity of the first item, and the time-decayed popularity of the second item. 
     
     
         4 . The method as recited in  claim 3 , wherein the similarity score is determined at least in part by dividing the number of instances in which both the first item and the second item have been of interest by the square root of the sum of the squares of the time-decayed popularity of the first item and the time-decayed popularity of the second item. 
     
     
         5 . The method as recited in  claim 2 , wherein determining a second item to recommend comprises selecting a pair of items from a set of one or more pairs of items, each of which includes the first item. 
     
     
         6 . The method as recited in  claim 5 , wherein each pair in the set of one or more pairs of items comprises a pair of items that were of interest to at least one user within a sliding time window having a predefined window size. 
     
     
         7 . The method as recited in  claim 5 , wherein for each pair in the set of one or more pairs a similarity score that quantifies the degree of similarity between the items comprising the pair is determined and a pair of items is selected from the set of one or more pairs of items based at least in part on the respective similarity scores of the one or more pairs. 
     
     
         8 . The method as recited in  claim 2 , wherein the second item is one of a plurality of items selected to be recommended. 
     
     
         9 . The method as recited in  claim 2 , further comprising adding the second item to a list of items suitable for being recommended to the user. 
     
     
         10 . The method as recited in  claim 2 , further including determining that the second item is already associated with the user and not recommending the second item to the user based at least in part on the determination that the second item is already associated with the user. 
     
     
         11 . The method as recited in  claim 10 , wherein determining that the second item is already associated with the user includes consulting a record of items associated with the user. 
     
     
         12 . The method as recited in  claim 10 , wherein determining that the second item is already associated with the user includes determining whether the second item is associated with a client application associated with the user. 
     
     
         13 . The method as recited in  claim 10 , wherein determining that the second item is already associated with the user includes checking user data stored on a host associated with the user. 
     
     
         14 . The method as recited in  claim 2 , further comprising selecting the second item based at least in part on what the user has been indicating an interest in most recently. 
     
     
         15 . The method as recited in  claim 2 , further comprising selecting the second item based at least in part on what the user is indicating an interest in at a current moment. 
     
     
         16 . The method as recited in  claim 2 , further comprising selecting the second item based at least in part on what the user has rated highly in the past. 
     
     
         17 . The method as recited in  claim 2 , wherein the first item and the second item comprise digital content. 
     
     
         18 . A system for determining an item to recommend, comprising:
 a processor configured to:
 receive an indication that a first item is of interest to a user; 
 determine a second item to recommend to the user based at least in part on a degree of similarity of the second item to the first item; and 
 provide the second item as a recommendation to the user in the event that the second item is not already associated with the user; 
 wherein the degree of similarity is determined based at least in part on a similarity score that quantifies the degree of similarity of the second item to the first item based at least in part on a number of instances in which both the first item and the second item have been of interest to one or more other users, and a time-decayed popularity of the second item; wherein, when determining the time-decayed popularity of a respective item, a more recent instance in which the respective item has been of interest to another user contributes more to the time-decayed popularity of the respective item than a less recent instance in which the respective item has been of interest to another user; and 
   a memory coupled to the processor and configured to provide instructions to the processor.   
     
     
         19 . The system as recited in  claim 18 , wherein the similarity score that quantifies the degree of similarity of the second item to the first item is based at least in part on the number of instances in which both the first item and the second item have been of interest to one or more other users, a time-decayed popularity of the first item, and the time-decayed popularity of the second item. 
     
     
         20 . A non-transitory computer readable storage medium storing computer instructions that when executed by a processor of a computer system cause the computer system to:
 receive an indication that a first item is of interest to a user;   determine a second item to recommend to the user based at least in part on a degree of similarity of the second item to the first item; and   provide the second item as a recommendation to the user in the event that the second item is not already associated with the user;   wherein the degree of similarity is determined based at least in part on a similarity score that quantifies the degree of similarity of the second item to the first item based at least in part on a number of instances in which both the first item and the second item have been of interest to one or more other users, and a time-decayed popularity of the second item; wherein, when determining the time-decayed popularity of a respective item, a more recent instance in which the respective item has been of interest to another user contributes more to the time-decayed popularity of the respective item than a less recent instance in which the respective item has been of interest to another user.   
     
     
         21 . The non-transitory computer readable storage medium as recited in  claim 20 , wherein the similarity score that quantifies the degree of similarity of the second item to the first item is based at least in part on the number of instances in which both the first item and the second item have been of interest to one or more other users, a time-decayed popularity of the first item, and the time-decayed popularity of the second item.