Selecting content objects for recommendation based on content object collections

Abstract

Each image of a plurality of user-defined collections of images has properties associated therewith. A seed image is defined as currently viewed or selected image and the properties associated with the seed image are retrieved. The seed image is used to identify seed collections as the collections of images which have the seed image as one of their images. A plurality of candidate images are identified from the seed collections. For each candidate image, a significance score is determined as a function of either the number of seed collections to which the candidate image belongs or the ratio of the number of seed collections to the set of all collections to which the candidate image belongs. Each candidate image is ranked based on the significance scores and the seed image properties. At least one candidate image is recommended to the user based on the ranking of the candidate images.

What is claimed is: 
     
         1 . A method of selecting an image for recommendation to a user of a content hosting service, the method operating on the content hosting service having a content storage module, a collection storage module and a processor and operative with a client module, the method comprising:
 storing a plurality of user-defined collections of images in said collection storage module from a plurality of said users of said content hosting service, each said image having properties associated therewith;   via said processor, defining a seed image as an image currently viewed or selected by a current user and retrieving said properties associated with said seed image;   via said processor, using said seed image to identify seed collections from among said user-defined collections of images which have said seed image as one of the images of said seed collection;   via said processor, identifying a plurality of candidate images for recommendation from said selected seed collections;   via said processor, determining for each candidate image, a significance score as a function of either the number of seed collections to which said candidate image belongs or the ratio of said number of seed collections to the set of all collections to which said candidate image belongs;   via said processor, ranking each candidate image based on said significance scores and on said seed image properties; and   via said processor, recommending at least one candidate image for recommendation to the user via said client module based on the ranking of the candidate images.   
     
     
         2 . The method of  claim 1 , wherein said seed image properties comprise a category of said seed image. 
     
     
         3 . The method of  claim 1 , wherein said seed image properties comprise tags associated with said seed image. 
     
     
         4 . The method of  claim 1 , wherein said seed image properties comprise characteristics of said seed image. 
     
     
         5 . The method of  claim 1 , and also comprising retrieving extrinsic properties of said seed image to use in said ranking. 
     
     
         6 . A method of selecting an image for recommendation to a user of a content hosting service, the method operating on the content hosting service having a content storage module, a collection storage module and a processor and operative with a client module, the method comprising:
 storing a plurality of user-defined collections of images in said collection storage module from a plurality of said users of said content hosting service, each said image having properties associated therewith;   via said processor, defining a set of seed image as a set of images currently viewed or selected by a current user and retrieving said per-seed-image properties associated with said set of seed images;   via said processor, using said set of seed images to identify seed collections from among said user-defined collections of images which have at least one of said seed images as one of the images of said seed collection;   via said processor, identifying a plurality of candidate images for recommendation from said selected seed collections;   via said processor, determining for each candidate image, a significance score as a function of either the number of seed collections to which said candidate image belongs or the ratio of said number of seed collections to the set of all collections to which said candidate image belongs;   via said processor, ranking each candidate image based on said significance scores and on said per-seed-image properties; and   via said processor, recommending at least one candidate image for recommendation to the user via said client module based on the ranking of the candidate images.   
     
     
         7 . The method of  claim 6 , wherein said per-seed-image properties comprise a category of said seed image. 
     
     
         8 . The method of  claim 6 , wherein said per-seed-image properties comprise tags associated with said seed image. 
     
     
         9 . The method of  claim 6 , wherein said per-seed-image properties comprise characteristics of said seed image. 
     
     
         10 . The method of  claim 6 , and also comprising retrieving extrinsic properties of said seed image to use in said ranking. 
     
     
         11 . A content hosting system comprising a processor and operative with a client module, the system comprising:
 a collection storage module to store a plurality of user-defined collections of images from a plurality of said users of said content hosting service, each said image having properties associated therewith; and   a content recommendation engine operative on said processor to select an image for recommendation to a user of said content hosting service, said engine comprising:   a seed object module to define a seed image as an image currently viewed or selected by a current user, to retrieve said properties associated with said seed image and to use said seed image to identify seed collections from among said user-defined collections of images which have said seed image as one of the images of said seed collection;   a candidate identification module to identify a plurality of candidate images for recommendation from said selected seed collections and to determine for each candidate image, a significance score as a function of either the number of seed collections to which said candidate image belongs or the ratio of said number of seed collections to the set of all collections to which said candidate image belongs;   a ranker to rank each candidate image based on said significance scores and on said seed image properties and to recommend at least one candidate image for recommendation to the user via said client module based on the ranking of the candidate images.   
     
     
         12 . The system of  claim 11 , wherein said seed image properties comprise a category of said seed image. 
     
     
         13 . The system of  claim 11 , wherein said seed image properties comprise tags associated with said seed image. 
     
     
         14 . The system of  claim 11 , wherein said seed image properties comprise characteristics of said seed image. 
     
     
         15 . The system of  claim 11 , and wherein said ranker further to retrieve extrinsic properties of said seed image to use in said ranking.