Item recommendation system, item recommendation method and program

Abstract

An item recommendation system includes: a feedback receiving unit receiving feedbacks with respect to items by users; a feedback history storage unit storing information of feedbacks received at present or in the past by the feedback receiving unit in units of combinations of users and items; a relevance calculation unit calculating relevance between items base on first information indicating the degree in which the same user gives feedbacks to respective items and second information indicating the degree in which many users give feedbacks by using feedback information stored by the feedback history storage unit when receiving the feedback by the feedback receiving unit; and a recommendation unit immediately recommending items having higher relevance with respect to the item to the user who has given the feedback to the item by the feedback receiving unit based on the relevance calculated by the relevance calculation unit.

1 .- 8 . (canceled) 
     
     
         9 . An item recommendation method comprising:
 receiving, by one or more processors, feedback for a plurality of items from a plurality of users;   recommending, by the one or more processors, at least one item to a user based on first information and second information, the first information indicating feedback given by the user for at least one item and second information indicating past feedback given by at one or more other users to a plurality of items.   
     
     
         10 . The recommendation method of  claim 9 , wherein feedback includes at least one of purchase history, evaluation values, or access history of detailed information about an item. 
     
     
         11 . The recommendation method of  claim 9 , further comprising:
 determining, by the one or more processors, a user taste by analyzing the feedback from the user;   extracting, by the one or more processors, an item group comprising a subset of the plurality of items relevant to the feedback received for the at least one item; and   generating, by the one or more processors, a recommendation list based on the user taste comprising one or more items from the item group,   wherein   the at least one item recommended in the recommending step is selected from the recommendation list.   
     
     
         12 . The recommendation method of  claim 11 , wherein determining user taste further comprises assigning scores to items for which feedbacks for one or more items are received, the scores based on purchase frequency of the one or more items, evaluation value for the one or more items, and access frequency of detailed information about the one or more items. 
     
     
         13 . An item recommendation system comprising:
 a memory unit having instructions stored thereon;   one or more processors configured to execute the stored instructions, which cause the one or more processors to:
 receive feedback for a plurality of items from a plurality of users; 
 recommend at least one item to a user based on first information and second information, the first information indicating feedback given by the user for at least one item and second information indicating past feedback given by at least one another user to a plurality of items. 
   
     
     
         14 . The recommendation system of  claim 13 , wherein feedback includes at least one of purchase history, evaluation values, or access history of detailed information about an item. 
     
     
         15 . The recommendation system of  claim 13 , wherein the one or more processors are further configured to:
 determine a user taste by analyzing the feedback from the user;   extract an item group comprising a subset of the plurality of items relevant to the feedback received for the at least one item; and   generate a recommendation list based on the user taste comprising one or more items from the item group,   wherein   the at least one recommended item is selected from the recommendation list.   
     
     
         16 . The recommendation system of  claim 15 , wherein the one or more processors determine user taste by assigning scores to items for which feedbacks for one or more items are received, the scores based on purchase frequency of the one or more items, evaluation value for the one or more items, and access frequency of detailed information about the one or more items. 
     
     
         17 . A non-transitory computer readable medium having stored thereon instructions which when executed cause one or more processors to perform a recommendation method comprising:
 receiving feedback for a plurality of items from a plurality of users;   recommending at least one item to a user based on first information and second information, the first information indicating feedback given by the user for at least one item and second information indicating past feedback given by at least one another user to a plurality of items.   
     
     
         18 . The recommendation method of  claim 17 , wherein feedback includes at least one of purchase history, evaluation values, or access history of detailed information about an item. 
     
     
         19 . The recommendation method of  claim 17 , further comprising:
 determining, by the one or more processors, a user taste by analyzing the feedback from the user;   extracting, by the one or more processors, an item group comprising a subset of the plurality of items relevant to the feedback received for the at least one item; and   generating, by the one or more processors, a recommendation list based on the user taste comprising one or more items from the item group,   wherein   the at least one item recommended in the recommending step is selected from the recommendation list.   
     
     
         20 . The recommendation method of  claim 19 , wherein determining user taste further comprises assigning scores to items for which feedbacks for one or more items are received, the scores based on purchase frequency of the one or more items, evaluation value for the one or more items, and access frequency of detailed information about the one or more items.