Method and system for search refinement

Abstract

A search refinement method and system to select items from a plurality of items using a user device connected via the internet to a server apparatus. The server apparatus includes a database of items and item attributes. Images of a first subset of the items are transmitted to the user device where they are presented on a display. A change indication selected by a user of the user device is received by the server apparatus, the change indication including at least one of an attribute selection and a reference item selected from the first subset. Responsive to receiving the change indication, an item score is computed for each of the items based on the item's attributes and the change indication. Images of a second subset of the items are presented on the user device. Each time the user initiates a change indication, another subset is presented on the user device, and the user can iteratively refine the search by finding items similar to the items represented by the change indication.

1 . A search refinement method comprising:
 by a processing device according to processing instructions embedded in one or more non-transitory computer readable medium:   establishing an internet connection between the processing device and a user device;   accessing a database of items and item attributes;   transmitting to the user device images of a first subset of the items and user interface instructions structured to render the images on the user device;   receiving via the internet connection a change indication selected by a user of the user device, the change indication including at least one of an attribute selection and a reference item selected from the first subset; and   responsive to receiving the change indication, computing an item score for each of the items based on the item's attributes and the change indication;   forming, based on the item scores, a second subset of the items; and   serving user interface instructions structured to render images of the second subset of the items on the user device.   
     
     
         2 . A method as in  claim 1 , wherein computing an item score includes multiplying a vector corresponding to the item's attributes to a change vector corresponding to the change indication. 
     
     
         3 . A method as in  claim 3 , wherein forming a second subset of the items includes selecting the items with the highest scores. 
     
     
         4 . A method as in  claim 1 , wherein forming a second subset of the items includes: comparing the attributes of the items to the attributes corresponding to the change indication; and selecting items based on the comparison. 
     
     
         5 . A method as in  claim 1 , further comprising:
 before serving user interface instructions structured to render images of a first subset of the items:   serving user interface instructions structured to render images of a first set of subcategory choices, each subcategory choice in the first set of subcategory choices having attributes;   receiving via the internet connection an indication of a subcategory choice selected from the first set of subcategory choices by the user of the user device; and   including in the first subset items with attributes similar to attributes of the subcategory choice, the similar items including items with matching attributes and items with non-matching attributes; and   excluding from the first subset items with attributes dissimilar to the subcategory choice.   
     
     
         6 . A method as in  claim 5 , further comprising:
 computing item scores for each of the items based on the item's attributes and the subcategory choice; and identifying similar and dissimilar items using the item scores.   
     
     
         7 . A method as in  claim 1 , further comprising:
 receiving via the internet connection an indication of a filtering facet selected by the user of the user device;   if the filtering facet is inclusive, including in one of the first subset and the second subset only items matching corresponding attributes of the filtering facet; and   if the filtering facet is exclusive, including in one of the first subset and the second subset only items not matching corresponding attributes of the filtering facet.   
     
     
         8 . A method as in  claim 1 , further comprising storing a query including the change indication. 
     
     
         9 . A method as in  claim 2 , further comprising storing an alert parameter, periodically executing the query, and outputting an alert if the alert parameter is satisfied. 
     
     
         10 . A method as in  claim 1 , further comprising compiling information including the change indication, and selling the information so compiled. 
     
     
         11 . A search refinement method comprising:
 by a processing device according to processing instructions embedded in one or more non-transitory computer readable medium:   accessing a recommendation set of items having item attributes;   receiving from a user an indication of a reference item, the reference item having reference item attributes;   receiving an item attribute selection from the user;   computing item scores for each of the items based on the item attributes, the item attribute selection and the reference item attributes;   ranking the items based on the item scores; and   serving user interface instructions for rendering images of at least a subset of the items on the user device, the at least a subset of the items selected according to the ranking of the items.   
     
     
         12 . A search refinement system comprising:
 a database of items and item attributes;   user interface instructions operable in a user device and structured to cause the user device to receive images of a first subset of the items and of a second subset of the items; to present the images; and to transmit a change indication selected by a user of the user device, the change indication including at least one of an attribute selection and a reference item selected from the first subset; and   a server apparatus including a processing device, one or more non-transitory computer readable medium, and a plurality of modules including:
 a communication module structured to establish an internet connection between the processing device and a user device, to transmit the images of the first subset of items and of the second subset of items; and to receive the change indication; 
 a database interface module structured to access the database of items and item attributes; 
 a user interface module structured to generate the user interface instructions; and 
 a processing module structured to compute an item score for each of the items based on the item's attributes and the change indication; and to form, based on the item scores, the second subset of the items. 
   
     
     
         13 . A system as in  claim 12 , wherein the processing module computes the item score by multiplying a vector corresponding to the item's attributes to a change vector corresponding to the change indication. 
     
     
         14 . A system as in  claim 13 , wherein the processing module forms the second subset by selecting the items with the highest scores. 
     
     
         15 . A system as in  claim 12 , wherein the processing module forms the second subset by comparing the attributes of the items to the attributes corresponding to the change indication; and selecting items based on the comparison. 
     
     
         16 . A system as in  claim 12 , wherein the user interface instructions are structured to present images of a first set of subcategory choices, each subcategory choice in the first set of subcategory choices having attributes; and to transmit via the internet connection an indication of a subcategory choice selected from the first set of subcategory choices by the user of the user device; and wherein the processing module is structured to include in the first subset items with attributes similar to attributes of the subcategory choice, the similar items including items with matching attributes and items with non-matching attributes; and to exclude from the first subset items with attributes dissimilar to the subcategory choice. 
     
     
         17 . A system as in  claim 16 , wherein the processing module is structured to compute item scores for each of the items based on the item's attributes and the subcategory choice; and to identify similar and dissimilar items using the item scores. 
     
     
         18 . A system as in  claim 12 , wherein the user interface instructions are structured to transmit via the internet connection an indication of a filtering facet selected by the user of the user device; and wherein the processing module is structured to
 if the filtering facet is inclusive, include in one of the first subset and the second subset only items matching corresponding attributes of the filtering facet; and   if the filtering facet is exclusive, include in one of the first subset and the second subset only items not matching corresponding attributes of the filtering facet.