Relevance scores for result ranking and arrangement

• US 8,001,141 B1
• Filed: 12/01/2009
• Issued: 08/16/2011
• Est. Priority Date: 03/29/2007
• Status: Active Grant

• Current Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Original Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Inventors: Bar, Tamar
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Wilcox; James J
• Application Number:
  US12/628,751
• Time in Patent Office: 623 Days
• Field of Search: None
• US Class Current: 707/769
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]

Abstract

Methods and systems for configuring the display of items in a network-based merchandising environment are provided. In one embodiment, the present invention identifies items within a catalog of items, computes a browse relevance score for each item, and configures the display of a Web page such that items with the highest browse relevance score are prominently displayed. In at least one embodiment, browse data, such as category fit data, popularity data, newness data, etc., are used to compute the browse relevance score for each item.

Claims

1. A computer-implemented method of configuring items for display in an electronic marketplace, comprising:

under control of one or more computer systems configured with executable instructions,

receiving item data for each of a plurality of items available for consumption through the electronic marketplace, each item defining a product, each item being associated with at least one category node in a hierarchical categorization tree comprising a plurality of category nodes, each category node including a plurality of product categories, wherein each item is associated with at least one product category of the plurality of product categories included in each category node;

obtaining browse relevance data for each of the plurality of items, the browse relevance data representing previous user actions with respect to any of the category nodes associated with each item;

for each item, determining a browse relevance score for each category node associated with the item based at least in part upon the browse relevance data, the browse relevance score including a score for the item in a product category with which the item is associated, the score for the item including a popularity score for the item, the popularity score being determined based on a prediction boost score that provides a boost for the score for the item based at least in part on whether the item is based on a popular character;

generating a ranked list of browse relevance scores for each category node, each ranked list including the browse relevance scores for each item associated with the category node;

in response to receiving a user-generated request corresponding to one of the category nodes, determining an arrangement of at least a portion of the items associated with the category node based at least in part upon the ranked list of browse relevance scores for the category node, the arrangement including images of two or more items with the highest browse relevance scores among browse relevance scores determined for the items associated with the category node, the arrangement further including a title of a product category with which each of the two or more items is associated; and

providing a page including the arrangement for display to the user having generated the request.

2. The computer-implemented method of claim 1, wherein the browse relevance data for each item includes at least one of browse data relating to browse habits of a plurality of users with respect to each item, popularity data relating to each item, newness data relating to an amount of time that each item has been available, and category fit data indicating a determined measure of relatedness between each item and each associated category node.

3. The computer-implemented method of claim 1, wherein the browse relevance score for each item for each associated category node is determined by summing scores determined using the browse relevance data, the scores including at least one of a category fit score, a browse score, and a popularity score for the item and the category node.

4. A computer-implemented method of arranging items for display, comprising:

under control of one or more computer systems configured with executable instructions,

obtaining item data for each of a plurality of items, each item being associated with at least one category node in a hierarchical categorization tree comprising a plurality of category nodes;

obtaining browse relevance data for each of the plurality of items, each item defining a product, the browse relevance data representing previous user actions with respect to any of the category nodes associated with each item, each category node including a plurality of product categories, wherein each item is associated with at least one product category of the plurality of product categories included in each category node;

for each item, determining a browse relevance score for each category node associated with the item based at least in part upon the browse relevance data, the browse relevance score including a score for the item in a product category with which the item is associated, the score for the item including a popularity score for the item, the popularity score being determined based on a prediction boost score that provides a boost for the score for the item based at least in part on whether the item is based on a popular character;

in response to receiving a user-generated request corresponding to one of the category nodes, determining an arrangement of at least a portion of the items associated with that category node based at least in part upon the browse relevance score for each of the items with respect to that category node, the arrangement including images of two or more items with the highest browse relevance scores among browse relevance scores determined for the items associated with the category node, the arrangement further including a title of a product category with which each of the two or more items is associated; and

providing the arrangement for display to the user having generated the request.

5. The computer-implemented method of claim 4, wherein determining a browse relevance score for each category node associated with the item includes generating a ranked list of relevance scores for each item associated with the category node.

6. The computer-implemented method of claim 4, wherein the browse relevance data for each item includes at least one of browse data relating to browse habits of a plurality of users with respect to each item, popularity data relating to each item, newness data relating to an amount of time that each item has been available, and category fit data indicating a determined measure of relatedness between each item and each associated category node.

7. The computer-implemented method of claim 4, wherein the browse relevance score for each item for each associated category node is determined by summing scores determined using the browse relevance data, the scores including at least one of a category fit score, a browse score, and a popularity score for the item and the category node.

8. The computer-implemented method of claim 7, wherein the browse relevance score is further determined by multiplying the summed scores by a price range score for the item.

9. The computer-implemented method of claim 7, wherein the popularity score is determined based at least in part upon a sales rank score for the item with respect to an associated category node.

10. The computer-implemented method of claim 4, wherein the arrangement is provided as part of a Web page requested by the user.

11. A system for arranging items for display, comprising:

a catalog server, including at least one processor, operable to execute instructions to provide item data for each of a plurality of items, each item defining a product, each item being associated with at least one category node in a hierarchical categorization tree comprising a plurality of category nodes, each category node including a plurality of product categories, wherein each item is associated with at least one product category of the plurality of product categories included in each category node;

a relevance server, including at least one processor, operable to execute instructions to obtain the item data and further to obtain browse relevance data for each of the plurality of items, the browse relevance data representing previous user actions with respect to any of the category nodes associated with each item, the relevance server being further operable to determine a browse relevance score for each item for each category node associated with the item, based at least in part upon the browse relevance data, the browse relevance score including a score for the item in a product category with which the item is associated, the score for the item including a popularity score for the item, the popularity score being determined based on a prediction boost score that provides a boost for the score for the item based at least in part on whether the item is based on a popular character; and

a retail server, including at least one processor, operable to execute instructions to, in response to receiving a user-generated request corresponding to one of the category nodes, determine an arrangement of at least a portion of the items associated with that category node based at least in part upon the browse relevance score for each of the items with respect to that category node, the arrangement including images of two or more items with the highest browse relevance scores among browse relevance scores determined for the items associated with the category node, the arrangement further including a title of a product category with which each of the two or more items is associated, the retail server being further operable to provide the arrangement for display to the user having generated the request.

12. The system of claim 11, further comprising:

at least one data store configured to store at least one of the item data and the browse relevance data.

13. The system of claim 11, wherein the relevance server is further operable to update, over time, the browse relevance scores for each item for each category associated with the item.

14. A system for arranging items for display, comprising:

a processor; and

memory including instructions that, when executed by the processor, cause the processor to;

obtain item data for each of a plurality of items, each item defining a product, each item being associated with at least one category node in a hierarchical categorization tree comprising a plurality of category nodes, each category node including a plurality of product categories, wherein each item is associated with at least one product category of the plurality of product categories included in each category node;

obtain browse relevance data for each of the plurality of items, the browse relevance data representing previous user actions with respect to any of the category nodes associated with each item;

for each item, determine a browse relevance score for each category node associated with the item based at least in part upon the browse relevance data, the browse relevance score including a score for the item in a product category with which the item is associated, the score for the item including a popularity score for the item, the popularity score being determined based on a prediction boost score that provides a boost for the score for the item based at least in part on whether the item is based on a popular character;

in response to receiving a user-generated request corresponding to one of the category nodes, determine an arrangement of at least a portion of the items associated with that category node based at least in part upon the browse relevance score for each of the items with respect to that category node, the arrangement including images of two or more items with the highest browse relevance scores among browse relevance scores determined for the items associated with the category node, the arrangement further including a title of a product category with which each of the two or more items is associated; and

provide the arrangement for display to the user having generated the request.

15. The system of claim 14, wherein determining a browse relevance score for each category node associated with the item includes generating a ranked list of relevance scores for each item associated with the category node.

16. The system of claim 14, wherein the browse relevance data for each item includes at least one of browse data relating to browse habits of a plurality of users with respect to each item, popularity data relating to each item, newness data relating to an amount of time that each item has been available, and category fit data indicating a determined measure of relatedness between each item and each associated category node.

17. The system of claim 14, wherein the browse relevance score for each item for each associated category node is determined by summing scores determined using the browse relevance data, the scores including at least one of a category fit score, a browse score, and a popularity score for the item and the category node.

18. The system of claim 17, wherein the browse relevance score is further determined by multiplying the summed scores by a price range score for the item.

19. The system of claim 17, wherein the popularity score is determined based at least in part upon a sales rank score for the item with respect to an associated category node.

20. A computer-readable storage medium storing instructions for arranging items for display in an electronic environment, the instructions when executed by a processor causing the processor to:

obtain item data for each of a plurality of items, each item defining a product, each item being associated with at least one category node in a hierarchical categorization tree comprising a plurality of category nodes, each category node including a plurality of product categories, wherein each item is associated with at least one product category of the plurality of product categories included in each category node;

obtain browse relevance data for each of the plurality of items, the browse relevance data representing previous user actions with respect to any of the category nodes associated with each item;

for each item, determine a browse relevance score for each category node associated with the item based at least in part upon the browse relevance data, the browse relevance score including a score for the item in a product category with which the item is associated, the score for the item including a popularity score for the item, the popularity score being determined based on a prediction boost score that provides a boost for the score for the item based at least in part on whether the item is based on a popular character; and

in response to receiving a user-generated request corresponding to one of the category nodes, determine an arrangement of at least a portion of the items associated with that category node based at least in part upon the browse relevance score for each of the items with respect to that category node, the arrangement including images of two or more items with the highest browse relevance scores among browse relevance scores determined for the items associated with the category node, the arrangement further including a title of a product category with which each of the two or more items is associated; and

provide the arrangement for display to the user having generated the request.

21. The computer-readable storage medium of claim 20, wherein determining a browse relevance score for each category node associated with the item includes generating a ranked list of relevance scores for each item associated with the category node.

22. The computer-readable storage medium of claim 20, wherein the browse relevance data for each item includes at least one of browse data relating to browse habits of a plurality of users with respect to each item, popularity data relating to each item, newness data relating to an amount of time that each item has been available, and category fit data indicating a determined measure of relatedness between each item and each associated category node.

23. The computer-readable storage medium of claim 20, wherein the browse relevance score for each item for each associated category node is determined by summing scores determined using the browse relevance data, the scores including at least one of a category fit score, a browse score, and a popularity score for the item and the category node.

24. The computer-readable storage medium of claim 23, wherein the browse relevance score is further determined by multiplying the summed scores by a price range score for the item.

25. The computer-readable storage medium of claim 23, wherein the popularity score is determined based at least in part upon a sales rank score for the item with respect to an associated category node.