COMPUTER PROCESSES FOR IDENTIFYING RELATED ITEMS AND GENERATING PERSONALIZED ITEM RECOMMENDATIONS

• US 20080033821A1
• Filed: 08/20/2007
• Published: 02/07/2008
• Est. Priority Date: 09/18/1998
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Eric Benson, Gregory Linden, Jennifer Jacobi
• Inventors: Benson, Eric, Jacobi, Jennifer, Linden, Gregory
• Granted Patent: US 7,921,042 B2
• Time in Patent Office: Days
• US Class Current: 705/14
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0253 During e-commerce, i.e. online transactions
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0271 Personalized advertisement
• CPC Class Codes: G06Q 30/0277 Online advertisement
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A computer-implemented service analyzes purchase histories and/or other types of behavioral data of users on an aggregated basis to detect and quantify associations between particular items represented in an electronic catalog. The detected associations are stored in a mapping structure that maps items to related items, and is used to recommend items to users of the electronic catalog.

Claims

1. A computer-implemented method, comprising:

maintaining transaction histories of each of a plurality of users in computer storage, said transaction histories identifying specific items for which transactions were conducted by the users;

generating data values that represent degrees to which specific items are related to each other, each data value corresponding to a pair of items, and being dependent upon how frequently the items of said pair co-occur in the transaction histories of the plurality of users;

using the data values to select, for each of a plurality of items, a corresponding set of related items;

generating, in computer storage, a mapping that maps each of said plurality of items to the respective set of related items; and

using the mapping structure, in combination with corresponding ones of said data values, to generate personalized item recommendations for each of a plurality of users.

2. The method of claim 1, wherein the transaction histories include item purchase histories.

3. The method of claim 1, wherein the method comprises generating personalized item recommendations for a target user, at least in part, by selecting a first plurality of items that are known to be of interest to the target user, and using the mapping and associated data values to map the first plurality of items to a second plurality of items to recommend to the target user.

4. The method of claim 3, wherein selecting the first plurality of items comprises selecting items from a purchase history of the user.

5. The method of claim 4, wherein generating the personalized item recommendations further comprises giving different amounts of weights to different items in the purchase history of the target user such that more weight is given to recently purchased items than to less recently purchased items.

6. The method of claim 4, wherein selecting the first plurality of items additionally comprises selecting items from an item ratings profile of the target user, said items ratings profile based on explicit ratings of items by the target user.

7. The method of claim 3, wherein selecting the first plurality of items additionally comprises selecting items from an item ratings profile of the target user, said items ratings profile based on explicit ratings of items by the target user.

8. The method of claim 7, wherein generating the personalized item recommendations further comprises giving different amounts of weights to different items in the items ratings profile of the target user, said amounts of weight being dependent upon the explicit ratings of the corresponding items.