SERVICE FOR PROVIDING ITEM RECOMMENDATIONS

• US 20080243632A1
• Filed: 03/30/2007
• Published: 10/02/2008
• Est. Priority Date: 03/30/2007
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Srinivasan, ShenbagaKumaran, Tavis, Matthew T., Smith, Brent R., Miller, Jocelyn M., Kane, Francis J.
• Granted Patent: US 7,881,984 B2
• Time in Patent Office: Days
• US Class Current: 705/26
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/10 Multimedia information
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A service is disclosed for enabling web sites and other entities to provide item recommendations and other behavior-based content to end users. The service can be implemented as a web service that is remotely accessible over the Internet. Web sites use the web service'"'"'s interface to report events descriptive of item-related actions performed by end users (e.g., item views, item purchases, searches for items, etc.). The web service analyzes the reported event data on an aggregated basis to detect various types of associations between particular items, and stores resulting datasets that map items to associated items. The web service'"'"'s interface also provides various API calls for enabling the web sites to request item recommendations and other behavior-based content, including but not limited to personalized recommendations that are based on the event history of the target user. Advantageously, the web sites need not host the infrastructure for providing such content.

Claims

1. A system that provides a web service that is accessible over a network to a plurality of web service clients, the system comprising:

a computer system that implements a web service interface for communicating with the web service clients, the web service interface including at least one event reporting application program interface (API) used by the web service clients to report events to the web service, said events generated by particular end users and corresponding to particular items, the web service interface further including at least one recommendations API which enables the web service clients to request personalized item recommendations for particular end users;

an event data repository that stores event data descriptive of the events reported by the web service clients, wherein the event data repository records the reported events in association with the items, end users, and web service clients to which such events correspond;

an association mining system that analyzes the event data reported by each web service client to detect behavior-based associations between particular items, wherein the association mining component generates item-association datasets that correspond to particular web service clients, said item-association datasets reflecting the detected behavior-based associations; and

a recommendation engine that generates said personalized item recommendations in response to requests received by the computer system via the recommendations API, the recommendation engine being responsive to a request from a web service client for recommendations for a specified end user by using an event history of the specified end user, in combination with an item-association dataset corresponding to the requesting web service client, to select a set of items to recommend to the specified end user.

2. The system of claim 1, wherein the web service interface additionally includes a bulk upload API that enables a web service client to perform a bulk upload of event data to concurrently report a plurality of past events to the web service.

3. The system of claim 1, wherein the web service interface implements an additional recommendations API that enables a web service client to pass a list of two or more source items to the web service and to obtain a corresponding list of recommended items.

4. The system of claim 1, wherein the web service interface includes an authentication component that authenticates requests received from the web service clients via said APIs.

5. The system of claim 1, wherein the event data repository stores event timestamps associated with particular events, said event timestamps reported by the web service clients via the event reporting API.

6. The system of claim 5, wherein the recommendation engine uses the event timestamps to give different amounts of weight to different items included in the event history of the end user.

7. The system of claim 1, wherein the event data repository stores event type identifiers associated with particular events, said event type identifiers reported by the web service clients via the event reporting API.

8. The system of claim 7, wherein the association mining system uses the event type identifiers to generate, for a given web service client, a plurality of item-association datasets, each of which corresponds to a particular type of event.

9. The system of claim 7, wherein the recommendations API includes an event type parameter that enables a requesting web service client to request recommendations that are specific to a particular event type.

10. The system of claim 1, wherein the recommendations API includes an item type identifier that enables a requesting web service client to request recommendations that are specific to a particular type of item.

11. The system of claim 1, wherein the event data repository stores event data descriptive of search query submission events of end users, and the association mining system analyzes the search query event data, in combination with event data descriptive of item selection actions, to generate item-association datasets that map particular search queries to particular items that are related to such search queries.

12. The system of claim 1, wherein the web service interface further includes at least one API that enables a web service client to specify a persistent stop list identifying items to be excluded from the item recommendations, and wherein the system uses the stop list to filter out item recommendations generated by the recommendation engine.

13. The system of claim 1, further comprising a web services description language (WSDL) file represented in computer storage, said WSDL file describing the web service interface, and being configured for use in generating client-side code for interacting with the web service.

14. The system of claim 1, wherein the web service interface includes a billing component that causes operators of said web service clients to be charged for at least some types of requests.

15. The system of claim 1, in combination with a web site that acts as one of said web service clients, said web site being remote from said computer system.

16. The system of claim 15, wherein the web site is configured to report events via the event reporting API in response to browsing actions of end users of the web site, and is further configured to request end-user-specific item recommendations via the recommendations API, and to dynamically incorporate said item recommendations into web pages.

17. A computer-implemented method, comprising:

receiving requests from each of a plurality of clients to store event data descriptive of user-generated events, at least some of the event data comprising identifiers of particular items and particular users to which the events correspond, the requests being received over a computer network via a service interface;

storing the event data received via said requests in a data repository in association with the clients to which such event data corresponds;

generating and storing at least one item-association dataset for each client, wherein each item-association dataset is based on event data reported by a particular client, and identifies item-to-item associations detected from collectively analyzing the event data of a plurality of users;

receiving, via the service interface, a request from a client for item recommendations for a specified user; and

responding to said request by returning item recommendations based on information in the stored event data associated with the specified user, and being further based on an item-association dataset corresponding to said client.

18. The method of claim 17, wherein responding to said requests by returning item recommendations comprises:

(a) identifying a plurality of source items represented in the stored event data associated with the specified user, (b) using the plurality of source items, in combination with the item-association dataset, to select a plurality of additional items to recommend to the specified user, and (c) returning a list of at least some of the additional items to the web service client.

19. The method of claim 17, wherein at least some of the clients are web sites.

20. The method of claim 17, wherein the service interface is a web service interface.

21. The method of claim 20, further comprising transmitting a web services description language (WSDL) file to an operator of at least one of the clients, said WSDL file describing the web service interface, being configured for use in generating client-side code for interacting with the web service interface.

22. The method of claim 17, further comprising authenticating at least some of said requests.

23. The method of claim 17, wherein the event data comprises event type identifiers that specify at least two different types of events, and wherein the method comprises generating, for client, a first item-association dataset that is specific to a first event type, and a second item-association dataset that is specific to a second event type.

24. The method of claim 23, wherein said request for recommendations includes an event type identifier, and the method comprises using the event type identifier to select a corresponding item-association dataset for generating the item recommendations.

25. The method of claim 17, wherein the event data comprises item type identifiers, and the method comprises using said item type identifiers to limit the item recommendations to an item type specified in said request for item recommendations.

26. The method of claim 17, wherein said request for recommendations includes an event type identifier, and the method comprises using said event type identifier to select, from the specified user'"'"'s event data, items that correspond to the event type identifier.

27. The method of claim 17, wherein the event data comprises event timestamps reported by the clients, and wherein the method comprises taking said event timestamps into consideration in selecting items to recommend.

28. The method of claim 17, further comprising receiving a bulk upload file from a client, and storing event data contained therein in said data repository, wherein the bulk upload file contains event data descriptive of a plurality of past events.

29. The method of claim 17, further comprising receiving a stop list from said client via the service interface, and using the stop list to filter out at least some items selected to recommend.

30. A computer system programmed to perform the method of claim 17.

31. A computer-readable medium having stored thereon executable instructions that embody the method of claim 17.

32. A computer-implemented method, comprising:

receiving a request from a remote web site for item recommendations, said request including an identifier of a user of the web site, and being received via a service interface; and

in response to the request, identifying a plurality of items represented in an event history of said user, and using said plurality of items, in combination with an item-association dataset associated with said web site, to generate and return personalized item recommendations for the user.

33. The method of claim 32, wherein the method comprises receiving the request via a web service interface.

34. The method of claim 32, wherein the request includes an event type identifier, and the method comprises using the event type identifier to select said plurality of items from the event history of the user.

35. The method of claim 32, wherein the request includes an item type identifier, and the method comprises using the item type identifier to limit the personalized item recommendations to items of a corresponding type.

36. The method of claim 32, further comprising authenticating the request using a signature included with the request.

37. The method of claim 32, further comprising charging an operator of the web site a fee associated with the request.

38. The method of claim 32, further comprising generating the item-association dataset based on user-generated events reported by the web site for a plurality of users.