Computer services for assisting users in locating and evaluating items in an electronic catalog based on actions performed by members of specific user communities

• US 6,963,850 B1
• Filed: 08/19/1999
• Issued: 11/08/2005
• Est. Priority Date: 04/09/1999
• Status: Expired due to Term

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon.com, Inc.
• Inventors: McAuliffe, Jon, Spiegel, Joel R., Bezos, Jeffrey P.
• Primary Examiner(s): Cogg, Wynn W.
• Assistant Examiner(s): ZURITA, JAMES H
• Application Number:
  US09/377,447
• Time in Patent Office: 2,273 Days
• Field of Search: 705/26, 705/27, 705/25, 705/10, 705/35, 705/37, 704/13
• US Class Current: 705/7.29
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0613 Third-party assisted
• CPC Class Codes: G06Q 30/0623 Item investigation
• 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: G06Q 30/0645 Rental transactions; Leasing transactions
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A Web based system provides informational services for assisting customers in selecting products or other types of items from an electronic catalog of a merchant. Users of the system can create and join user communities, such as communities based on user hobbies, localities, professions, and organizations. The system also supports implicit membership communities that are based on email addresses (e.g., all users having a “nasa.com” email address), shipping/billing addresses, and other known user information. Using purchase history data collected for online users, the system automatically identifies and generates lists of the most popular items (and/or items that are becoming popular) within particular communities, and makes such information available to users for viewing. For example, in the context of an online book store users of the nasa.com community may automatically be presented a Web page which lists the bestselling book titles among nasa.com users, or may be sent email notifications of purchase events or hotselling books within the community. Another feature involves automatically notifying users interested in particular products of other users (preferably other members of the same community) that have purchased the same or similar products. For example, in one embodiment, when a user accesses a book detail page, the detail page is customized to include the names and email addresses of other members of the user'"'"'s community that recently purchased the same book.

Claims

1. A method of assisting users in selecting items from an electronic catalog of items, the catalog accessible to users of an online sales system that provides services for allowing users to purchase items from the catalog, the method comprising:

providing a database which contains information about a plurality of user communities, wherein different communities represent different subsets of users of the sales system;

tracking, via computer, online purchases of items from the sales system by the users to generate purchase history data, and storing the purchase history data in a computer memory;

processing, via computer, at least the purchase history data to identify at least one item which, based on actions of both members and non-members of a selected community of said plurality of user communities, has a popularity level that is substantially greater within the selected community than outside the selected community; and

electronically notifying users that the at least one item is popular within the selected community to assist users in selecting items from the electronic catalog.

2. The method of claim 1, wherein electronically notifying users comprises generating a Web page which includes a community-based most popular items list.

3. The method of claim 2, wherein the most popular items list is a bestsellers list.

4. The method of claim 1, wherein electronically notifying users comprises automatically generating and sending an email message to members of the selected community.

5. The method of claim 4, wherein the email message contains contact information of at least one member of the community that has purchased an item described in the email message.

6. The method of claim 1, wherein processing the purchase history data to identify at least one item comprises identifying a set of characterizing purchases for the community.

7. The method of claim 1, wherein the community is an implicit membership community.

8. The method of claim 7, wherein the community is based on email addresses of users.

9. The method of claim 1, wherein the community is an explicit membership community.

10. The method of claim 1, wherein the community is derived from an electronic address book of a user.

11. The method of claim 1, wherein the community is a composite community which comprises multiple other communities of the database.

12. The method as in claim 1, wherein processing the purchase history data comprises excluding substantially, from consideration purchase history data falling outside a selected look-back period, so that calculated item popularity levels strongly reflect current interests of users.

13. A system for assisting users of an online sales system in selecting items from an electronic catalog of items, the system comprising:

at least one electronic database which contains purchase history data for users of the sales system, and which contains information about a plurality of user communities wherein different communities represent different subsets of users of the sales system; and

a computer system that executes a process which analyzes at least the purchase history data to identify items that have substantially higher popularity levels within particular communities of the plurality of communities relative to their respective popularity levels among a general user population, and which notifies users of the sales system of the items and associated communities for which such popularity level disparities exist, to thereby assist such users in selecting items from the electronic catalog.

14. The system of claim 13, wherein the process comprises a first process which generates a table which contains bestselling items list for at least some of the communities, and a second process which selects items from the table to display to users.

15. The system of claim 13, further comprising a user interface which allows users to select and join at least some of the user communities.

16. The system of claim 13, further comprising a user interface which allows a user to define a composite community that includes multiple communities of the database, and to initiate generation of a popular items list for the composite community.

17. The system of claim 13, wherein at least some of the communities are implicit membership communities.

18. The system of claim 13, wherein at least some of the communities are based on email addresses of users.

19. The system of claim 13, wherein at least some of the communities are based on electronic address books of the users.

20. The system of claim 13, wherein the process generates and displays community bestsellers lists for at least some of the communities.

21. The system of claim 13, wherein the process compares a popularity of an item within a community to a popularity of the item among non-members of the community.

22. The system of claim 13, wherein the process sends to the users notification emails that include descriptions of the items that are popular within particular communities.

23. The system of claim 22, wherein at least some of the notification emails include contact information of users that have purchased items described therein.

24. The system of claim 22, wherein at least some of the notification emails specify a level of acceptance an item has attained within a particular community.

25. The system of claim 13, wherein the process identifies items that are have substantially higher popularity levels within particular communities by at least identifying a set of items purchased by members of a community that distinguishes the community from a general user population.

26. The system of claim 25, wherein the process uses a censored chi-square algorithm to identify the set of items.

27. The system of claim 13, wherein the computer process excludes, substantially, from consideration purchase history data falling outside a selected look-back period.

28. A method of assisting users in selecting items from an electronic catalog of items, the catalog accessible to users of an online sales system that provides services for allowing users to purchase items from the catalog, the method comprising steps, each step performed by computer, of:

identifying a subset of users of the online sales system that have email addresses that satisfy a particular criterion, wherein the subset comprises a plurality of users;

identifying at least one item that characterizes the subset of users, wherein the step of identifying comprises

processing purchase history data of the subset of users and of users falling outside said subset; and

electronically notifying users of the online sales system of a relationship between the at least one item and the subset of users, to assist users in selecting items from the electronic catalog.

29. The method of claim 28, wherein identifying a subset of users comprises identifying all users having a selected domain name within their respective email addresses.

30. The method of claim 29, wherein the selected domain name is a domain name of a selected company.

31. The method of claim 28, wherein identifying a subset of users comprises identifying all users that have one of a selected group of domain name within their respective email addresses.

32. The method of claim 28, wherein electronically notifying comprises generating a Web page which includes a list of items that characterize the subset of users.

33. The method of claim 28, wherein electronically notifying comprises sending email notification messages to at least some of the users of the subset.

34. The method of claim 28, wherein identifying at least one item that characterizes the subset of users comprises comparing a popularity level of the item among the subset of users to a popularity level of the item among a general population of users.

35. The method of claim 28, wherein identifying at least one item that characterizes the subset of users comprises identifying a set of items that distinguish the subset from a general population of users.

36. The method as in claim 35, wherein identifying a set of items that distinguish the subset from a general population of users comprises applying a censored chi-square type algorithm to the purchase history data.

37. The method as in claim 28, wherein identifying at least one item that characterizes the subset of users comprises at least one of comparing a velocity of an item within the community to a velocity of the item within a general user population;

and comparing an acceleration of the item within the community to an acceleration of the item within a general population.

38. A method of assisting users in selecting items from a catalog of items, comprising:

electronically identifying a subset of a general population of users;

tracking, via computer, at least one type of user activity that indicates user affinities for particular items of the catalog to generate activity history data;

processing, via computer, the activity history data of the general population of users, including the subset of users, to identify a set of items that distinguish the subset from the general population; and

electronically notifying users of the catalog of a relationship between the set of items and the subset of users, to thereby assist the users of the catalog in selecting items therefrom.

39. The method of claim 38, wherein processing the activity history data comprises processing purchase history data, and the set of items consists essentially of items purchased by members of the subset.

40. The method of claim 38, wherein tracking at least one type of user activity comprises tracking item viewing events.

41. The method of claim 38, wherein processing the activity history data comprises applying a censored chi-square algorithm to the activity history data.

42. The method of claim 38, wherein the subset consists of members of an implicit-membership community.

43. A method of assisting users in selecting items from an electronic catalog of items, comprising:

electronically identifying among a population of users of the electronic catalog, a subset of users that, based on user shipping addresses, reside in a common geographic region;

electronically monitoring actions of the subset of users, and of users falling outside the subset, to generate user activity history data reflective of user actions performed with respect to items in the electronic catalog;

electronically processing the user activity history data to identify an item whose popularity level among the subset of users is substantially higher than a popularity level of the item among the population of users; and

electronically notifying at least one user that a relationship exists between the item and the geographic region to assist the at least one user in selecting items from the electronic catalog.

44. The method as in claim 43, wherein processing the user activity history data to identify an item whose popularity level among the subset of users is substantially higher than a popularity level of the item among the population of users comprises at least one of (a) comparing a velocity of the item within the subset to a velocity of the item within a general user population;

and (b) comparing an acceleration of the item within the subset to an acceleration of the item within a general user population.

45. The method as in claim 43, wherein processing the user activity history data to identify an item whose popularity level among the subset of users is substantially higher than a popularity level of the item among the population of users comprises applying a censored chi-square type algorithm to the user activity history data.

46. The method as in claim 43, wherein notifying at least one user comprises presenting to the user a list of items that are popular in the geographic region relative to popularity levels outside the region.

47. The method as in claim 43, wherein the user actions performed with respect to items in the electronic catalog comprise item purchases.

48. The method as in claim 43, wherein the user actions performed with respect to items in the electronic catalog comprise item viewing events.

49. The method as in claim 43, wherein the geographic region corresponds to a particular city.

50. The method as in claim 43, wherein processing the user activity history data comprises applying a time window to the user activity history data such that a popularity level of the item is measured substantially from a most recent set of the user activity history data.

51. A computer-implemented method of assisting users in selecting items from an electronic catalog of items, the method comprising:

storing email addresses of users of the electronic catalog;

electronically identifying, for each of a plurality of organizations, a respective group of said users who are deemed to be affiliated with the respective organization by virtue of a domain name of the respective organization occurring within an email address of each such user, to thereby identify a plurality of organization-specific groups of users;

electronically tracking at least one type of user activity that indicates user affinities for particular items in the electronic catalog to generate activity history data for the users;

electronically analyzing for each organization-specific group of users, the activity history data to generate a list of items within the electronic catalog that distinguish the organization-specific group from a general population of the users, to thereby generate a plurality of organization-specific popular items lists; and

electronically exposing the organization-specific popular items lists to the users of the electronic catalog in association with the organizations to which such popular items lists correspond, to allow users to identify items that are popular within specific organizations.

52. The method as in claim 51, wherein exposing the organization-specific popular items lists to the users comprises providing to a target user a personalized web page that displays organization-specific popular items lists of organizations with which the target user is affiliated.

53. The method as in claim 51, wherein analyzing the activity history data comprises excluding, substantially, from consideration activity history data falling outside a selected look-back period.

54. The method as in claim 51, wherein analyzing the activity history data comprises applying a censored chi-square type algorithm to the activity history data.

55. The method as in claim 51, wherein at least some of the organizations are companies.

56. A system for assisting users in selecting items from an electronic catalog of items, the system comprising:

at least one electronic database which contains email addresses of users of the electronic catalog; and

a computer system that executes a computer program to

identify, for each of a plurality of organizations, a respective group of said users who are deemed to be affiliated with the respective organization by virtue of a domain name of the respective organization occurring within an email address of each such user, to thereby identify a plurality of organization-specific groups of users;

track at least one type of user activity that indicates user affinities for particular items in the electronic catalog to generate activity history data for the users;

analyze, for each organization-specific group of users, the activity history data to generate a list of items within the electronic catalog that distinguish the organization-specific group from a general population of the users, to thereby generate a plurality of organization-specific popular items lists; and

expose the organization-specific popular items lists to the users of the electronic catalog in association with the organizations to which such popular items lists correspond, to allow users to identify items that are popular within specific organizations.