Method, system, and computer readable medium for identifying erroneous duplicate orders

• US 7,987,118 B1
• Filed: 11/28/2007
• Issued: 07/26/2011
• Est. Priority Date: 05/24/2004
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Jais, Stephane J., Lendvai, Robert, Sandridge, J. Gray, Gupta, Diwakar, Silver, Scott M., Gellman, David I.
• Primary Examiner(s): ALLEN, WILLIAM J
• Application Number:
  US11/946,756
• Time in Patent Office: 1,336 Days
• Field of Search: 705 26- 27, 705 261- 272
• US Class Current: 705/26.81
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing
• CPC Class Codes: G06Q 30/0635 Processing of requisition or of purchase orders
• CPC Class Codes: G06Q 30/0637 Approvals

Abstract

A method, system, and computer-readable medium for assisting in automatically identifying and handling erroneous orders is described. In some situations, an automatic identification is made of received orders from users that are duplicates of one or more other orders recently placed by those users. When orders are identified as being potentially erroneous, fulfillment of those orders may be delayed while automatically querying the users to obtain manual confirmation to continue with the order fulfillment. In other situations, other types of orders are analyzed, orders are identified as being potentially erroneous in other ways, and such orders are handled in ways other than based on querying for a manual confirmation or rejection response. This abstract is provided to comply with rules requiring an abstract, and it is submitted with the intention that it will not be used to interpret or limit the scope or meaning of the claims.

Claims

1. A computer-implemented method for identifying erroneous duplicate orders, the method comprising:

receiving an indication of a current order placed by a user;

automatically determining whether the current order is a duplicate with one or more prior orders that were previously placed by the user, the automatic determining being performed by one or more configured computing systems in a manner that is personalized to the user, wherein the manner personalized to the user differs from one or more other manners for automatically determining if other orders placed by other users are duplicates;

when it is determined that the current order is not a duplicate, automatically proceeding to fulfill the current order without further intervention by the user, the automatic proceeding being performed by the one or more configured computing systems; and

when it is determined that the current order is a duplicate, querying the user to obtain manual confirmation before proceeding to fulfill the current order.

2. The method of claim 1 wherein the one or more prior orders are orders that were previously placed within a specified prior period of time.

3. The method of claim 1 wherein the one or more prior orders are orders previously placed by the user that satisfy one or more specified criteria.

4. The method of claim 1 including, when a manual response from the user confirms the current order, proceeding to fulfill the current order without further user intervention.

5. The method of claim 1 including, when a manual response from the user rejects the current order, canceling the current order.

6. The method of claim 1 including analyzing manual feedback from one or more users regarding one or more orders placed by those users in order to identify aspects of the orders that assist in predicting whether future orders are potentially erroneous, and wherein future automatic determining of whether an order is a duplicate is based on comparing aspects of the order to the identified order aspects.

7. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders includes comparing types of information in the current and prior orders that are selected in a manner specific to the user.

8. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders includes selecting the prior orders in a manner specific to the user.

9. The method of claim 1 wherein orders may be placed by users using multiple distinct types of ordering processes, and wherein the automatic determining of whether the current order is a duplicate with one or more prior orders includes comparing the types of ordering processes used to place the current and prior orders.

10. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders in a manner specific to the user is based at least in part on analyzing prior behavior of the user.

11. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders in a manner specific to the user is based at least in part on one or more preferences previously indicated by the user.

12. The method of claim 1 wherein the method is performed for each of multiple current orders received from multiple users, and wherein a second current order that is identical to the current order but is placed by a second user is not determined to be a duplicate.

13. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders is performed in a manner specific to a group of multiple related users that includes the user and that is automatically identified based on behavior of those users.

14. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders is performed in a manner specific to a group of multiple related users that includes the user and that is automatically identified based on characteristics of those users.

15. The method of claim 1 wherein the querying of the user to obtain manual confirmation is performed automatically by the one or more configured computing systems in a manner specific to the user.

16. The method of claim 1 including maintaining information about orders that were placed within a specified prior period of time, and wherein the prior orders are at least some of the orders whose information is maintained.

17. The method of claim 16 wherein the maintaining is performed repeatedly such that the orders whose information is maintained are orders placed within a specified prior period of time relative to a current time.

18. The method of claim 16 including updating the maintained information to reflect changes to orders by customers after those orders were placed.

19. The method of claim 1 wherein the determining of whether the current order is a duplicate with one or more prior orders includes comparing a subset of information included as part of each of the orders.

20. The method of claim 1 wherein the determining of whether the current order is a duplicate with one or more prior orders includes comparing a manner of placing the orders.

21. The method of claim 1 wherein the determining of whether the current order is a duplicate with one or more prior orders includes determining whether a degree of similarity between the current order and one or more of the prior orders exceeds a specified threshold.

22. The method of claim 1 wherein the determining of whether the current order is a duplicate with one or more prior orders is part of determining whether the current order is potentially erroneous, and including querying the user to obtain manual confirmation before proceeding to fulfill the current order any time that it is determined that the current order is potentially erroneous.

23. The method of claim 1 including automatically determining whether another current order that has not yet been placed by a user is a duplicate with one or more prior orders that were previously placed by that user and, when it is determined that the another current order is a duplicate, warning the user that the another current order is a duplicate.

24. The method of claim 1 wherein the automatic determining in the manner personalized to the user of whether the current order is a duplicate is based at least in part on feedback obtained from the user related to one or more prior orders.

25. The method of claim 1 wherein the automatic determining of whether the current order is a duplicate in the manner personalized to the user is based at least in part on one or more prior interactions with the user.

26. The method of claim 25 wherein the one or more prior interactions with the user with which the automatic determining of whether the current order is a duplicate is personalized to the user include obtaining responses by the user to prior querying for confirmations to proceed with prior orders by the user that were automatically determined to be duplicates.

27. The method of claim 25 wherein the one or more prior interactions with the user with which the automatic determining of whether the current order is a duplicate is personalized to the user include obtaining preferences specified by the user regarding how to perform future automatic determining of whether an order of the user is a duplicate.

28. The method of claim 1 further comprising, before the automatic determining in the manner personalized to the user of whether the current order is a duplicate, automatically identifying the personalized manner for the user based at least in part on modifying a default manner of determining whether a current order for a user is a duplicate.

29. The method of claim 1 wherein the automatic determining in the manner personalized to the user of whether the current order is a duplicate includes determining that the current order is a duplicate, and wherein the method further comprises:

receiving an indication of a second current order placed by a distinct second user; and

automatically determining whether the second current order is a duplicate with one or more prior orders that were previously placed by the second user, the automatic determining for the second current order being performed in a manner that is distinct from the manner personalized to the user such that the second current order is not determined to be a duplicate, the second current order being substantially identical to the current order, and the one or more prior orders that were previously placed by the second user being substantially identical to the one or more prior orders that were previously placed by the user.

30. The method of claim 1 wherein the automatic determining in the manner personalized to the user of whether the current order is a duplicate includes determining that the current order is not a duplicate, and wherein the method further comprises:

receiving an indication of a second current order placed by a distinct second user; and

automatically determining whether the second current order is a duplicate with one or more prior orders that were previously placed by the second user, the automatic determining for the second current order being performed in a manner that is distinct from the manner personalized to the user such that the second current order is determined to be a duplicate, the second current order being substantially identical to the current order, and the one or more prior orders that were previously placed by the second user being substantially identical to the one or more prior orders that were previously placed by the user.

31. A computing device for identifying potentially erroneous orders, comprising:

one or more processors;

an erroneous order identifier component that is configured to, when executed by at least one of the one or more processors, determine whether orders placed by users are potentially erroneous based at least in part on actions previously taken by the users regarding other orders, the determining being performed for each of at least some of the users in a manner that is personalized to that user based at least in part on prior interactions with that user so as to differ from one or more manners used for other users for determining if other orders placed by those other users are potentially erroneous; and

a user notifier component that is configured to, when executed by at least one of the one of more processors, and when it is determined that an order placed by a user is potentially erroneous, notify the user of the determining for the order so that the user can provide a response regarding an action to take for the order.

32. The computing device of claim 31 wherein the determining of whether an order for a user is potentially erroneous includes determining that the order is potentially erroneous if the order is a duplicate of a prior order that the user previously placed.

33. The computing device of claim 31 wherein the determining of whether an order for a user is potentially erroneous includes determining that the order is potentially erroneous if it is sufficiently likely that the user did not intend to place the order.

34. The computing device of claim 31 wherein the determining of whether an order for a user is potentially erroneous includes determining that the order is potentially erroneous if it is sufficiently likely that placing of the order by the user was a mistake.

35. The computing device of claim 31 wherein the notifying of a user of the determining for an order includes querying the user and obtaining a manual confirmation response from the user to fulfill the order.

36. The computing device of claim 31 further comprising a user feedback handler component that is configured to analyze feedback from one or more users regarding one or more orders for those users in order to identify aspects of the orders that assist in predicting whether future orders are potentially erroneous.

37. The computing device of claim 31 wherein the erroneous order identifier component and the user notifier component each includes software instructions for execution by the computing device.

38. The computing device of claim 31 wherein the erroneous order identifier component consists of a means for determining whether orders for users are potentially erroneous based at least in part on actions previously taken by the users regarding other orders, and wherein the user notifier component consists of a means for notifying the user of the determining for the order so that the user can provide a response regarding an action to take for the order after it is determined that an order for a user is potentially erroneous.

39. A non-transitory computer-readable medium whose contents configure a computing system to identify potentially erroneous orders, by performing a method comprising:

automatically determining whether an order placed by a user is potentially erroneous based at least in part on actions previously taken by the user regarding other prior orders, the determining being performed by the configured computing system in a manner that is personalized to the user based at least in part on the previously taken actions, wherein the personalized manner for the user differs from one or more other manners used for other users for determining if distinct orders placed by those other users are potentially erroneous; and

when it is automatically determined that the order placed by the user is potentially erroneous, automatically notifying the user of the determining for the order to enable the user to provide a response regarding an action to take for the order.

40. The non-transitory computer-readable medium of claim 39 wherein the automatic determining of whether the order placed by the user is potentially erroneous includes determining that the order is potentially erroneous if the order is a duplicate of a prior order that the user previously placed.

41. The non-transitory computer-readable medium of claim 40 wherein the automatic notifying of the user of the determining for the order includes querying the user and obtaining a manual confirmation response from the user to fulfill the order, and wherein the method further comprises automatically initiating fulfillment of the order placed by the user based on the obtained manual confirmation response.

42. The non-transitory computer-readable medium of claim 41 wherein the querying of the user is performed automatically by the configured computing system in a manner specific to the user.

43. The non-transitory computer-readable medium of claim 39 wherein the other prior orders are orders that were previously placed by the user within a specified prior period of time.

44. The non-transitory computer-readable medium of claim 39 wherein the other prior orders are orders previously placed by the user that satisfy one or more specified criteria.

45. The non-transitory computer-readable medium of claim 39 wherein the method further comprises receiving the response from the user and, if the response from the user rejects proceeding with the order, canceling the order.

46. The non-transitory computer-readable medium of claim 39 wherein the method further comprises, before the automatic determining of whether the order placed by the user is potentially erroneous, analyzing manual feedback from one or more users regarding one or more orders placed by those users in order to identify aspects of the orders that assist in predicting whether future orders are potentially erroneous, and wherein the automatic determining of whether the order placed by the user is potentially erroneous is based on comparing aspects of the order to the identified order aspects.

47. The non-transitory computer-readable medium of claim 39 wherein the automatic determining of whether the order is potentially erroneous based at least in part on the actions previously taken by the user regarding the other prior orders includes multiple of comparing types of information in the order and in the other prior orders in a manner specific to the user, of selecting the other prior orders in a manner specific to the user, of comparing in a manner specific to the user one or more of multiple alternative types of ordering processes used to place the order and the other prior orders, and of using one or more preferences previously indicated by the user.

48. The non-transitory computer-readable medium of claim 39 wherein the automatic determining of whether the order placed by the user is potentially erroneous includes automatically analyzing the previously taken actions of the user.

49. The non-transitory computer-readable medium of claim 39 wherein the determining of whether the order is potentially erroneous includes one or more of comparing a subset of information included in the order and in the other prior orders, of comparing a manner of placing the order and the other prior orders, and of determining whether a degree of similarity between the order and one or more of the prior other orders exceeds a specified threshold.

50. The non-transitory computer-readable medium of claim 39 wherein the method further comprises automatically determining whether a distinct potential order that has not yet been placed by the user is potentially erroneous based on one or more of the other prior orders and, if it is determined that the potential order is potentially erroneous, warning the user that the potential order is potentially erroneous.

51. The non-transitory computer-readable medium of claim 39 wherein the automatic determining in the manner personalized to the user of whether the order is potentially erroneous is based at least in part on feedback obtained from the user related to one or more of the other prior orders.

52. The non-transitory computer-readable medium of claim 39 wherein the previously taken actions by the user include providing responses by the user to prior querying for confirmations to proceed with prior orders by the user that were automatically determined to be potentially erroneous.

53. The non-transitory computer-readable medium of claim 39 wherein the previously taken actions by the user include specifying preferences of the user regarding how to perform future automatic determining of whether an order of the user is potentially erroneous.

54. The non-transitory computer-readable medium of claim 39 wherein the method further comprises, before the automatic determining in the manner personalized to the user of whether the order is potentially erroneous, automatically identifying the personalized manner for the user based at least in part on modifying a default manner of determining whether a current order for a user is potentially erroneous.

55. The non-transitory computer-readable medium of claim 39 wherein, if it is automatically determined that the order placed by the user is not potentially erroneous, automatically initiating fulfillment of the order placed by the user without further intervention by the user, the automatic initiating of the fulfillment of the order being performed by the configured computing system.

56. The non-transitory computer-readable medium of claim 39 wherein the computer-readable medium is a memory of the configured computing system, and wherein the contents are instructions that when executed program the configured computing system to perform the method.

57. A computer-implemented method for identifying erroneous duplicate orders, the method comprising:

receiving an indication of a current order placed by a user;

automatically determining whether the current order is a duplicate with one or more prior orders that were previously placed by the user, the automatic determining being performed by one or more configured computing systems;

if it is determined that the current order is not a duplicate, automatically proceeding to fulfill the current order without further intervention by the user, the automatic proceeding being performed by the one or more configured computing systems; and

if it is determined that the current order is a duplicate,

querying the user to obtain manual confirmation before proceeding to fulfill the current order; and

after obtaining the manual confirmation from the user for the current order, automatically adjusting how a future automatic determination of whether an order placed by the user is identified as a duplicate will be performed to reflect a lower likelihood that the user has erroneously placed that order, the automatic adjusting being performed by the one or more configured computing systems.

58. The method of claim 57 wherein the automatic adjusting of how a future automatic determination will be performed of whether a future order for the user is identified as a duplicate includes modifying which types of previously placed prior orders are used for comparison to the future order.

59. The method of claim 57 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders that were previously placed by the user is based at least in part on one or more predetermined criteria for comparing the current order and the one or more prior orders, and wherein the automatic adjusting of how a future automatic determination will be performed of whether a future order for the user is identified as a duplicate includes modifying the criteria used for determining whether the future order is a duplicate with one or more previously placed prior orders.

60. The method of claim 57 wherein the automatic determining of whether the current order is a duplicate with one or more prior orders that were previously placed by the user includes selecting the one or more prior orders based at least in part on prior orders that were placed during a period of time of a predetermined length prior to the current order, and wherein the automatic adjusting of how a future automatic determination will be performed of whether a future order for the user is identified as a duplicate includes modifying the length of the period of time such that the prior orders used for comparison to the future order are prior orders previously placed during a period of time of the modified length prior to the future order.

61. The method of claim 57 wherein the automatic adjusting of how a future automatic determination will be performed of whether a future order for the user is identified as a duplicate includes modifying a manner of querying the user to obtain manual confirmation for the future order.

62. The method of claim 57 wherein the automatic adjusting of how a future automatic determination will be performed of whether a future order for the user is identified as a duplicate includes determining not to query the user to obtain manual confirmation for the future order.

63. The method of claim 57 further comprising, after receiving a manual cancellation of the current order from the user in response to the querying, automatically adjusting how a future automatic determination of whether an order placed by the user is identified as a duplicate will be performed so as to reflect a higher likelihood that the user has erroneously placed that order.

64. A computer-implemented method for identifying erroneous duplicate orders, the method comprising:

receiving an indication of a current order placed by a user;

automatically determining whether the current order is a duplicate with one or more prior orders that were previously placed by the user, the automatic determining being performed by one or more configured computing systems;

if it is determined that the current order is not a duplicate, automatically proceeding to fulfill the current order without further intervention by the user, the automatic proceeding being performed by the one or more configured computing systems; and

if it is determined that the current order is a duplicate,

querying the user to obtain manual confirmation before proceeding to fulfill the current order; and

after obtaining the manual confirmation from the user for the current order, automatically adjusting how a future automatic determination of whether any order for any user is identified as a duplicate will be performed to reflect a lower likelihood that that user has erroneously placed that order, the automatic adjusting being performed by the one or more configured computing systems.