Identifying the items most relevant to a current query based on items selected in connection with similar queries

• US 20050177569A1
• Filed: 04/15/2005
• Published: 08/11/2005
• Est. Priority Date: 03/03/1998
• Status: Active Grant

• Current Assignee: Dwayne Bowman, Joel R. Spiegel
• Original Assignee: Dwayne Bowman, Joel R. Spiegel
• Inventors: Spiegel, Joel R., Bowman, Dwayne
• Granted Patent: US 7,305,390 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/3349 Reuse of stored results of previous queries
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: Y10S 707/913 Multimedia
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

The present invention provides a software facility for identifying the items most relevant to a current query based on items selected in connection with similar queries. In preferred embodiments of the invention, the facility receives a query specifying one or more query terms. In response, the facility generates a query result identifying a plurality of items that satisfy the query. The facility then produces a ranking value for at least a portion of the items identified in the query result by combining the relative frequencies with which users selected that item from the query results generated from queries specifying each of the terms specified by the query. The facility identifies as most relevant those items having the highest ranking values.

Claims

1-20. -20. (canceled)

21. A method in a computing system for ranking items in a search result based on relevance of the items, the method comprising:

receiving a search query from a user that specifies one or more terms;

generating a query result identifying a plurality of items satisfying the query;

ranking multiple of the identified items by,

identifying one or more demographic groups to which the user belongs;

for each of the multiple items, generating a rating for the item based on a frequency with which other users in the identified demographic groups purchased the item during a prior period of time; and

generating a ranking for each of the multiple items relative to the other multiple items based on the generated ratings; and

displaying the multiple items to the user in an order based on the generated ranking for those items,

so that items purchased most frequently by other users in a demographic group to which a user belongs are displayed to the user before items purchased less frequently by those other users.

22. The method of claim 21 including, before the ranking of the multiple identified items, tracking frequencies with which users in multiple predefined demographic groups purchase items during one or more periods of time.

23. The method of claim 22 wherein the tracking of the frequencies with which users in the predefined demographic groups purchase items includes, for each of multiple other queries that are each from a user and each specify one or more terms:

identifying one or more of the predefined demographic groups to which the user belongs;

providing an indication to the user of a search result that includes items satisfying the query; and

for each of the included items that the user purchases and for each of the identified predefined demographic groups, increasing for each of the specified terms a rating value corresponding to a combination of the purchased item and the predefined demographic group for a specified period of time.

24. The method of claim 21 wherein the item purchases during the prior period of time that are used for the generating of the rating for an item are purchases of the item by users after the item was presented to the users as part of search results.

25. The method of claim 21 wherein the item purchases during the prior period of time that are used for the generating of the rating for an item are purchases of the item by users after the users select the item in a manner other than from a search result.

26. The method of claim 21 wherein the item purchases during the prior period of time that are used for the generating of the rating for an item are purchases of the item by users after the item was in shopping carts of the users.

27. The method of claim 21 wherein the generating of the ratings for items based on prior purchases is performed so as to weight recent purchases more heavily than less recent purchases.

28. The method of claim 21 wherein the generating of the ratings for items based on prior purchases is performed so as to reflect changes in purchases during the prior period of time from another period of time earlier than the prior period.

29. A method in a computing system for ranking items in a search result, the method comprising:

receiving one or more search terms for a user;

identifying a plurality of items that match at least one of the terms;

identifying one or more demographic groups to which the user belongs;

generating a relevance ranking for each of multiple of the identified items based on a number of selections of that item by users in the identified demographic groups during at least one prior period of time; and

providing an indication of the generated relevance rankings for the multiple items.

30. The method of claim 29 wherein the selections of an item during a prior period of time that are used for the generating of the relevance ranking for the item are each purchases of the item during the prior period of time.

31. The method of claim 30 wherein each of the item purchases during the prior period of time are purchases of an item by a user after the item was presented to the user as part of search results.

32. The method of claim 29 wherein the selections of an item during a prior period of time that are used for the generating of the relevance ranking for the item are each to display additional information about the item during the prior period of time.

33. The method of claim 29 wherein the selections of an item during a prior period of time that are used for the generating of the relevance ranking for the item are each to add the item to a tentative list of purchases during the prior period of time.

34. The method of claim 29 wherein the selections of an item during a prior period of time that are used for the generating of the relevance ranking for the item are each to review information about the item for a length of time during the prior period of time.

35. The method of claim 29 wherein the selections of an item during a prior period of time that are used for the generating of the relevance ranking for the item are each to take a number of actions when reviewing information about the item during the prior period of time.

36. The method of claim 29 wherein the providing of the indication of the generated relevance rankings for the multiple items includes displaying those items in an order based on the generated relevance rankings for those items.

37. The method of claim 29 wherein the generating of the relevance ratings for items based on selections during at least one prior period of time is performed so as to weight recent selections more heavily than less recent selections.

38. The method of claim 29 wherein the generating of the relevance ratings for items based on numbers of selections during at least one prior period of time is performed so as to reflect changes in the numbers of selections during the at least one prior period of time from one or more other earlier periods of time.

39. The method of claim 29 wherein the generating of the relevance ratings for each of the items is further based on the number of selections of the item during multiple prior periods of time, and wherein the generating of the relevance ratings is further performed so as to weight selections in more recent of the multiple prior periods of time more heavily than selections in less recent of the multiple prior periods of time.

40. The method of claim 29 wherein the number of selections of an item during a prior period of time that is used for the generating of the relevance ranking for the item represents a frequency with which the item was selected during the prior period of time.

41. The method of claim 29 including, before the generating of the relevance rankings for the multiple identified items, tracking selections of the items by users in multiple predefined demographic groups during one or more periods of time.

42. The method of claim 41 wherein the tracking of the selections of the items by users in multiple predefined demographic groups during one or more periods of time includes, for each of multiple search queries that are each from a user:

identifying one or more of the predefined demographic groups to which the user belongs;

providing an indication to the user of a search result that includes items satisfying the query; and

for each of the included items that the user selects during the one or more periods of time and for each of the identified predefined demographic groups, increasing a rating value corresponding to a combination of the selected item and the predefined demographic group.

43. The method of claim 29 wherein the identified demographic groups are each based at least in part on an age of users in the group.

44. The method of claim 29 wherein the identified demographic groups are each based at least in part on a sex of users in the group.

45. The method of claim 29 wherein the identified demographic groups are each based at least in part on income of users in the group.

46. The method of claim 29 wherein the identified demographic groups are each based at least in part on geography of users in the group.

48. The method of claim 29 wherein the generating of the relevance ratings for each of multiple of the items identified as matching at least one term for a user is further based on behavioral information about another user.

49. The method of claim 29 wherein the generating of the relevance ratings for each of multiple of the items identified as matching at least one term for a user is further based on preference information for the user.

50. The method of claim 29 wherein each of the items is a book.

51. The computer-readable medium whose contents cause a computing system to rank items in a query result, by performing a method comprising:

for each of multiple items identified for a query from a user, ranking the item based on prior selections of that item by users in one or more selected demographic groups; and

providing an indication of the rankings for the multiple items.

52. The computer-readable medium of claim 51 where in the prior selections of an item are each purchases of the item occurring during a specified prior period of time.

53. The computer-readable medium of claim 52 where each of the item purchases during the prior period of time are purchases of an item by a user after the item was presented to the user as a result for a query.

54. The computer-readable medium of claim 51 wherein the selected demographic groups are demographic groups to which the user belongs.

55. The computer-readable medium of claim 51 wherein the ranking of each of the items based on prior selections of the item is performed so as to weight recent selections more heavily than less recent selections.

56. The computer-readable medium of claim 51 wherein the ranked items are a subset of items selected as results for the query, and wherein the ranking of the items is performed so as to reflect relevance of the items to the query.

57. The computer-readable medium of claim 51 wherein the computer-readable medium is a memory of a computing system.

58. The computer-readable medium of claim 51 wherein the contents include one or more user behavior data structures for use in ranking relevance of items in a query result, the data structure comprising a plurality of rating scores that each correspond to an item and one of multiple demographic groups, each rating score quantitatively reflecting an extent to which users in the corresponding demographic group have selected the item from query results, such that the data structure may be used to rank items in a later query result for a user by retrieving from the data structure at least one rating score corresponding to the item and to a demographic group to which that user belongs.

59. A computing system configured to rank items in a query result, comprising:

a query result ranking component configured to, for each of multiple items identified among query results for a query from a user, generate a rank for the item based on prior selections of that item by users in one or more demographic groups for the user; and

an output device with which the query result ranking component can provide the generated rankings for the multiple items.

60. The computing system of claim 59 wherein the prior selections of an item are each purchases of the item occurring during a specified prior period of time.

61. The computing system of claim 59 wherein the query result ranking component is executing in memory of the computing system.