Neural network system and method for controlling information output based on user feedback

• US 20050086186A1
• Filed: 08/30/2004
• Published: 04/21/2005
• Est. Priority Date: 02/02/1999
• Status: Active Grant

• Current Assignee: Hyper Search LLC (IPValuation Partners, LLC)
• Original Assignee: Thinkalike LLC
• Inventors: Sullivan, Alan, Pope, Ivan
• Granted Patent: US 7,120,615 B2
• Time in Patent Office: Days
• US Class Current: 706/16
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/063 using electronic means
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: H04L 51/212 using filtering or selective blocking
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

A system and method for controlling information output based on user feedback about the information is provided that comprises a plurality of information sources providing information. The information sources may be electronic mail providers, chat participants, or page links. At least one neural network module selects one or more of a plurality of objects to receive information from the plurality of information sources based at least in part on a plurality of inputs and a plurality of weight values during that epoch. At least one server, associated with the neural network module, provides one or more of the objects to a plurality of recipients. The objects may comprise electronic mail messages, chat participants viewers, or slots within a link directory page. The recipients provide feedback about the information during an epoch. At the conclusion of an epoch, the neural network takes all of the feedback that has been provided from the recipients and generates a rating value for each of the plurality of objects. Based on the rating value and the selections made, the neural network redetermines the weight values within the network. The neural network then selects the objects to receive information during a subsequent epoch using the redetermined weight values and the inputs for that subsequent epoch.

Claims

1-38. -38. (canceled)

39. A system for controlling information output based on user feedback about the information comprising:

a plurality of information sources providing information;

an artificial intelligence agent that selects one or more of a plurality of objects in which information from the plurality of information sources is placed based at least in part on a plurality of inputs and a plurality of weight values, the plurality of inputs including a new query;

at least one server, associated with the artificial intelligence agent, that provides one or more of the objects to one or more recipients, each of the one or more recipients having a display on which to display the information in the selected objects;

the recipients enabling for one or more users to generate feedback about the information; and

wherein the artificial intelligence agent redetermines at least one of the weight values at the end of an epoch, the epoch comprising an act of selecting an object corresponding to one of the information sources and selects, in response to a subsequent query, which objects to receive information using the at least one of the redetermined weight values and inputs associated with the subsequent query.

40. The system of claim 39, wherein the information sources are web pages and the information comprises web page links.

41. The system of claim 39, wherein the one or more objects comprise slots in a link directory page.

42. The system of claim 39, wherein the epoch comprises a selection of a web link.

43. The system of claim 39, wherein the epoch comprises a visit to a web page.

44. A system for controlling information output based on user feedback comprising:

a server that provides pieces of information to a recipient via a plurality of slots that present the pieces of information to a user; and

an artificial intelligence module in communication with the server and operable to rank the pieces of information such that the pieces of information are placed in given ones of the plurality of slots,

wherein when one of the slots is selected by the user, such act of selection comprises feedback to the artificial intelligence module about a corresponding piece of information and the artificial intelligence module redetermines the ranking for the pieces of information, and

wherein, in response to a subsequent query, at least one of the pieces of information is either placed in a slot different from a slot in which it was previously placed and presented to the user as a result of the redetermined ranking or not placed in any of the slots presented to the user.

45. The system of claim 44, wherein one or more of the pieces of information comprise links to web sites.

46. The system of claim 44, wherein one or more of the pieces of information comprise chat participants.

47. The system of claim 44, wherein one or more of the pieces of information comprise email participants.

48. The system of claim 44, wherein an attribute of the user is considered in redetermining the ranking of the pieces of information is placed.

49. In a networked computer system, a system for controlling information output based on user feedback comprising:

means for conveying information to a user in response to a user query, said information being displayed in a ranked list; and

means for receiving an indication of a selection of a portion of the information in the ranked list, wherein the act of selection comprises an epoch in which the selection is used to redetermine how at least some of said information is ranked in the list, wherein, in response to a subsequent query, at least a portion of said information is displayed in a different position in the ranked list.

50. The system of claim 49, wherein the means for receiving an indication receives the indication from a pointing device of a computer.

51. The system of claim 50, wherein the pointing device comprises a computer mouse.

52. The system of claim 49, wherein the information comprises links to web sites.

53. The system of claim 49, wherein the information comprises chat participants.

54. The system of claim 49, wherein the information comprises email participants.

55. The system of claim 49, wherein attributes of a particular user are considered in redetermining how said information is ranked in the list.

56. A method of controlling information output based on user feedback comprising:

conveying a list of information to a user in response to a user query;

detecting an act of selecting an entry in the list of information, wherein the act of selecting comprises an epoch; and

causing an update to an order of the list of information at an end of the epoch such that in response to a subsequent query, the entry is placed in a different position or placed in none of the positions in the list of information that is conveyed to a subsequent user that made the subsequent query.

57. The method of claim 56, wherein the list of information comprises links to web sites.

58. The method of claim 56, wherein the list of information comprises chat participants.

59. The method of claim 56, wherein the list of information comprises email participants.

60. The method of claim 56, further comprising:

considering attributes of a particular user in determining the position of the entry in the list of information.

61. The method of claim 56, wherein the act of selecting an entry comprises a mouse click.

62. The method of claim 56, wherein the act of selecting an entry comprises visiting a web site.

63. A method for controlling information output based on user feedback comprising:

sending, from a server and over an electronic network, information to a user in response to a user query, the information being presented to the user in a series of slots;

detecting that the user has selected one of the slots, the act of selecting the one of the slots comprising feedback to the server; and

updating, based on the feedback, the order of how the information is to be presented to a subsequent user in response to a subsequent query.

64. The method of claim 63, wherein the information comprises links to web sites.

65. The method of claim 63, wherein the information comprises chat participants.

66. The method of claim 63, wherein the information comprises email participants.

67. The method of claim 63, further comprising:

considering attributes of a particular user in the step of updating.