Method and apparatus for utilizing user feedback to improve signifier mapping
First Claim
1. A method, comprising:
- processing a user input comprising a signifier for a target resource with reference to a heuristic knowledge base utilizing a processor of a computing platform to determine a possible target resource, wherein the possible target resource is determined at least in part without regard to any comparison of linguistic relationships of the signifier to content and/or metadata contained within the possible target resource, wherein the heuristic knowledge base is dynamically updated utilizing the processor in view of feedback relating to one or more prior responses to one or more corresponding signifier inputs from one or more users, and wherein the signifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention; and
updating the heuristic knowledge base in accordance with one or more express inputs received from one or more additional users identifying additional and/or desired pairings of one or more target resources to be associated with one or more respective signifiers.
7 Assignments
0 Petitions
Accused Products
Abstract
An apparatus for finding resources on a network comprises a finder server having access to: (a) a database including: (i) an index of resources available on network of interconnected computers on which a plurality of resources reside; and (ii) information regarding user feedback gathered from previous operations of the apparatus by a user and plural previous users; and (b) a learning system operable to access and learn from information contained on the database. The finder server is operable to locate, in response to entry by the user of a resource identity signifier, a single intended target resource intended by the user to uniquely correspond to the resource identity signifier, from among a plurality of resources located on the network, by: receiving a resource identity signifier from the user; accessing the database to determine, based on the information in the database, which, if any, of the indexed resources is likely to be the intended target resource; and directing a computer of the user so as to cause that computer to connect the user to the address of the resource, if any, determined as likely to be the intended target resource.
78 Citations
72 Claims
-
1. A method, comprising:
-
processing a user input comprising a signifier for a target resource with reference to a heuristic knowledge base utilizing a processor of a computing platform to determine a possible target resource, wherein the possible target resource is determined at least in part without regard to any comparison of linguistic relationships of the signifier to content and/or metadata contained within the possible target resource, wherein the heuristic knowledge base is dynamically updated utilizing the processor in view of feedback relating to one or more prior responses to one or more corresponding signifier inputs from one or more users, and wherein the signifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention; and updating the heuristic knowledge base in accordance with one or more express inputs received from one or more additional users identifying additional and/or desired pairings of one or more target resources to be associated with one or more respective signifiers. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A method for finding a target, comprising:
-
processing a plurality of user inputs comprising one or more signifiers for a target resource from a respective plurality of users with reference to a heuristic knowledge base utilizing a processor of a computing platform to determine a possible target resource, wherein the possible target resource is determined at least in part without regard to any comparison of linguistic relationships of the one or more signifiers to content and/or metadata contained within the possible target resource, wherein the one or more signifiers comprise natural language expressions processed without regard to any linguistic constraints imposed by a standard and/or convention; updating said heuristic knowledge base at least in part in response to receiving feedback from the plurality of users; and updating said heuristic knowledge base in accordance with one or more express inputs received from one or more of the plurality of users identifying additional and/or desired pairings of one or more target resources to be associated with one or more respective signifiers. - View Dependent Claims (16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28)
-
-
29. A method of finding an intended target resource, comprising:
-
receiving a resource identity signifier intended to uniquely correspond to the target resource located on a network, wherein the resource identity signifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention; accessing database information that includes an index of available resources on the network, and includes information indicative of associated confidence levels and associated social usage of the resource identity signifier from multi-user feedback, both the associated confidence levels and social usage relating to previous results of attempts to connect to intended target resources; determining which of a plurality of matched indexed resources is likely to be the intended target resource that uniquely corresponds to the resource identity signifier using the associated confidence level of the resource identity signifier, wherein said determining is performed at least in part without regard to any comparison of linguistic relationships of the resource identity signifier to content and/or metadata contained within the plurality of matched indexed resources; updating the database information to add information usable to increase the confidence level associated with a mapping between the resource identity signifier and an address of the determined intended target resource if feedback indicates that the connected intended target resource corresponds to the resource identity signifier; updating the database information so as add information usable to decrease the confidence level associated with a mapping between the resource identity signifier and the address of the determined intended target resource if feedback indicates that the connected intended target resource fails to correspond to the resource identity signifier; and updating the database information provisionally to add an available resource after locating the resource. - View Dependent Claims (30, 31, 32, 33, 34, 35, 36, 37, 38)
-
-
39. A method of determining, in response to entry of an item identity signifier, an intended target item, comprising:
-
receiving the item identity signifier; accessing a database on a finder server on a network; learning a social usage of the item identity signifier from multi-user feedback inputs based on previous results by the finder server; and determining an indexed item as likely uniquely corresponding to the item identity signifier based on the social usage of the item identity signifier, wherein the indexed item is determined at least in part without regard to any comparison of linguistic relationships of the item identity signifier to content and/or metadata contained within the indexed item, wherein the item identity signifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention. - View Dependent Claims (40, 41, 42, 43, 44, 45, 46, 47, 48)
-
-
49. A method of interpreting user-inputted identifiers as unique item identifications, comprising:
-
enabling a population of users to submit identifier-item pairs, an identifier and an item in each pair being associated with one another; learning a social usage of identifiers associated with unique items by the population of users based at least in part on feedback from previous operations of the method from multiple members of the population; and for any user in the population of users, mapping an identifier submitted by the user to a single corresponding item based on the social usage of the entered identifier, wherein the identifier is mapped at least in part without regard to any comparison of linguistic relationships of the identifier to content and/or metadata contained within the single corresponding item, wherein the identifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention. - View Dependent Claims (50, 51, 52, 53, 54, 55, 56, 57, 58, 59)
-
-
60. A method of translating user-inputted identifiers into unique item identifications, comprising:
-
enabling a population of users to submit identifier-item pairs to be associated with one another; learning a social usage of identifiers associated with unique items by a population of multiple users sharing a defined subject or interest; and for any user, mapping a user-entered identifier to a single corresponding item known to the user based on the social usage of the entered identifier, wherein the user-entered identifier is mapped at least in part without regard to any comparison of linguistic relationships of the user-entered identifier to content and/or metadata contained within the single corresponding item, wherein the identifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention. - View Dependent Claims (61, 62, 63, 64, 65, 66, 67, 68, 69)
-
-
70. A method of connecting to an intended target resource, comprising:
-
receiving at a server a resource identity signifier intended to uniquely correspond to a resource located on a network; accessing database information on the server that includes an index of available resources on the network, and includes information indicative of associated confidence levels and associated social usage of the resource identity signifier derived from multi-user feedback, both the associated confidence levels and social usage relating to previous results of attempts to connect to intended target resources; determining which of a plurality of matched indexed resources is likely to be the intended target resource that uniquely corresponds to the recognized resource identity signifier using the associated confidence level associated with a mapping between the resource identity signifier and an address of the matched target resource, wherein the intended target resource is determined at least in part without regard to any comparison of linguistic relationships of the resource identity signifier to content and/or metadata contained within the intended target resource, wherein the confidence is for the signifier-item association, wherein the resource identity signifier comprises a natural language expression processed without regard to any linguistic constraints imposed by a standard and/or convention; updating the database information using a multi-user feedback to add information usable to increase the confidence level associated with a mapping between the resource identity signifier and an address of the determined intended target resource if feedback indicates that the connected intended target resource corresponds to the resource identity signifier; updating the database information using a multi-user feedback to add information usable to decrease the confidence level associated with a mapping between the resource identity signifier and the address of the determined intended target resource if feedback indicates that the connected intended target resource fails to correspond to the resource identity signifier; updating the database information provisionally to add an available resource after locating the resource; updating the database information to modify the social usage using the feedback; generating a list displayed on a computer of one or more links to a possible matched resource, the list being ordered according to confidence level, with a resource having a highest confidence level being ranked highest; and providing a selective option to enable any user to update the database with an available resource. - View Dependent Claims (71, 72)
-
Specification