Computerized system for enhancing expert-based processes and methods useful in conjunction therewith
First Claim
1. A method for enhancing expert-based processes when receiving input from a plurality of experts operating a corresponding plurality of computerized expert-based processes on a body of data, said data including an agreement set including at least one point of agreement regarding which all of said plurality of experts agree, the method comprising:
- determining a discrepancy set including at least one point of discrepancy regarding which less than all of the plurality of experts agree, said determining including performing a computerized comparison of input received from said plurality of experts thereby to identify points of discrepancy;
providing at least a portion of said discrepancy set to an oracle and receiving oracle input from the oracle resolving at least said point of discrepancy; and
selecting a subset of better computerized expert-based processes, from among said plurality of computerized expert-based processes, based on said oracle input,wherein said oracle input resolves less than all of a total number of points of discrepancy in said discrepancy set by randomly sampling a subset of the total number of points of discrepancy,and wherein each of the expert-based processes has a binary output defining first and second output values and wherein said discrepancy set includes a first set of points of discrepancy in which first and second experts from among said plurality of experts output the first and second output values respectively and a second set of points of discrepancy in which first and second experts from among said plurality of experts output the second and first output values respectively.
2 Assignments
0 Petitions
Accused Products
Abstract
A computerized system for enhancing expert-based processes, the system comprising a computerized expert based data analyzer receiving input from a plurality of experts by operating a corresponding plurality of expert-based processes on a body of data, the input including a discrepancy set including at least one point of discrepancy regarding which less than all of the plurality of experts agree and an agreement set including at least one point of agreement regarding which all of the plurality of experts agree; and an oracle from which oracle input is received resolving at least the point of discrepancy and not resolving any point of agreement in the agreement set; wherein the computerized analyzer is operative to select and to subsequently actuate for purposes of receiving input regarding the body of data, a subset of better experts from among the plurality of experts based on the oracle input.
-
Citations
29 Claims
-
1. A method for enhancing expert-based processes when receiving input from a plurality of experts operating a corresponding plurality of computerized expert-based processes on a body of data, said data including an agreement set including at least one point of agreement regarding which all of said plurality of experts agree, the method comprising:
-
determining a discrepancy set including at least one point of discrepancy regarding which less than all of the plurality of experts agree, said determining including performing a computerized comparison of input received from said plurality of experts thereby to identify points of discrepancy; providing at least a portion of said discrepancy set to an oracle and receiving oracle input from the oracle resolving at least said point of discrepancy; and selecting a subset of better computerized expert-based processes, from among said plurality of computerized expert-based processes, based on said oracle input, wherein said oracle input resolves less than all of a total number of points of discrepancy in said discrepancy set by randomly sampling a subset of the total number of points of discrepancy, and wherein each of the expert-based processes has a binary output defining first and second output values and wherein said discrepancy set includes a first set of points of discrepancy in which first and second experts from among said plurality of experts output the first and second output values respectively and a second set of points of discrepancy in which first and second experts from among said plurality of experts output the second and first output values respectively. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A computer program product, comprising a non-transitory computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method for enhancing expert-based processes, the method comprising:
-
receiving input from a plurality of experts by operating a corresponding plurality of expert-based processes on a body of data, the input including a discrepancy set including at least one point of discrepancy regarding which less than all of said plurality of experts agree and an agreement set including at least one point of agreement regarding which all of said plurality of experts agree; receiving oracle input from an oracle which receives at least a portion of said discrepancy set and resolving at least said point of discrepancy and not resolving any point of agreement in said agreement set; and selecting, and subsequently actuating for purposes of receiving input regarding the body of data, a subset of better experts from among the plurality of experts based on said oracle input, wherein said oracle input resolves less than all of a total number of points of discrepancy in said discrepancy set by randomly sampling a subset of the total number of points of discrepancy, and wherein each of the expert-based processes has a binary output defining first and second output values and wherein said discrepancy set includes a first set of points of discrepancy in which first and second experts from among said plurality of experts output the first and second output values respectively and a second set of points of discrepancy in which first and second experts from among said plurality of experts output the second and first output values respectively.
-
-
16. A computerized system for enhancing expert-based processes, the system comprising:
-
a computerized expert based data analyzer residing on a tangible non-transitory computer readable storage medium and operative for receiving input from a plurality of experts by operating a corresponding plurality of expert-based processes on a body of data, the input including a discrepancy set including at least one point of discrepancy regarding which less than all of said plurality of experts agree and an agreement set including at least one point of agreement regarding which all of said plurality of experts agree; an oracle, receiving at least a portion of said discrepancy set and from which oracle input is received resolving at least said point of discrepancy and not resolving any point of agreement in said agreement set; and wherein said computerized analyzer is operative to select and to subsequently actuate for purposes of receiving input regarding the body of data, a subset of better experts from among the plurality of experts based on said oracle input, wherein said oracle input resolves less than all of a total number of points of discrepancy in said discrepancy set by randomly sampling a subset of the total number of points of discrepancy, and wherein each of the expert-based processes has a binary output defining first and second output values and wherein said discrepancy set includes a first set of points of discrepancy in which first and second experts from among said plurality of experts output the first and second output values respectively and a second set of points of discrepancy in which first and second experts from among said plurality of experts output the second and first output values respectively. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)
-
Specification