| CPC G06Q 10/06398 (2013.01) [G06Q 50/01 (2013.01)] | 24 Claims |
|
1. A method, using a processor of a device, for determining a final quantitative measure for each evaluator j, wherein a quantitative measure represents a rating acumen of each evaluator, wherein evaluators with a higher final quantitative measure are considered better evaluators than those with a lower quantitative measure, the final quantitative measure of an evaluator used for determining a final rating by the evaluators of an evaluee, the method comprising:
(a) a plurality of evaluators and a plurality of evaluees each using a network connected device for uploading and downloading information to a server, wherein evaluator information comprises evaluator personal information, and wherein evaluee information comprises evaluee performance statistics, metrics, and media, wherein using the network connected device each evaluator can access the server to review the performance statistics, the metrics, and the media of each evaluee;
(b) displaying a graphical user interface on the device, each evaluator j providing an unweighted rating of an evaluee using the graphical user interface, the unweighted ratings stored in memory, wherein an evaluator j may provide an unweighted rating of same evaluees and/or of different evaluees as another evaluator j, wherein the evaluators j and the evaluees i are distributed over a geographic area thereby requiring use of the network-connected device, wherein the evaluators j can provide their unweighted ratings at different times and of different evaluees, the unweighted rating of an evaluee related to a skill or to an athletic or performance related attribute of the evaluee;
(c) assigning an initial quantitative measure to each evaluator j without consideration of the rating acumen of the evaluator j, wherein the initial quantitative measure is stored in memory;
(d) applying the initial quantitative measure for each evaluator j, as assigned at step (c), to the respective unweighted rating of each evaluee i by each evaluator j, as provided at step (b), to generate an initial weighted rating of each evaluee i by each evaluator j, wherein the initial weighted ratings of all evaluees i by all evaluators j are stored in a memory;
(e) summing the initial weighted ratings of all evaluators j for each evaluee i and normalizing a resultant sum by dividing by a sum of the initial quantitative measures of all evaluators j who evaluated the each evaluee i to generate an initial overall rating of each evaluee i, the initial overall rating of an evaluee considered an initial crowd-sourced rating of the evaluee;
(f) for each evaluator j, determining an absolute value of a differential between the unweighted rating by each evaluator j of each evaluee i, as provided at step (b), and the initial overall rating of each evaluee i, as determined at step (e), and summing the absolute values for each evaluator j over all evaluees i, as assessed by evaluator j, to generate a sum for each evaluator j, assigning an updated quantitative measure to each evaluator j based on the sum for the respective evaluator j;
(g) applying the respective updated quantitative measure for each evaluator j, as assigned at step (f), to the respective unweighted rating of each evaluee i, by each evaluator j, as provided at step (b), to generate an updated weighted rating of each evaluee i by each evaluator j;
(h) summing the updated weighted ratings of each evaluee i by all evaluators j, as determined at step (g), and normalizing a resultant sum by dividing by a sum of the updated quantitative measures of all evaluators j who evaluated the each evaluee i to generate an updated overall rating of each evaluee i, the updated overall rating of an evaluee considered a more accurate crowd-sourced rating of the evaluee than the initial overall rating of the evaluee;
(i) for each evaluator j, determining an absolute value of a differential between the unweighted rating by each evaluator j of each evaluee i, as provided at step (b), and the updated overall rating of each evaluee i, as determined at step (h), and summing the absolute values for each evaluator j over all evaluees i to generate a sum for each evaluator j, assigning an updated quantitative measure to each evaluator j based on the sum for the respective evaluator j;
(j) to increase an accuracy of each updated quantitative measure for each evaluator j and thereby a more accurate rating of each evaluee i, repeating steps (g) through (i), wherein each iteration through steps (g) through (i) generates another updated quantitative measure for each evaluator j that is used for the next iteration, wherein repeating steps (g) through (i) iteratively converges the updated quantitative measure such that a differential between two most recent updated quantitative measures for an evaluator j is less than a predetermined value, or repeating steps (g) through (i) a predetermined number of iterations, wherein a most recent updated quantitative measure is referred to as a final quantitative measure when the differential is less than the predetermined value or when the steps (g) through (i) have been repeated the predetermined number of iterations, wherein the final quantitative measure of an evaluator j represents a rating acumen of evaluator j; and
(k) after determining the final quantitative measure of each evaluator j according to step (j), applying the final quantitative measure of each evaluator j to the unweighted rating of an evaluee i as provided by the evaluator j, wherein use of the final quantitative measure of an evaluator in conjunction with evaluation of an evaluee provides a more accurate crowd-sourced evaluation of the evaluee i, and wherein the final quantitative measure is available for future use in conjunction with future evaluations by the evaluator j, wherein the evaluation is related to a skill or to an athletic or performance related attribute of an evaluee;
(l) as additional evaluators j, additional evaluees i, and/or additional ratings are entered into the network via the network of connected devices, steps (b) to (k) may be repeated, thereby dynamically responding to the crowd-sourcing population of evaluators and evaluees.
|