US 12,007,869 B2
Systems and methods for modeling computer resource metrics
Igor Trubin, Glen Allen, VA (US); Mark Schutt, Chesterfield, VA (US); and Jeffery Robinson, Richmond, VA (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Dec. 23, 2021, as Appl. No. 17/561,358.
Application 17/561,358 is a continuation of application No. 16/549,770, filed on Aug. 23, 2019, granted, now 11,243,863.
Application 16/549,770 is a continuation of application No. 15/184,501, filed on Jun. 16, 2016, granted, now 10,437,697, issued on Oct. 8, 2019.
Claims priority of provisional application 62/180,777, filed on Jun. 17, 2015.
Prior Publication US 2022/0114073 A1, Apr. 14, 2022
Int. Cl. G06F 11/34 (2006.01); G06Q 10/067 (2023.01)
CPC G06F 11/3442 (2013.01) [G06F 11/3452 (2013.01); G06Q 10/067 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
generating statistical models indicative of correlations between aggregated interaction data for a plurality of interaction types and aggregated resource utilization data for a plurality of devices respectively mapped to the interaction types, the interaction types comprising a mobile-device interaction type and a non-mobile-device interaction type, the statistical models comprising a first model and a second model;
for each model of the statistical models, scoring the model for one or more interaction types based on respective levels of the correlations (i) indicated by the model and (ii) related to resource type usages of the one or more interaction types; and
remapping the devices and the interaction types based on the scoring of the statistical models by:
(i) using the first model to remap one or more first devices having a first resource type to the mobile-device interaction type based on the first model indicating a high level of correlation between the mobile-device interaction type and the first resource type; and
(ii) using the second model to remap one or more second devices having a second resource type to the non-mobile-device interaction type based on the second model indicating a high level of correlation between the non-mobile-device interaction type and the second resource type, the second resource type being different from the first resource type.