# US 2019/0377653 A1 · Systems and methods for modeling computer resource metrics > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0377653 A1 - Document ID: US20190377653A1 - Status: Published application - Kind: A1 - Application number: 16/549,770 - Filed: 2019-08-23 - Published: 2019-12-12 - Earliest claimed: 2015-06-17 - Assignees: Capital One Services, LLC - Applicants: Capital One Services, LLC - Inventors: Igor Trubin; Mark Schutt; Jeffery Robinson - Web page: https://patentgrep.com/patent/US20190377653A1 ## CPC - G06F 11/3442: {for planning or managing the needed capacity} - G06F 11/3452: {Performance evaluation by statistical analysis} - G06Q 10/067: Enterprise or organisation modelling - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR ## Abstract This disclosure relates generally to system modeling, and more particularly to systems and methods for modeling computer resource metrics. In one embodiment, a processor-implemented computer resource metric modeling method is disclosed. The method may include detecting one or more statistical trends in aggregated interaction data for one or more interaction types, and mapping each interaction type to one or more devices facilitating the transactions. The method may further include generating one or more linear regression models of a relationship between device utilization and interaction volume, and calculating one or more diagnostic statistics for the one or more linear regression models. A subset of the linear regression models may be filtered out based on the one or more diagnostic statistics. One or more forecasts may be generated using the remaining linear regression models, using which a report may be generated and provided. ## Claims ### Claim at position 1 -20. (canceled) ### Claim 21 A processor-implemented computer resource metric modeling method, comprising: aggregating transaction volume data of one or more interaction types; mapping, for each interaction type, one or more devices; aggregating resource utilization data for the one or more devices; generating one or more multivariate models indicative of relationships between variations in the transaction volume data and the resource utilization data, wherein: a multivariate model of the multivariate models is generated for a group of the one or more devices, each group of one or more devices being associated with a traffic channel; calculating diagnostic statistics for the multivariate models; filtering out a first subset of the multivariate models based on the diagnostic statistics, with a second subset of the multivariate models remaining; and generating instructions to modify the one or more devices for load balancing based on the second subset of the multivariate models. ### Claim 22 (depends on 21) The method of claim 21, wherein the one or more interaction types includes a service provided via a mobile user device. ### Claim 23 (depends on 21) The method of claim 21, wherein the diagnostic statistics comprise at least one of a slope value, an intercept value, an R2 fit value, or a root-mean-squared error (RMSE) value. ### Claim 24 (depends on 21) The method of claim 21, wherein the one or more devices include one or more servers providing a service related to the one or more interaction types. ### Claim 25 (depends on 21) The method of claim 21, further comprising detecting statistical trends in the transaction volume data and the resource utilization data via multivariate adaptive statistical filtering. ### Claim 26 (depends on 21) The method of claim 21, further comprising: scoring the one or more multivariate models based on a strength of the multivariate models; and modifying the mapping of the one or more devices to each interaction type based on the scoring. ### Claim 27 (depends on 21) The method of claim 21, wherein the resource utilization data further comprises at least one of a disk utilization data, a memory utilization data, or a network utilization data. ### Claim 28 (depends on 21) The method of claim 21, further comprising: generating weekly profiles for the one or more devices, the weekly profiles comprising at least one of an average resource utilization value, a maximum resource utilization value, or a minimum resource utilization value plotted against each day of a week. ### Claim 29 (depends on 28) The method of claim 28, further comprising: comparing the weekly profiles; and detecting redundant devices based on the comparison. ### Claim 30 (depends on 21) The method of claim 21, further comprising: detecting an actual relationship between variations in user demand for the interaction type and the resource utilization data; comparing the actual relationship to the one or more multivariate models over time; and modifying the second subset of the multivariate models based on the comparison. ### Claim 31 A computer resource metric modeling system, comprising: at least one memory storing instructions; and at least one processor executing the instructions to perform operations comprising: aggregating transaction volume data of one or more interaction types; mapping for each interaction type, one or more devices; aggregating resource utilization data for the one or more devices; generating one or more multivariate models indicative of relationships between variations in the transaction volume data and the resource utilization data, wherein: a multivariate model of the multivariate models is generated for a group of the one or more devices, each group of one or more devices being associated with a traffic channel; calculating diagnostic statistics for the multivariate models; filtering out a first subset of the multivariate models based on the diagnostic statistics, with a second subset of the multivariate models remaining; and generating instructions to modify the one or more devices for load balancing based on the second subset of the multivariate models. ### Claim 32 (depends on 31) The system of claim 31, wherein the one or more interaction type includes a service provided via a mobile user device. ### Claim 33 (depends on 31) The system of claim 31, wherein the diagnostic statistics comprises at least one of a slope value, an intercept value, an R2 fit value, or a root-mean-squared error (RMSE) value. ### Claim 34 (depends on 31) The system of claim 31, wherein the one or more devices include one or more servers providing a service related to the one or more interaction types. ### Claim 35 (depends on 31) The system of claim 31, wherein the operations further comprise detecting statistical trends in the transaction volume data and the resource utilization data via multivariate adaptive statistical filtering. ### Claim 36 (depends on 31) The system of claim 31, wherein the operations further comprise: scoring the one or more multivariate models based on a strength of the multivariate models; and modifying the mapping of the one or more devices to each interaction type based on the scoring. ### Claim 37 (depends on 31) The system of claim 31, wherein the resource utilization data further comprises at least one of a disk utilization data, a memory utilization data, or a network utilization data. ### Claim 38 (depends on 31) The system of claim 31, wherein the operations further comprise: generating weekly profiles for the one or more devices, the weekly profiles comprising at least one of an average resource utilization value, a maximum resource utilization value, or a minimum resource utilization value plotted against each day of a week. ### Claim 39 (depends on 38) The system of claim 38, wherein the operations further comprise: comparing the weekly profiles; and detecting redundant devices based on the comparison. ### Claim 40 (depends on 31) The system of claim 31, wherein the operations further comprise: detecting an actual relationship between variations in user demand for the interaction type and the resource utilization data; comparing the actual relationship to the one or more multivariate models over time; and modifying the second subset of the multivariate models based on the comparison.