Back to results
US 12,737,710 B2Granted
Customer experience index estimator for a smart service analyzer
AssigneeRakuten Symphony, Inc.
InventorTasnim Ahmed
Life of this patent
Estimating the term…
Abstract
A method includes normalizing one or more key performance indicator (KPI), where the one or more KPI include user-level quantitative KPI or user-level qualitative KPI; determining whether weights are available for each normalized KPI; and in response to the weights being available for each normalized KPI, converting, based on the weights for each normalized KPI, each normalized KPI to a customer experience index (CEI) for each user and for each network service. The method further includes determining whether a trend shift has occurred based on the CEI; and automatically generating an alert in response to a determination that the trend shift occurred.
Claims20
A method, comprising:
- providing probes for obtaining data about a mobile network;
- providing the data about the mobile network obtained by the probes to a key performance indicator (KPI) generator;
- based on the data about the mobile network obtained by the probes, generating one or more KPI at the KPI generator;
- applying a machine learning (ML) model to normalize the one or more KPI in a performance range using KPI Performance Thresholds, where the one or more KPI include user-level quantitative KPI or user-level qualitative KPI;
- determining whether weights are available for each of the one or more normalized KPI;
- in response to the determining the weights are available for each of the one or more normalized KPI, converting, based on the weights for each of the one or more normalized KPI, each of the one or more normalized KPI to a customer experience index (CEI) for each user and for each network service;
- calculating a shift ratio of a trend shift count representing the one or more normalized KPI that are equal to zero divided by a total number of iterations;
- determining whether a trend shift representing the shift ratio having a value greater than a predetermined threshold has occurred based on the CEI;
- automatically generating an alert in response to a determination that the trend shift occurred;
- electronically sending the alert indicating that the trend shift has occurred to stakeholders for review;
- receiving confirmation of the trend shift from the stakeholders;
- and
- in response to receiving the confirmation, creating new KPI performance thresholds based on the trend shift.
10 more elements
An apparatus configured to:
- provide probes for obtaining data about a mobile network;
- provide the data about the mobile network obtained by the probes to a key performance indicator (KPI) generator;
- based on the data about the mobile network obtained by the probes, generate one or more KPI at the KPI generator;
- apply a machine learning (ML) model to normalize the one or more KPI in a performance range using KPI Performance Thresholds, where the one or more KPI include user-level quantitative KPI or user-level qualitative KPI;
- determine whether weights are available for each of the one or more normalized KPI;
- in response to the determining the weights are available for each of the one or more normalized KPI, convert, based on KPI weights, each of the one or more normalized KPI to a customer experience index (CEI) for each user and for each network service;
- calculate a shift ratio of a trend shift count representing the one or more normalized KPI that are equal to zero divided by a total number of iterations;
- determine whether a trend shift representing the shift ratio having a value greater than a predetermined threshold has occurred based on the CEI;
- automatically generate an alert in response to a determination that the trend shift occurred;
- electronically send the alert indicating that the trend shift has occurred to stakeholders for review;
- receive confirmation of the trend shift from the stakeholders;
- and
- in response to receiving the confirmation, create new KPI performance thresholds based on the trend shift.
10 more elements
A non-transitory computer-readable media having computer-readable instructions stored thereon, which when executed perform operations to:
- provide probes for obtaining data about a mobile network;
- provide the data about the mobile network obtained by the probes to a key performance indicator (KPI) generator;
- based on the data about the mobile network obtained by the probes, generate one or more KPI at the KPI generator;
- apply a machine learning (ML) model to normalize the one or more KPI in a performance range using KPI Performance Thresholds, where the one or more KPI include user-level quantitative KPI or user-level qualitative KPI;
- determine whether weights are available for each of the one or more normalized KPI;
- in response to the determining the weights are available for each of the one or more normalized KPI, convert, based on KPI weights, each of the one or more normalized KPI to a customer experience index (CEI) for each user and for each network service;
- calculate a shift ratio of a trend shift count representing the one or more normalized KPI that are equal to zero divided by a total number of iterations;
- determine whether a trend shift representing the shift ratio having a value greater than a predetermined threshold has occurred based on the CEI;
- automatically generate an alert in response to a determination that the trend shift occurred;
- electronically send the alert indicating that the trend shift has occurred to stakeholders for review;
- receive confirmation of the trend shift from the stakeholders;
- and
- in response to receiving the confirmation, create new KPI performance thresholds based on the trend shift.
10 more elements
Dependencies are inferred from the claim text.