Patent Issued for Learning based metric determination and clustering for service routing (USPTO 11140268): United Services Automobile Association
2021 OCT 21 (NewsRx) -- By a
The assignee for this patent, patent number 11140268, is
Reporters obtained the following quote from the background information supplied by the inventors: “An organization may use any number of computing systems, communications networks, data storage devices, or other types of systems to provide service to individuals. An organization may also employ service representatives that use the various systems to assist individuals in service sessions that are conducted over the telephone, in a video conference, through text chat sessions, in person, and/or over other communication channels. Organizations may strive to provide an efficient and productive service interaction between service representatives and the individuals being serviced, while maintaining an appropriate quality level for the service provided by service representatives.”
In addition to obtaining background information on this patent, NewsRx editors also obtained the inventors’ summary information for this patent: “Implementations of the present disclosure are generally directed to routing service requests. More specifically, implementations are directed to using machine learning techniques, such as deep learning that includes classification, clustering, and/or other techniques, to determine metric(s) for a service session, the metric(s) providing a prediction of survey scores that would be submitted by serviced individuals in a post-session survey, clustering users based at least partly on the determined metric(s), and routing incoming service requests based on the clustering.
“In general, innovative aspects of the subject matter described in this specification can be embodied in methods that includes actions of: receiving a session record of communications between a service representative (SR) and an individual during a service session in a service environment; providing the session record as input to at least one computer-processable model that determines, based on the session record, at least one metric for the service session, the at least one model having been trained, using machine learning and based at least partly on survey data for previous service sessions, to provide the at least one metric that is a prediction of at least one survey score for the service session; determining a cluster for the individual based at least partly on the at least one metric, the cluster including a plurality of individuals having at least one similar metric determined by the at least one model; in response to receiving a subsequent service request from the individual within the service environment, selecting at least one SR based on the cluster of the individual; and initiating a subsequent service session between the individual and the at least one SR that is selected based on the cluster of the individual.
“Implementations can optionally include one or more of the following features: the cluster is further determined based on the plurality of individuals in the cluster having a similarity in one or more of: a product discussed by the individuals during service sessions; a location of the individuals; and a characteristic of the individuals; the at least one SR is selected based on determining that individuals in the cluster each have a respective metric that exceeds a threshold value and that is determined, by the at least one model, based on a respective session record of communications between the respective individual and the at least one SR; the at least one SR is selected based on determining that an average metric for individuals in the cluster exceeds a threshold value, the average metric being an average of metrics that are determined, by the at least one model, based on session records of communications between the individuals and the at least one SR; the service session is an audio call between the SR and the individual; the session record includes an audio record of at least a portion of the audio call; the operations further include developing the at least one model based on training data that describes, for each of a plurality of previous service sessions: a previous session record of communications during a respective previous service session; and at least one survey score provided to rate the respective previous service session; the operations further include receiving the at least one survey score provided, by the individual, to rate the service session; the operations further include employing the at least one survey score and the session record to further train the at least one model; the at least one model is trained to identify a plurality of features present in the session record; each of the at least one metric is determined based on a strength of one or more corresponding features present in the session record; the machine learning is deep learning in which a separate model is trained to identify each of the plurality of features; the at least one model is a concatenated model that is a combination of a language model and an acoustic model; the language model is output from a language classifier recurrent neural network (RNN); and/or the acoustic model is output from an acoustic feature layer convolutional neural network (CNN).
“Other implementations of any of the above aspects include corresponding systems, apparatus, and computer programs that are configured to perform the actions of the methods, encoded on computer storage devices. The present disclosure also provides a computer-readable storage medium coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations in accordance with implementations of the methods provided herein. The present disclosure further provides a system for implementing the methods provided herein. The system includes one or more processors, and a computer-readable storage medium coupled to the one or more processors having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations in accordance with implementations of the methods provided herein.
“Implementations of the present disclosure provide one or more of the following advantages. Through use of machine learning developed models to determine metrics that are predicted survey results, implementations provide a technique to determine survey results in a service environment even in instances when the individual being serviced does not complete a survey. Moreover, the model(s) enable an objective determination of survey results across a wider range of individuals than would ordinarily complete survey results, thus providing more accurate, objective, and comprehensive view of the service provided in a service environment. By providing more accurate survey results through use of predictive modeling, implementations may avoid repeatedly requesting that individual(s) complete surveys following a service session, thus avoiding the expenditure of processing power, network bandwidth, storage space, active memory, and/or other computing resources that may be expended in traditional service environments. In instances where the predictive model(s) are sufficiently developed to enable session metric(s) to be determined without surveying individuals, implementations also avoid the expenditure of computing resources that would otherwise be consumed to survey individuals following a service session.”
The claims supplied by the inventors are:
“1. A computer-implemented method performed by at least one processor, the method comprising: receiving, by the at least one processor, a session record of communications between a service representative (SR) and a customer during a service session in a service environment; providing, by at least one processor, the session record as input to at least one computer-processable model that determines, based on the session record, a prediction of at least one survey score that would likely be submitted by the customer for the service session; determining, by the at least one processor, a cluster for the customer based at least partly on the prediction of the at least one survey score, the cluster including a plurality of other customers; in response to receiving a subsequent service request from the customer within the service environment, selecting, by the at least one processor, at least one SR based on determining that an average survey score for customers in the cluster exceeds a threshold value, the average survey score being an average of survey scores that are determined, by the at least one model, based on session records of communications between the customers and the at least on SR; and responsive to identifying that the customer is included in the cluster, bypassing a default interactive voice response process by directly initiating, by the at least one processor, a subsequent service session between the customer and the at least one SR that is selected based on the cluster of the customer.
“2. The method of claim 1, wherein the cluster is further determined based on the plurality of other customers in the cluster having a similarity in one or more of: a product discussed by the plurality of other customers during service sessions; a location of the plurality of other customers; and a characteristic of the plurality of other customers.
“3. The method of claim 1, wherein the at least one SR is selected based on determining that respective customers in the cluster each have a respective survey score that exceeds a threshold value and that is determined, by the at least one model, based on a respective session record of communications between the respective customer and the at least one SR.
“4. The method of claim 1, wherein: the service session is an audio call between the SR and the customer; and the session record includes an audio record of at least a portion of the audio call.
“5. The method of claim 1, further comprising: developing, by at least one processor, the at least one model based on training data that describes, for each of a plurality of previous service sessions: a previous session record of communications during a respective previous service session; and at least one survey score provided to rate the respective previous service session.
“6. The method of claim 5, further comprising: receiving, by at least one processor, the at least one survey score provided, by the customer, to rate the service session; and employing, by at least one processor, the at least one survey score and the session record to further train the at least one model.
“7. The method of claim 1, wherein: the at least one model is trained to identify a plurality of features present in the session record; and the prediction of each of the at least one survey score is determined based on a strength of one or more corresponding features present in the session record.
“8. The method of claim 7, wherein the at least one model is a deep learning model in which a separate model is trained to identify each of the plurality of features.
“9. The method of claim 1, wherein: the at least one model is a concatenated model that is a combination of a language model and an acoustic model; the language model is output from a language classifier recurrent neural network (RNN); and the acoustic model is output from an acoustic feature layer convolutional neural network (CNN).
“10. The method of claim 1, further comprising: receiving a completed survey from the customer; comparing the completed survey with the prediction of the at least one survey score for the customer to provide training feedback data; and employing the training feedback data to refine the at least one computer-processable model.
“11. A system, comprising: at least one processor; and a memory communicatively coupled to the at least one processor, the memory storing instructions which, when executed by the at least one processor, cause the at least one processor to perform operations comprising: receiving a session record of communications between a service representative (SR) and a customer during a service session in a service environment; providing the session record as input to at least one computer-processable model that determines, based on the session record, a prediction of at least one survey score that would likely be submitted by the customer for the service session; determining a cluster for the customer based at least partly on the prediction of the at least one survey score, the cluster including a plurality of other customers; in response to receiving a subsequent service request from the customer within the service environment, selecting at least one SR based on determining that an average survey score for customers in the cluster exceeds a threshold value, the average survey score being an average of survey scores that are determined, by the at least one model, based on session records of communications between the customers and the at least one SR; and responsive to identifying that the customer is included in the cluster, bypassing a default interactive voice response process by directly initiating a subsequent service session between the customer and the at least one SR that is selected based on the cluster of the customer.
“12. The system of claim 11, wherein the cluster is further determined based on the plurality of other customers in the cluster having a similarity in one or more of: a product discussed by the plurality of other customers during service sessions; a location of the plurality of other customers; and a characteristic of the plurality of other customers.
“13. The system of claim 11, wherein the at least one SR is selected based on determining that respective customers in the cluster each have a respective survey score that exceeds a threshold value and that is determined, by the at least one model, based on a respective session record of communications between the respective customer and the at least one SR.
“14. The system of claim 11, wherein: the service session is an audio call between the SR and the customer; and the session record includes an audio record of at least a portion of the audio call.
“15. The system of claim 11, the operations further comprising: developing the at least one model based on training data that describes, for each of a plurality of previous service sessions: a previous session record of communications during a respective previous service session; and at least one survey score provided to rate the respective previous service session.
“16. The system of claim 15, the operations further comprising: receiving the at least one survey score provided, by the customer, to rate the service session; and employing the at least one survey score and the session record to further train the at least one model.
“17. The system of claim 11, wherein: the at least one model is trained to identify a plurality of features present in the session record; and the prediction of each of the at least one survey score is determined based on a strength of one or more corresponding features present in the session record.
“18. The system of claim 11, wherein: the at least one model is a concatenated model that is a combination of a language model and an acoustic model; the language model is output from a language classifier recurrent neural network (RNN); and the acoustic model is output from an acoustic feature layer convolutional neural network (CNN).
“19. The system of claim 11, wherein the operations further comprise: receiving a completed survey from the customer; comparing the completed survey with the prediction of the at least one survey score for the customer to provide training feedback data; and employing the training feedback data to refine the at least one computer-processable model.
“20. One or more non-transitory computer-readable media storing instructions which, when executed by at least one processor, cause the at least one processor to perform operations comprising: receiving a session record of communications between a service representative (SR) and a customer during a service session in a service environment; providing the session record as input to at least one computer-processable model that determines, based on the session record, a prediction of at least one survey score that would likely be submitted by the customer for the service session; determining a cluster for the customer based at least partly on the prediction of the at least one survey score, the cluster including a plurality of other customers; in response to receiving a subsequent service request from the customer within the service environment, selecting at least one SR based on determining that an average survey score for customers in the cluster exceeds a threshold value, the average survey score being an average of survey scores that are determined, by the at least one model, based on session records of communications between the customers and the at least one SR; and responsive to identifying that the customer is included in the cluster, bypassing a default interactive voice response process by directly initiating a subsequent service session between the customer and the at least one SR that is selected based on the cluster of the customer.”
For more information, see this patent: Chadha, Bipin. Learning based metric determination and clustering for service routing.
(Our reports deliver fact-based news of research and discoveries from around the world.)
University of Copenhagen Reports Findings in Malaria (An ethnographic study of how health system, socio-cultural and individual factors influence uptake of intermittent preventive treatment of malaria in pregnancy with sulfadoxine-pyrimethamine …): Mosquito-Borne Diseases – Malaria
Researchers Submit Patent Application, “Integrated Digital-Analog Archiving Systems And Methods For Document Preservation”, for Approval (USPTO 20210311907): Patent Application
Advisor News
- Goldman Sachs, others weigh in on what’s ahead for the market in 2025
- How is consumers’ investing confidence impacting the need for advisors?
- Can a balance of withdrawals and annuities outshine the 4% rule?
- Economy ‘in a sweet spot’ but some concerns ahead
- Trump, Capitol allies likely looking at quick cryptocurrency legislation
More Advisor NewsAnnuity News
Health/Employee Benefits News
- Studies from University of Wisconsin Describe New Findings in Insurance (Care Transition Management and Patient Outcomes In Hospitalized Medicare Beneficiaries): Insurance
- Shock poll: 41 percent of young voters find killing of UnitedHealthcare CEO acceptable
- Former Bright Health, once a Minnesota IPO darling, goes private in $1.3B deal
- Ask the Medicare Specialist
- Murphy, safely re-elected, attacks medical insures
More Health/Employee Benefits NewsLife Insurance News