United Services Automobile Association (usaa)

Bipin Chadha

Vijay Jayapalan

Gregory B Yarbrough

Eric J Smith

John McChesney TenEyck, Jr

Lambros Petropoulos

Michael Joseph Gaeta

Oscar Roberto Tijerina

Michael P Gout

Jeremy Kolt Loukas

Robert Bruno Pace, Jr

Brett Andrew Walker

Jerome J Gainer, Jr

Samuel E Tumlinson

Jose Esteban Deleon, Iii

Techniques are described for generating metric(s) that predict survey score(s) for a service session. Model(s) may be trained, through supervised or unsupervised machine learning, using training data such as communications from previous service sessions between service representative(s) and individual(s), and survey scores provided by the serviced individual to rate the session on one or more criteria (e.g., survey questions). The model(s) may be trained to output, based on an input session record, metric(s) that each correspond to a survey score that would have been provided by the individual had they completed the survey. The model may be a concatenated model that combines a language model output from a language classifier recurrent neural network, and an acoustic model output from an acoustic feature layer convolutional neural network. Individuals can be clustered according to the metric(s) and/or other factors, and the cluster(s) can be employed for routing incoming service requests.

Learning based metric determination and clustering for service routing

A system includes a server having a database stored thereon, an operator device, and a coordinating device communicatively coupled with the server and with the operator device, wherein the coordinating device comprises a processor and a memory, and wherein the memory is configured to store instructions thereon that, when executed by the processor, cause the coordinating device to perform various steps. The steps include receiving account data, from the server, corresponding with an account of a plurality of accounts saved to the database, receiving preference data, from the operator device, corresponding with preferences of an operator of the operator device and relating to the account data. The steps also include generating an avatar reflecting the account data in accordance with the preference data, and displaying the avatar on a display.

Real time avatar

Systems and methods simulate call centers networks and call loads to test load balancing and routing. The simulation can be used for generating, using a load balancer, a call score for the one or more calls based on the call information and selecting, using the load balancer, one of the simulated call centers as a selected call center based on the call score and the respective response entity profile.

Systems and methods for simulating multiple call center balancing

Various embodiments of the present disclosure relate generally to providing virtual assistants. More specifically, various embodiments of the present disclosure relate to methods and systems for providing a simulation-based virtual advisor. In some embodiments, the systems and methods combine and/or integrate the functionality of a conversational virtual agent (e.g., a virtual assistant) with the functionality of a simulation virtual agent, to provide users, customers, or members with targeted, user-specific information in response to questions or other intentions posed by the users via messages exchanged between the users and the agents.

Simulation-based virtual advisor

Systems and methods simulate call centers networks and call loads to test load balancing and routing. Performance can be logged to memorialize the load balancing and routing techniques' handling of different loads. This allows the testing of new algorithms or architectures and stress testing of existing architectures in non-production environments. Load balancing can be based on a call score, and routing can invoke proprietary routing protocols.

Techniques are described for generating metric(s) that predict survey score(s) for a service session. Model(s) may be trained, through supervised or unsupervised machine learning, using training data from previous service sessions between service representative(s) and individual(s). Training data may include, for previous service session(s), a session record (e.g., audio record) of the session and a set of survey scores provided by the serviced individual to rate the session on one or more criteria (e.g., survey questions). The model(s) may be trained to output, based on an input session record, metric(s) that each correspond to a survey score that would have been provided by the individual had they completed the survey. The model may be a concatenated model that is a combination of a language model output from a language classifier recurrent neural network, and an acoustic model output from an acoustic feature layer convolutional neural network.

Learning based metric determination for service sessions

Systems and methods balance loads of traffic directed to various response entities. Systems and methods alternatively or complementarily generate simulated loads of traffic for handling by various response entities or load balancing algorithms.

Systems and methods solve functions relating to load balancing, call routing, and costs in call center networks for specific parameters. Systems and methods can also utilize machine learning to provide specific parameters relating to load balancing, call routing, and costs in call center networks.

Call center load balancing and routing management

The method may include simulating a call and determining a propensity score associated with the call. The method may also include selecting a first destination, from a plurality of destinations, for the call based on a routing algorithm and the propensity score. The method may include routing the call to the first destination and updating a first call load associated with the first destination based on the call. The method may also include comparing the first call load to a plurality of other call loads. Each of the other call loads may be associated with at least one of the plurality of destinations. The method may also include updating the routing algorithm based on the comparing.

Growing community of inventors

San Antonio, TX, United States of America

Bipin Chadha