	US 12,079,821 B2
	Assistance for customer service agents
James Ellison, Issaquah, WA (US); Mark Hanson, Plano, TX (US); Joel Werdell, Seattle, WA (US); Stephen King, Seattle, WA (US); Christopher Mills, Bellevue, WA (US); Phoebe Parsons, Redmond, WA (US); Kasey Snow, Renton, WA (US); and Rudy Bourcelot, Everett, WA (US)
Assigned to T-Mobile USA, Inc., Seattle, WA (US)
Filed by T-Mobile USA, Inc., Bellevue, WA (US)
Filed on May 11, 2021, as Appl. No. 17/317,765.
Claims priority of provisional application 63/032,438, filed on May 29, 2020.
Claims priority of provisional application 63/023,077, filed on May 11, 2020.
Prior Publication US 2021/0350385 A1, Nov. 11, 2021
Int. Cl. G06Q 30/016 (2023.01); G06N 20/00 (2019.01)

CPC G06Q 30/016 (2013.01) [G06N 20/00 (2019.01)]

18 Claims

1. A computer-implemented method, comprising:

receiving, by a computing device, customer interaction data that reflects an interaction between a first user and a second user;

providing, by the computing device, the customer interaction data as an input to a first machine learning-trained model that is configured to generate a first model output that includes a characteristic of the interaction between the first user and the second user;

receiving, by the computing device and from the first machine learning-trained model, the first model output that includes the characteristic of the interaction between the first user and the second user;

accessing, by the computing device, a model publishing service that is configured to publish model outputs from the first machine-learning trained model to micro-applications;

based on accessing the model publishing service that is configured to publish model outputs from the first machine-learning trained model to micro-applications, determining, by the computing device, that a first micro-application that is presenting data to the second user subscribes to the model outputs of the first machine learning-trained model;

based on determining that the first micro-application that is presenting data to the second user subscribes to the model outputs of the first machine learning-trained model, providing, by the computing device and to the first micro-application, the first model output that includes the characteristic of the interaction between the first user and the second user;

providing, by the computing device, (i) the first model output that is output by the first machine learning-trained model and that includes the characteristic of the interaction between the first user and the second user and (ii) the customer interaction data as inputs to (i) a second machine learning-trained model that is configured to generate a second model output that includes instructions for the second user to continue interacting with the first user during the interaction between the first user and the second user and (ii) a third machine learning-trained model;

receiving, by the computing device and from the second machine learning-trained model, the second model output that includes the instructions for the second user to continue interacting with the first user during the interaction between the first user and the second user;

receiving, by the computing device and from the third machine learning-trained model, the third model output;

accessing, by the computing device, the model publishing service that is configured to (i) publish model outputs from the second machine learning-trained model to micro-applications and (i) publish model outputs from the third machine learning-trained model to micro-applications;

based on accessing the model publishing service that is configured to publish model outputs from the second machine learning-trained model to micro-applications, determining, by the computing device, that a second micro-application that is presenting data to the second user subscribes to the model outputs of the second machine learning-trained model;

based on accessing the model publishing service that is configured to publish model outputs from the third machine learning-trained model to micro-applications, determining, by the computing device, that no micro-applications subscribe to the model outputs of the third machine learning-trained model;

based on determining that the second micro-application that is presenting data to the first user or the second user subscribes to the model outputs of the second machine learning-trained model, providing, for output to the second micro-application, the second model output that includes the instructions for the second user to continue interacting with the first user during the interaction between the first user and the second user; and

based on determining that no micro-applications subscribe to the model outputs of the third machine learning-trained model, disregarding, by the computing device, the model outputs of the third machine learning-trained model.