Amazon Technologies, Inc.

Thomas Boyd Johnston

Pasquale DeMaio

Wei Niu

Attila Narin

Jon R Jay

Yung-Chun Lin

Juliana Saussy

David Arthur Stanford, Jr

Paul Horvath

Tak Chung Lung

Erin Nicole Rifkin

Phillip H Keung

Vinay Kale

Spyridon Garyfallos

Jing Zhang

Yiwei Li

Manjeshwar Donthi

Richard Julian Rodriguez

Saurabh Dwivedi

Max Moon Ramsay

Lawrence Mark Pizette

Paul Nau

Johnathon Paul Anthony Meichtry

Michael Lee Garza

Philip Fitzsimons

Correcting time series predictions for seasonality using trends may be performed. An initial time series prediction for a period of time using a machine learning model trained on a time series may be made. Data of the time series from more than an analysis time period prior to the period of time may be obtained. The time series prediction and the obtained data of the time series may be decomposed in respective components including a trend component. The respective trend component of the obtained data may evaluated to determine whether a trend replacement criteria is satisfied. The respective trend component of the time series prediction may be replaced with a portion corresponding to the period of time in a prior analysis time period from the respective component of the obtained data of the time series to correct the time series prediction for seasonality.

Correcting time series predictions for seasonality with trends

Methods, systems, and computer-readable media for accurate usage forecasting for virtual contact centers are disclosed. A contact center management system configures a contact center instance associated with a client. The contact center instance comprises a plurality of queues configured to store contacts. At least a portion of the contacts are routed to a plurality of agents. The contact center management system determines, using one or more machine learning models associated with the contact center instance, a plurality of predictions for a plurality of metrics for the contact center instance for a plurality of time horizons. At least a portion of the predictions are generated for individual queues of the contact center instance.

Accurate individual queue level metric forecasting for virtual contact center queues with insufficient data, using models trained at higher granularity level

This disclosure describes management of a contact center executing in a service provider network. The management may be performed by using one or more trained deep learning/machine learning (ML) models. The ML models may be trained using metrics and data gathered from the various services and systems in the service provider network used to implement the contact center. The trained models may be used for forecasting staffing needs, e.g., agents, for the contact center, scheduling agents, detecting anomalies with regard to customer traffic, e.g., received communications from customers, at the contact center, adherence of agents to the scheduling, and dynamically routing of received communications to appropriate agents (e.g., based on skills of the agent and other factors) for handling of the communications.

Dynamic communication routing at contact centers

Technology is described for providing assessments of resource deployments in a computing service environment. A group of related resources may be identified in a computing service environment. The group of related resources may be associated with a single customer. A rating for the group of related resources may be determined based in part on design characteristics of the group of related resources and runtime behavior of the group of related resources. A resource configuration recommendation may be provided based on the rating. The resource configuration recommendation may include alternative resource configurations to improve the rating for the group of related resources.

Assessing and rating deployments of resources

Barrier for isolation of metal structures

Growing community of inventors

Seattle, WA, United States of America

Thomas Boyd Johnston