AT&T Intellectual Property I, LP

James Fan

Jennifer K Lam

Richard T Kuo

Jeffrey Aaron

David Chen

Thomas Barrett

Cheng P Liu

Venson Shaw

Alireza Hooshiari

Dan Celenti

Deva-Datta Sharma

Sanjay Agraharam

Steven R Polston

David Feng-Lin Chen

Michelle Martens

Eric Towell

Arun Gupta

Hieu Le

Lipton Chin-Loy

Chandrasekaran Sundaramurthy

Kevin Xiangkai Meng

Chou Lan Pok

Vimal Begwani

Stephen Kurpinsky

Al Hooshiari

Eric Forbes

Arthur Richard Brisebois

Rittwik Jana

Raghvendra Savoor

Michael Raftelis

Srilal Weerasinghe

Manoop Talasila

Dick Kwan

Cheng P Lui

Ojus Parikh

Max Faulkner, Jr

Concepts and technologies disclosed herein are directed to interpretation workflows for machine learning-enabled event tree-based diagnostic and customer problem resolution. According to one aspect, a system can receive a workflow construction specification derived from a machine learning-enabled event tree ('MLET'). The MLET can be generated for use by a customer service agent to resolve a customer problem. The workflow construction specification can include a plurality of objects, each of which represents a navigation path through the MLET. The system can traverse the workflow construction specification and can create a set of workflow creation commands based upon at least one policy. The system can generate a workflow visualization interpretation file based upon the set of workflow creation commands. The workflow visualization interpretation file can identify how the MLET derived a root cause of the customer problem. The system can then present the workflow visualization interpretation file to the customer service agent.

Interpretation workflows for machine learning-enabled event tree-based diagnostic and customer problem resolution

An example method includes receiving a plurality of data request specifications from a plurality of user endpoint devices of requesting users, aggregating a subset of the plurality of data request specifications which share a similarity into a super request, retrieving a plurality of datasets, wherein the plurality of datasets is needed to fulfill the super request, aggregating the plurality of datasets into a single data superset, and delivering the single data superset to the plurality of user endpoint devices of the requesting users via a data pipeline, where the delivering includes segregating the single data superset so that each user endpoint device of the plurality of user endpoint devices of the requesting users receives a portion of the single data superset corresponding to one data request specification of the plurality of data request specifications that was submitted by the each user endpoint device.

Secure pipeline-based data delivery

A processing system including at least one processor may obtain a first ontology of a first type of data pipeline component, map the first ontology to a second ontology for a second type of data pipeline component that is stored in a catalog of data pipeline component types, provide a second data schema for the second type of data pipeline component as a template for a first data schema for the first type of data pipeline component, and add the first type of data pipeline component to the catalog of data pipeline component types, where the adding comprises storing the first ontology and the first data schema for the first type of data pipeline component in the catalog of data pipeline component types.

Data pipeline controller

A method includes creating one or more first policy shims to be applied to a ML/AI module, applying the one or more first policy shims to an input or an output of the ML/AI module and executing the ML/AI module on a data set in response to the applying step. The one or more first policy shims includes an input policy shim and an output policy shim and the applying step includes applying the input policy shim to the data set prior to the executing step and applying the output policy shim to an output of the executing step.

Public policy rule enhancement of machine learning/artificial intelligence solutions

Concepts and technologies disclosed herein are directed to managing enterprise software licenses for virtual network functions ('VNFs'). According to one aspect disclosed herein, a system can acquire a software license for a software asset to be instantiated and used by a cloud computing environment associated with an enterprise. The system can prepare, with an enterprise anchor point ('EAP') module managed by a vendor of the software asset, a certificate validation process that uses an enterprise security certificate to ensure the software license is valid for an execution instance of the software asset. The system can instantiate the execution instance of the software asset in the cloud computing environment. The system can validate, by the EAP module, the enterprise security certificate to ensure the software asset is instantiated and used in accordance with the software license.

Managing enterprise software licenses for virtual network functions

Concepts and technologies disclosed herein are directed to a flexible behavioral chain ('FBC') framework for permission-based enterprise-focused blockchain applications. According to one aspect disclosed herein, a chain framework module ('CFM') executed by a FBC system can receive a new block request to add a new block to an FBC. The new block request can be created by a trigger management module (&#x201c;TMM&#x201d;) executed by the FBC system in response to a trigger, such as a customer touchpoint, a customer engagement, a subscribed event, and/or a virtual event. In response to the new block request, the CFM can generate the new block. A customer value score derivation module (&#x201c;CVSDM&#x201d;) executed by the FBC system can determine a customer value score (&#x201c;CVS&#x201d;) for a customer. The CFM can incorporate the CVS into the new block. The CFM can connect the new block to the FBC.

Flexible behavioral chain framework for permission-based enterprise-focused blockchain applications

A method includes defining a plurality of variables to modify in a control loop; collecting first data using a first variable of the plurality of variables while executing the control loop, generating a first result based on the collecting first data step, substituting a second variable of the plurality of variables for the first variable, collecting second data using the second variable while executing the control loop, generating a second result based on the collecting second data step, comparing the first result and the second result; and taking an action based on the comparing step.

Automatic control loop decision variation

A method includes receiving, by a computing device of a media service provider, a search request associated with a first product. The search request is received from a media device associated with a subscriber of a service provided by the media service provider. The method includes, in response to the search request, determining, by the computing device, a responsible party associated with the first product based on a purchase date of the first product and a return period for the first product. The method also includes sending, from the computing device to the media device, contact information of the responsible party. The contact information is to be displayed by the media device to a display device with an option to contact the responsible party.

System and method of providing post-purchase media content to a subscriber

A processing system including at least one processor may obtain a first request for delivery of a first data set to a first destination, map the first request to a first information model, obtain a second request for delivery of a second data set to a second destination, map the second request to a second information model, and identify that a portion of data is part of both data sets. The processing system may next determine a plan for configuring data pipeline components for delivering the first data set to the first destination and the second data set to the second destination, the plan comprising: a combination of the first information model and the second information model, and at least one modification to the combination. The processing system may then configure the data pipeline components in accordance with the plan.

Data blending for multiple data pipelines

Concepts and technologies disclosed herein are directed to automated control loop searching ('ACLS'). According to one aspect disclosed herein, an ACLS system can create a search model that provides high-level information regarding what the ACLS system should search for when a search pattern is detected within data that is output from execution of a control loop. The ACLS system can activate a control loop system that executes the control loop to yield the data as output. The ACLS system can detect the search pattern within data, and in response, the ACLS system can execute, based upon the search model, a search of the data. The ACLS system can collect search results of the search and select additional data from the search results.

Automatic control loop searching

A processing system may obtain a request to deploy a service comprising a first virtual network function (VNF) to a multi-cloud environment comprising cloud infrastructures, the request including metadata associated with the service, decompose the service into a plurality of modules comprising: a plurality of VNFs and/or a plurality of components of the first VNF, select at least a first deployment model for the service from among a plurality of deployment models, based upon the metadata, each deployment model comprising rules for assigning the plurality of modules to cloud infrastructures, and deploy a first module to a first cloud infrastructure and a second module to a second cloud infrastructure.

Application deployment in multi-cloud environment

A system that incorporates teachings of the present disclosure may include, for example, receiving location information associated with a mobile communication device, determining a first location of the mobile communication device based on the location information, selecting a first camera from a group of cameras based on the determined first location, receiving at least one first image from the selected first camera that captures at least a portion of the first location, performing image recognition on at least one second image to identify a user associated with the mobile communication device, selecting another camera from the group of cameras based on a determined position of the identified user, and receiving at least another image from the selected other camera. Other embodiments are disclosed.

Method and apparatus for managing a camera network

Fast-aiding radio access network intelligent controllers (RICs) can assist other RICs that are experiencing performance issues. The system can take into account the availability of other RICs that can aid the RIC needing assistance. Therefore, multiple RICs cooperatively working together can generate synergies, and allow differentiated RICs to operate in both predefined and/or dynamically create RIC groups. Once a RIC that needs assistance is identified, other RICs can offload functionalities that are less important than the functionality of the RIC that is experiencing the performance issues. After the assisting RIC has offloaded its functionalities, it can then devote resources to the RIC that needs the assistance.

Facilitation of fast aiding radio access network intelligent controllers for 5G or other next generation network

In one embodiment, a method includes monitoring, by a control loop including a processor and a memory, a first environment. The control loop includes one or more predetermined control loop parameters. The method also includes receiving, by the control loop and in response to monitoring the first environment, first data from the first environment and receiving, by the control loop, information from an adaptation control loop. The method also includes determining, by the control loop, to automatically adjust at least one of the one or more predetermined control loop parameters based at least in part on the information received from the adaptation control loop and automatically adjusting, by the control loop, the one or more predetermined control loop parameters. The method further includes determining, by the control loop, to initiate an action based on the first data collected from the first environment and the one or more adjusted control loop parameters.

Growing community of inventors

San Ramon, CA, United States of America

James Fan