International Business Machines Corporation

Other

Kyndryl, Inc.

Thomas Benjamin Harrison

Brian M O'Connell

Aaron K Baughman

Herbert Daniel Pearthree

Nicholas Albert McCrory

Michelle Welcks

Richard J Bogdany

Cody A Clements

Michael P Lazos

Robert J Lewis

Mary C Zabriskie

Clay T Upton

Shengzhi Sun

Cameron Niall-Kerr McAvoy

Cameron Naill-Kerr McAvoy

A computer processor generates a topic-based dataset based on parsing content received from a plurality of information sources, which includes historical data and scientific data, associated with a location of a natural resource. The processor generates a plurality of clusters, respectively corresponding to like-topic data of the topic-based dataset. The processor determines a plurality of hypotheses, respectively corresponding to the plurality of clusters of the like-topic data, wherein the plurality of hypotheses are based on features associated with each of the plurality of clusters of the like-topic data. The processor combines pairs of clusters, based on a similarity heuristic applied to the one or more pairs of clusters, and the processor determines a plurality of probabilities respectively corresponding to a validity of each hypothesis of the plurality of hypotheses, associated with the location of a natural resource.

Probability mapping model for location of natural resources

Social media inputs specific to a content of a broadcast program and received from a plurality of users are analyzed to determine a sentiment value expressed relative to a portion of the content. A group is formed of those users whose sentiment values expressed relative to the portion of the content are within a specified tolerance value of sentiment value and are expressed within a defined period during the broadcast of the program. A historical data from the inputs unrelated to the content is analyzed to extract an entity. A time zone of a geographical location of the entity is associated with the group. a sentiment of the group expressed relative to the portion of the content is shifted by an amount of time on a program continuum timeline, such that the sentiment of the group aligns with a time when the portion was broadcast in the time zone.

Event affinity detangling with broadcasts rolling over time zones

For distributed processing using drift-based dynamic clustering of Internet of Things (IoT) devices, at a central device, a data source to be used for processing a workload is determined. A set is selected of devices operating within a threshold distance from the data source at a first time. A first subset of the set of devices is selected to form a cluster of devices. Each device in the first subset satisfies a clustering condition. A first device in the first subset is instructed to configure an application at the first device to participate in the cluster and process the workload. From a performance check on the first device, a change is discovered in a performance metric. In response to the change resulting from an increased demand for a computing resource at the first device, the first device is replaced with a second device from the first subset.

Distributed load processing using drift-based dynamic clusters of internet of things devices

Approaches are provided for answering an inquiry of a cognitive distributed network. An approach includes receiving the inquiry at the cognitive distributed network. The approach further includes determining a classification for the inquiry based on natural language of the inquiry. The approach further includes classifying the inquiry as a single question class. The approach further includes determining, by at least one computing device, a type of introspection to be used by the cognitive distributed network on the inquiry. The approach further includes generating an answer to the inquiry based on the determined type of introspection.

Cognitive distributed network

Social media inputs from users, and specific to a content of a broadcast program, are analyzed to determine a sentiment value expressed relative to a portion of the content by a user. A group is formed of those users whose sentiment values expressed relative to the portion within a defined period during the broadcast of the program are within a specified tolerance value of sentiment value expressed by the user. A time zone of a location extracted from a social media input of the user is assigned to the group. A fabricated event that evokes a predicted sentiment is inserted in the broadcast program. A confidence value for the time zone is computed by comparing a sentiment value of a response by the group to the fabricated event with the predicted sentiment value. The group's placement in the time zone is confirmed when the confidence value exceeds a threshold.

Location-based event affinity detangling for rolling broadcasts

At a central device, a data source to be used for processing a workload is determined. A set is selected of devices operating within a threshold distance from the data source at a first time. A first subset of the set of devices is selected to form a cluster of devices. Each device in the first subset satisfies a clustering condition. A first device in the first subset is instructed to configure an application at the first device to participate in the cluster and process the workload. From a performance check on the first device, a change is discovered in a performance metric. In response to the change resulting from an increased demand for a computing resource at the first device, the first device is replaced with a second device from the first subset.

Distributed load processing using drift-based dynamic clusters of Internet of Things devices

Embodiments of the present invention provide an approach for allocating computing resources based on social networking/media trends in a networked computing environment (e.g., a cloud computing environment). In a typical embodiment, a baseline computing resource allocation will be determined for the networked computing environment based upon historical computing resource data (e.g., stored in at least one computer storage device). Social networking trend data corresponding to usage of a set of social networking websites may be received and analyzed to determine a forecasted computing resource allocation (e.g., based on social networking trends). The baseline computing resource allocation may be compared to the forecasted computing resource allocation to identify any difference therebetween. A computing resource allocation protocol/plan may then be determined based on the comparison (e.g., to address the difference).

Resource allocation based on social networking trends in a networked computing environment

For distributed processing using forecasted location-based IoT device clusters, at a central IoT device, a data source that is to be used and a duration for processing a workload is determined. A set of IoT devices operating within a threshold distance from the data source at a first time is selected. A subset of the set is selected to form a sub-cluster of IoT devices where a forecasted travel path of a member IoT device in the subset keeps the member within the threshold distance from the data source for the duration. A lightweight application is configured at a first IoT device in the subset which enables the first IoT device to participate in the sub-cluster and process the workload.

Distributed load processing using forecasted location-based internet of things device clusters

A method and associated systems for optimizing a computing platform. A processor joins sets of configurable parameters into groups that each identifies a configuration of the computing environment or of a component or subsystem of the computing environment. The processor generates a set of variations of each group, where each variation identifies a candidate configuration of the component, subsystem, or platform, and where each candidate configuration identifies a distinct set of values of the group of parameters associated with that component, subsystem, or platform. Each configuration of this first generation of configurations undergoes a massively parallel iterative procedure that generates a next generation of configurations by performing operations upon the first generation that are similar to those of a natural-selection process. The procedure repeats until successive generations converge, within resource constraints, to a fittest generation that represents an optimal or most nearly optimal configuration of the computing platform.

Optimizing a virtualized computing environment

For distributed processing using location-based IoT device clusters, using a processor and a memory at a central IoT device, a data source that is to be used for processing a workload is determined. A set of IoT devices that are operating within a threshold distance from the data source at a first time is selected. At the central IoT device, to form a cluster of IoT devices, a subset of the set of IoT devices is selected. Each IoT device in the subset satisfies a clustering condition. The processor at the central IoT device is instructed to configure a device application at a first IoT device in the subset of IoT devices, the device application enabling the first IoT device to participate in the cluster and process the workload.

Distributed load processing using location-based internet of things device clusters

For distributed processing using clustering of interdependent Internet of Things (IoT) devices, at a central device, a data source to be used for processing a workload is determined. A set is selected of devices operating within a threshold distance from the data source at a first time. A first subset of the set of devices is selected. Each device in the first subset satisfies a clustering condition. A first device in the subset is instructed to configure a lightweight application to participate in the cluster and process the workload. The processing of the workload is halted on a second device, where the first device has a processing dependency on the second device in processing the workload. A preserved current state of processing the workload is transferred from the first device to a third device. The processing of the workload is continued using the second device and the third device.

Distributed load processing using clusters of interdependent internet of things devices

For distributed processing using forecasted location-based IoT device clusters, at a central IoT device, a data source that is to be used and a duration for processing a workload is determined. A set of IoT devices operating within a threshold distance from the data source at a first time is selected. A first subset of the IoT devices is selected to form a cluster of IoT devices where each IoT device satisfies a clustering condition. A second subset of the first subset is selected to form a sub-cluster of IoT devices where a forecasted travel path of a member IoT device in the second subset keeps the member within the threshold distance from the data source for the duration. A lightweight application is configured at a first IoT device in the second subset which enables the first IoT device to participate in the sub-cluster and process the workload.

An approach is provided for allowing a network computing (e.g., cloud computing) infrastructure to modify its resource allocation plan (e.g., an instance count) by using a Kth derivative vector plot, which may be generated using historical logs. Among other things, this approach enables an infrastructure to project an allocation forecast for a specified duration and adapt to changes in network traffic.

Growing community of inventors

Holly Springs, NC, United States of America

Thomas Benjamin Harrison