Abstract: An embodiment includes specifying an original model to be trained by a machine learning platform using training data and an algorithm, wherein the training data includes a plurality of training input values upon which to make a prediction and a plurality of training output values representing target prediction values. The embodiment generates an updated model by training the original model to algorithmically approach the training output values as provided with the training data based on the plurality of input values. The embodiment generates an updated function map representative of output values of the updated model and versions the updated model according to a classification rule for assigning a version classification based on a comparison of the updated function map to an original function map representative of output values of the original model. The embodiment deploys the updated model into a production environment. to replace the original model.

Abstract: Embodiments of the present invention provide methods, computer program products, and systems. Embodiments of the present invention can be used to, in response to response to anticipating a maneuver to be performed by a vehicle, adjust one or more 5G dedicated traffic channels (DTCH) over a radio bearer network. Embodiments of the present invention can be used to revert the one or more adjusted 5G DTCH after the maneuver is performed by the vehicle.

Abstract: A method, a computer system, and a computer program product are provided. A computing system receives natural language first textual input from a second computing system and performs natural language processing. The natural language processing identifies vagueness and missing information in the first textual input. Additional information is obtained from at least one source responsive to identifying at least one of the vagueness and the missing information. The identified vagueness and the missing information are resolved based on the obtained additional information and a meaning of the first textual input is determined. Output is provided to the second computing system for presentation to a user.

Abstract: A 5G network conference call initiation is detected. Information associated with the conference call includes a topic, an agenda, participants, and goals, is received. The respective participants of the conference call are identified, based on a dedicate traffic channel (DTCH) of the respective participants of the conference call. An allocation threshold of conference call contribution associated with the respective participants is determined, based on the information received. Contributions of respective participants are calculated during the conference call based on a list of DTCHs of the respective participants. The contributions of the respective participants are compared to the allocation threshold and responsive to determining that the contributions of a first participant of the respective participants of the conference call exceed the allocation threshold, an action mitigating the excessive contribution in the conference call is performed.

Abstract: In an approach to defining and symphonizing serverless functions of hybrid multi-cloud services, responsive to receiving a message to deallocate one or more bearer channels, a count of deletion requests is retrieved, where the count of deletion requests includes the one or more deallocated bearer channels. Responsive to determining the type of bearer channel deleted, an appropriate connected device is selected, where the appropriate connected device is chosen from the group consisting of a connected user equipment, a peer base station, and a gateway. A hinting message is sent to the appropriate connected device that the one or more deallocated bearer channels are available.

Abstract: An approach to optimizing utilization of virtual agents within a virtual agent system. The approach may include monitoring the processing loads of virtual agents and identifying highly utilized virtual agents. The approach may also include configuring a pathway which directs a user query to the identified highly utilized virtual agent and allowing the highly utilized virtual agent to respond to the user query if the highly utilized virtual agent is capable of generating a satisfactory response. Additionally, the approach may include sending the user query to one or more other virtual agents if the highly utilized virtual agent is unable to generate a response above a confidence threshold.

Abstract: A method, system, and computer program product for generating and presenting an offline video presentation is provided. The method includes executing service software code within a service orchestration and programmability framework layer of a telecom network. Communications with static and dynamic hardware and software resources are established and dynamic video presentations with respect to channels generated with respect to selective broadcast control channels of the telecom network are generated. Video contents of the dynamic video presentations are retrieved logical channels associated with virtual network functions are generated. Information associated with the video contents fare transferred from a channel to physical network functions of the telecom network. Connections associated with detecting offline video streams are allocated and a connection to a logical network is established. The video contents are associated with an offline video stream in response to the connection.

Abstract: A method, system and computer-usable medium are disclosed for proactive data hinting in a telecom network. In certain embodiments, the invention relates to receiving a data hinting request for an application at a virtual network framework of the telecom network. A dedicated hinting channel is selected to provide the proactive data hinting; monitoring for the hinting request. Data is moved from the resource, such as edge cloud computing, where the application is implemented when the proactive data hinting is received. The application is used by user equipment, access points and Internet of Things (IoT) devices.

Abstract: A method, system and computer-usable medium are disclosed for proactive data hinting in a telecom network. In certain embodiments, the invention relates to receiving a data hinting request for an application at a virtual network framework of the telecom network. A dedicated hinting channel is selected to provide the proactive data hinting; monitoring for the hinting request. Data is moved from the resource, such as edge cloud computing, where the application is implemented when the proactive data hinting is received. The application is used by user equipment, access points and Internet of Things (IoT) devices.

Abstract: A method, a computer system, and a computer program product are provided. A computing system receives natural language first textual input from a second computing system and performs natural language processing. The natural language processing identifies vagueness and missing information in the first textual input. Additional information is obtained from at least one source responsive to identifying at least one of the vagueness and the missing information. The identified vagueness and the missing information are resolved based on the obtained additional information and a meaning of the first textual input is determined. Output is provided to the second computing system for presentation to a user.

Abstract: A processor may receive a media frame collection. The media frame collection may include two or more frames. The processor may identify, from the media frame collection, a location of an entity. The processor may determine whether the location of the entity is within a threshold location in regard to the other two or more frames in the media frame collection. The processor may display the one or more frames that exceed the threshold location to a user.

Abstract: An approach includes a system implemented in a computer infrastructure including a processor. The approach further includes the processor configured to receive dimensionally aware linkages. The approach further includes the processor configured to retrieve aggregated data from a data set. The approach further includes the processor configured to identify at least one affinity cluster related to the dimensionally aware linkages in at least two dimensions and the retrieved aggregated data by performing at least one lookup linkage which matches at least one dimension of the at least two dimensions of the dimensionally aware linkages with the retrieved aggregated data. The approach further includes the processor configured to provide personalized content to a user having same or similar user preferences as the received dimensionally aware linkages based on the identified at least one affinity cluster.

Abstract: A method and system for virtually detecting and modifying technical issues is provided. The method includes enabling virtual software of a hardware controller with respect to a software layer of a fifth generation (5G) telecom communication network. A signal is enabled by a 5G apparatus and a handshaking processes between the hardware controller and the 5G apparatus is executed. In response, metadata structures are parsed and classified. Technical issues associated with hardware and software of devices connected to the 5G telecom communication network are identified and the virtual software is modified for managing and repairing the technical issues. Self-learning software code for executing future processes associated with executing virtual detection and issue modification is generated.

Abstract: A middleware at a proxy node may cause a copy of a data object stored on a storage node to be updated. Copies of the data object may be stored on a plurality of storage nodes located in first and second geographic locations. A first processing activity for a storage node that corresponds with the application category of the data object may be determined. A first storage node may be selected for a first update based on being associated with the first processing activity in a first proportion. The copy of the data object stored at the first storage node may be updated before updating a copy of the data object stored at a second storage node. The second storage node may be associated with the first processing activity in a second proportion, with the second proportion being less than the first proportion.

Abstract: In an approach to training image classification models through the creation and utilization of stub images, one or more computer processors tag ground truth data utilizing subject matter expert inputs for a classification. The one or more computer processors extract one or more features and one or more boundaries from the tagged ground truth data. The one or more computer processors tag the one or more extracted features and the one or more extracted boundaries. The one or more computer processors generate one or more stub images by incorporating the one or more tagged features, and the one or more tagged boundaries at a confidence level. The one or more computer processors train a classification model utilizing the one or more generated stub images. The one or more computer processors deploy the trained classification model.

Abstract: A requirement process document is received. The requirement process document specifies a plurality of requirements for developing a product. Each of the plurality of requirements is classified, using machine learning, as either a business requirement or a technical requirement. A hierarchy including a plurality of levels of business requirements and technical requirements is determined from the requirement process document. A particular level in the hierarchy for each of the business requirements and technical requirements and one or more relationships between the business requirement and the technical requirement within the hierarchy is identified. A realizing component and associated capability for implementing one or more of the technical requirements is identified.

Abstract: A computer-implemented method includes: receiving, by a computer device, and from a head position sensor, head position information that indicates a position of a user's head; receiving, by the computer device, and from an eye position sensor, eye position information that indicates a position of a sensed eye of the user; producing, by the computer device, eye image data that represents a desired eye image based on the head position information and the eye position information; and transmitting, by the computer device, the eye image data to a wearable device worn by the user.

Abstract: An approach is provided in which a cognitive system engages in a conversation that includes multiple participants. The cognitive system detects a diversion indicator during the conversation that is a trigger to the cognitive system perform a diversion action. The cognitive system identifies the diversion action that corresponds to the diversion indicator and performs the diversion action during the conversation accordingly.

Abstract: A first user communication request by a first user is detected. The first user communication request is directed to a computer response system. In response to detecting the first user communication request a profile associated with the first user is determined. A first set of computing resources of the computer response system is allocated to a first automated respondent. The allocation is based on the determined profile. A communication between the first user and the first automated responded is assigned based on the determined profile.

Abstract: A computer-implemented method according to one embodiment includes receiving a training data set to be applied to a model; selecting a subset of the training data set as a sample set; for each of a plurality of predetermined augmentations, applying the predetermined augmentation to the sample set to create an augmented sample set, training the model with the augmented sample set, determining a performance of the trained model, and assigning a weight to the predetermined augmentation for the training data set, based on the determined performance; and selecting one or more of the plurality of predetermined augmentations to be applied to the training data set before the training data set is applied to the model, based on the weight assigned to each of the plurality of predetermined augmentations.