Abstract: The illustrative embodiments provide for adaptation of reference architecture documents across clouds. An embodiment includes loading a reference architecture document. The embodiment also includes populating a knowledge repository corpus where the knowledge repository corpus includes data from information technology (IT) products, cloud services, and integrations stored with associated metadata about functionalities. The IT products, the cloud services, and the integrations are stored with metadata. The embodiment also includes classifying, using an incident analysis module, the service incidents based on the reference architecture document. The embodiment also includes detecting flaws in the reference architecture document based on the classified service incidents. The embodiment also includes creating a weighted graph of the cloud services. The embodiment also includes traversing the weighted graph and creating an adapted reference architecture based on the traversed weighted graph.

Abstract: An embodiment for altering a machine based on a predicted problem in a multi-machine environment is provided. The embodiment may include receiving data relating to an activity and functionalities of a primary machine in a multi-machine environment. The embodiment may also include executing a digital twin simulation of a digital twin model of the primary machine performing the activity in accordance with the functionalities. The embodiment may further include identifying a problem condition with at least one component of the primary machine in executing one or more steps of the activity. The embodiment may also include in response to determining the primary machine is not capable of performing an alternative activity with the problem condition, deploying one or more secondary machines having at least one substitute component for the at least one component of the primary machine to collaborate with the primary machine.

Abstract: Embodiments of the present invention provide an approach for event damage assessment. Specifically, the approach provides for obtaining sensor data from objects with sensors in an activity area and detecting events based on this data. A real-time digital simulation is generated to estimate the damages caused by the detected event. A directed acyclic graph is created to model similar scenario causation factors, which is then refined using data analysis and simulations to create an augmented directed acyclic graph. A neural network architecture is applied to the augmented directed acyclic graph to classify faults. An insurance claim amount is calculated based on the estimated damages amount and the classified fault.

Abstract: Systems, computer program products and/or computer-implemented methods described herein relate to balancing agriculture risk with green energy practices to reduce a carbon footprint. A system can comprise a memory that stores computer executable components and a processor that executes the computer executable components. The computer executable components comprise an analysis component that determines crop metrics and terrace metrics for a terrace slope farming system by employing imaging data, a prediction component that generates, based on the crop metrics and terrace metrics, a crop-terrace pairing for a terrace of the terrace slope farming system, wherein the crop-terrace pairing is defined by one or more of the crop metrics that match to one or more of the terrace metrics, and an evaluation component that identifies the crop-terrace pairing as resulting in a first carbon footprint that is lower than a second carbon footprint that is identified for the same terrace.

Abstract: An example operation may include one or more of executing queries on one or more external data stores to retrieve domain data of one or more ranked lists of assets and variables corresponding to the domain data, generating a plurality of feature groups based on the retrieved domain data and variables, wherein the plurality of feature groups correspond to a plurality of features used to generate the one or more ranked lists, converting the plurality of features groups into a plurality of values via execution of an optimization engine, transforming the plurality of values into a plurality of sentences describing the plurality of feature groups, and displaying the plurality of sentences via a user interface.

Abstract: An embodiment for blending musical phenotypes with user selections is provided. The embodiment may include receiving a corpus of songs selected by a user and one or more preferences of the user relating to musical interests. The embodiment may also include converting each song in the corpus into a spectrogram. The embodiment may further include encoding each spectrogram into a chromosomal representation of integers. The embodiment may also include creating pools of chromosomal positions. The embodiment may further include processing the pools of chromosomal positions into a gene representation. The embodiment may also include translating the gene representation into a phenotype expression.

Abstract: An approach is provided for optimizing application caching and locking. Features specifying an operating environment of an application are extracted. The features include actual and forecasted central processing unit usage and memory, disk, and network pressure. A pairwise set of class-based and method-based ASTs and the extracted features are input into a logical neural network. Symbolic feature vectors are generated for the features by establishing bounds and flattening the features. The symbolic feature vectors and the set of class-based and method-based ASTs are input into a stacked transformer having encoders and decoders. The encoders and decoders are trained on word or token distributions of code ASTs and operating environment bounds associated with the ASTs. Using the stacked transformer, code is generated for replacing a portion of a method represented by a method-based AST. The code adds or changes caching or locking in the application.

Abstract: A method, a structure, and a computer system for diverse natural language generation. The exemplary embodiments may include training a data-to-text neural network (D2T NN) and training a text-to-text neural network (T2T NN), wherein the D2T NN and the T2T NN have identical transformer architectures, and wherein the training of the D2T NN and the T2T NN are indexed by time. The exemplary embodiments may further include interleaving weights between the D2T NN and the T2T NN, as well as generating a sentence based on the interleaved D2T NN and T2T NN.

Abstract: A computer implemented method processes log entries. A number of processor units identify log entries. The number of processor units determines whether anomalous content is present in the log entries. The number of processor units suppresses the anomalous content to form suppressed content in response to determining that the anomalous content is present in the log entries. According to other illustrative embodiments, a computer system and a computer program product for processing log entries are provided.

Abstract: A method, system, and computer program product for implementing machine learning natural language digital package generation and selection is provided. The method includes receiving from hardware and software sources, factual data associated with an event. In response, natural language digital templates comprising natural language phrase variants for each portion of the factual data is generated. Factual data phrases are generated and packaged into digital packages including at least one natural language phrase variant with respect to each portion of factual data. An initial package is selected by minimizing a number of repetitions of the factual data phrases across the digital packages and digital summaries are extracted. Alignment attributes associated with the digital summaries are determined with respect to the initial package and a final package is selected. A hardware device is enabled for presenting a video stream including the final package with respect to the event.

Abstract: A network usage plan is constructed using a network usage prediction model. The network usage plan comprises a first time and a first desired network configuration to be implemented at the first time. A network is configured according to the first desired network configuration for use at the first time. Responsive to determining, during a time period in which the network is configured according to the first network configuration, that a usage requirement has changed more than a threshold amount, the network is reconfigured according to a second network configuration meeting the changed usage requirement.

Abstract: A computer-implemented method for determining context granularity is provided. The computer-implemented method includes determining an activity for a user and a distance between the user and a location of the activity, ascertaining a contextual information granularity level for displaying contextual information relevant to the activity in an augmented reality (AR) interface in accordance with the distance, displaying the contextual information corresponding to the contextual information granularity level in the AR interface, updating the contextual information granularity level as the distance changes to obtain a new contextual information granularity level and displaying the contextual information corresponding to the new contextual information granularity level in the AR interface.

Abstract: Systems and methods are provided for recovering LED light in indoor multi-level farms. A plurality of cameras and sensors capture in real time in a vertical farm environment a plurality of camera images and sensor data. The camera images include physical images of the vertical farm environment, and the sensor data includes light data. The camera images and sensor data are analyzed to identify a pattern of light origin and direction in the vertical farm environment. The plant characteristics are identified based on comparing the captured camera images, using image recognition, against a model. Based on the identified plant characteristics, manipulating the light data. The plant characteristics include plant species growth state and general health. The light is manipulated by breaking it into component visible light wavelengths. The model predicts optimizing the redirecting of the optimal wavelengths for improved plant growth.

Abstract: The present inventive concept provides for a method of battery pack thermal and gaseous stress mitigation. The method includes obtaining data related to batteries within a battery pack. Features are extracted from the obtained data related to the batteries. The extracted features include effected batteries, battery positions, gas and temperature measurements, and gas and temperature thresholds. The extracted features are mapped. Effected battery patterns are identified. Space is created between the effected batteries and adjacent batteries based on the identified effected battery patterns.

Abstract: A method for detecting conflation errors in an augmented reality interface is disclosed. In one embodiment, such a method includes receiving an actual screenshot of an augmented reality interface. The method generates, using a generative adversarial network (GAN), a replica screenshot of the augmented reality interface that indicates how the actual screenshot is expected to appear. The replica screenshot is non-identical to the actual screenshot. The method receives the actual screenshot and the replica screenshot into a fractal-based convolutional neural network (CNN) to determine a correlation between the actual screenshot and the replica screenshot. In the event the correlation is below a designated threshold, the method flags the actual screenshot as having a potential conflation error. This conflation error may involve conflating a digital component for a physical component or vice versa. A corresponding system and computer program product are also disclosed.

Abstract: Within a video frame, elements are identified. A graph is constructed for a portion of video content including the video frame. Using the graph and an excitement level corresponding to an element in the plurality of elements, the video frame is divided into an alterable region and an unalterable region. By solving an optimization problem, a compute resource and a rendering application are selected, the compute resource represented by a runtime feature vector encoding a plurality of features describing execution of the rendering application on the compute resource. Using the compute resource and the rendering application, a background image corresponding to the alterable region is rendered. The unalterable region and the background image are combined into a rendered video frame, the rendered video frame replacing the video frame within the portion of video content.

Abstract: From data of a first interaction environment, user activity data while using the first interaction environment, and user biometric data while using the first interaction environment, a first cognitive load index is generated. A first directed acyclic graph corresponding to the first cognitive load index is generated. Using a graph convolutional network, the first directed acyclic graph, and a second directed acyclic graph generated from a second cognitive load index, a set of cognitive overload causation factors is generated. Using the set of cognitive overload causation factors, the first interaction environment is adjusted.

Abstract: A computer-implemented method includes generating two tensile circles based on a common circle created by overlapping two tensile spheres. An angle is determined using a modulo function and a predefined value. The angle is applied to both tensile circles. Next, multiplicands are determined for both tensile circles based on the angle applied to both tensile circles. The method then encrypts and/or decrypts data using a symmetric cryptography technique and the multiplicands.

Abstract: A computer-implemented method for detecting a change in a meta-environment is disclosed. The method comprises collecting historic parameter values from at least one actor being active in the meta-environment, wherein each of the at least one actors has a set of abilities represented by parameters relating to the collected parameter values, collecting current parameter values from the at least one actor being active in the meta-environment, determining weight values of the collected historic parameters and weight values of the collected current parameters, and upon a value of a function of the determined weight values of the historic parameters and the weight values of the collected current parameters being greater than a predefined threshold value, determining a detection of the change in the meta-environment.

Abstract: Using a logistic regression classification model executing on a processor, a topic is classified into an interaction type in a set of predefined interaction types. A set of documents corresponding to the topic is extracted from a document repository. Using a generative adversarial model executing on a processor, a sentiment corresponding to a reaction to a previous presentation is scored, the scoring resulting in a scored sentiment. Using a trained attention layer model, the interaction type, the set of documents, and the scored sentiment are weighted, the weighting generating a weighted interaction type, a weighted set of documents, and a weighted scored sentiment. Using a natural language generation transformer model executing on the processor, a document in the weighted set of documents is weighted according to the weighted interaction type and the weighted scored sentiment.