Abstract: A digital model is generated of a debris location where the digital model includes an augmented reality overlay of the debris location. Digital data from sensors at a debris field is received at a computer, and the digital data depicts debris in the debris field. Materials in the debris field are detected and identified based on the detected materials in the debris field. Layers of materials in the debris field are identified. A digital model is generated of the materials in the debris field including in the layers. Using the computer and an augmented reality (AR) device, an augmented reality overlay is generated based on the digital model of the materials in the debris fields. The augmented reality overlay is superimposed over the digital model of the materials in the debris field to indicate the materials in the debris field and the materials in the layers of the debris field.

Abstract: Systems and methods synchronize content of a virtual environment with a state of a physical environment. In aspects, a method includes obtaining sensor data from a network of remote sensors measuring a physical state of a location at a time; generating context specific parameter data based on the sensor data; obtaining context data from a remote virtual reality (VR) system, wherein the context data reflects a current state of virtual content in a virtual environment displayed by the remote VR system; selecting virtual content to be displayed in the virtual environment by the remote VR system based on the context specific parameter data, the context data, and rules; and sending the virtual content to the remote VR system to be displayed to a user, wherein the virtual content reflects a state of the physical location at the time.

Abstract: A computer-implemented method for identifying, by a processor set, a cluster of users by learning from their interactions in a first virtual environment. The processor set may further classify characteristics of an avatar associated with each user of the cluster of users in the first virtual environment and generate a secure portal visible only to users of the cluster of users to teleport to a second virtual environment. The processor set may also generate a guest profile and a guest avatar in the second virtual environment for each user of the cluster of users. The processor may also adapt the guest avatar for each user of the cluster of users based on a theme of the second virtual environment and generate a new profile prefilled with data collected from previous virtual environments that contain a subset of profile data required for the second virtual environment.

Abstract: A system for collecting and analyzing data to determine a contextual environment of the system, where the contextual environment of the system includes existing software and hardware resources available to the system, employing a matching algorithm to identify additional available software and hardware resources that complement the existing software and hardware resources available to the system, determining a task to be performed on the system, and generating a prioritized list of the additional available software and hardware that complement the system's existing software and hardware resources, the list being ordered based on a degree of relevance with respect to the task to be performed by the system.

Abstract: In an approach to coordinated emotion representation and expression in multi-agent digital assistants, one or more computer processors monitor a conversation of a first user. One or more computer processors determine an intervention opportunity is present in the conversation. One or more computer processors retrieve a profile associated with the first user. One or more computer processors determine a context of the conversation. Based on the profile associated with the first user and the context of the conversation, one or more computer processors generate one or more emotion-based responses. One or more computer processors determine a most appropriate response of the one or more emotion-based responses. One or more computer processors present the most appropriate response to the first user.

Abstract: The exemplary embodiments disclose a method, a computer program product, and a computer system for determining the context of media. The exemplary embodiments may include a user consuming media, collecting data of the media, extracting one or more features from the collected data, and determining a media context based on the extracted one or more features and one or more models.

Abstract: In an approach for providing a user with a set of information on a feature of a comparable product and enabling the user to assess the feature by overlaying the user's product with a digital twin of the comparable product, a processor identifies a first product in a surrounding of the user by comparing an image of the first product to one or more images stored in a product corpus. A processor searches, based on the first product, the product corpus for one or more products within a family of the first product. A processor selects a second product from the product corpus based in part on a set of product comparison preferences. A processor overlays the first product with a digital twin of the second product. A processor simulates a feature associated with the second product that is different from a feature associated with the first product.

Abstract: In an approach to improve power generation in both manual and autonomous vehicles embodiments of the present invention measure, by a set of sensors, a wind speed as it passes through a funnel system. Further, embodiments modulate, by the funnel system, the wind speed of wind that passes through the funnel system to a turbine array, where modulating the wind speed comprises adjusting, by a motor set, a cross-sectional area of an entry portion and an exit portion of the funnel system based on the measured wind speed. Additionally, embodiments proactively adjust, by the computing system, the funnel system to maintain a predetermined wind speed that is being fed to the turbine array through the funnel system by employing machine learning models and control algorithms, using sensor data, vehicle to everything communication, and predictive analysis of wind patterns and road conditions.

Abstract: In an approach for providing contextual conversational user assistance, a processor monitors a plurality of inputs by a user to a computing device. Responsive to determining, based on the monitoring, that an error rate associated with the plurality of inputs has exceeded a threshold level, a processor invokes a conversational large learning model (LLM). A processor queries the user regarding a mental state and a task of the user using the conversational LLM. A processor identifies one or more methods to assist the user based on a response of the user to the querying. A processor executes the one or more methods identified to assist the user.

Abstract: Described are techniques for multi-factor authentication. The techniques include defining a future date and time to implement an automated executable on a server using an agentless Information Technology (IT) automation tool. The techniques further include generating a Future Time-based One-Time Passwords (FTOTP) for the automated executable based on the future date and time. The techniques further include providing the FTOTP to the agentless IT automation tool as a second factor to establish communication with the server for purposes of implementing the automated executable.

Abstract: Text is received from a user describing item(s) for migration to a computing environment with cloud feature(s), resulting in item description(s), the text including unstructured text that are processed separately. Text mining is performed on the unstructured text to extract item feature(s). For each listing a portion of the unstructured text is extracted, resulting in an extracted text portion for each listing from which an entity is identified. Each entity or item feature is mapped to cloud feature(s) available from solution(s) with cloud feature(s). Based on the cloud feature(s), recommendation(s) are made to the user regarding cloud feature(s) of the solution(s) for optional consideration by the user. Explanation(s) for the recommended cloud feature(s) from explainability model(s) may be provided to the user.

Abstract: A context aware code snippet recommendation method, system, and computer program product that includes generating comments for source code, learning semantically meaningful sentence embeddings based on the comments, and retrieving code snippets from the source code based on the learned semantically meaningful sentence embeddings.

Abstract: Embodiments of the invention are directed to a computer-implemented method that includes receiving, using a processor, data representing a 3D physical object by a processing device, and controlling, using the processor, a 3D printing device to apply successive layers of a material by a printing head including a nozzle. The controlling includes depositing at least one layer by instructing a snake robot of a configurable track system to form a shape and define a movement track that corresponds to a movement profile aligned with the 3D model, moving the printing head along the movement track defined by the snake robot, and depositing the at least one layer via the nozzle as the printing head is moved along the movement track.

Abstract: Provided is a method, system, and computer program product for adaptatively training a large language model based on predicted changes in computing resources. A processor may identify data elements of computing resources for training a large language model. The identified data elements include a configured model topology. The processor may generate a forecast vector capturing a predicted change in the configured model topology over a time period. The processor may execute, using an optimization algorithm and the forecast vector, a series of optimization experiments using each training type of a plurality of training types to determine an optimal computing resource supply and demand over the time period. The processor may determine, based on the series of optimization experiments, an accuracy penalty value associated with each training type. The processor may train the large language model using a first training type that has a lowest accuracy penalty.

Abstract: A method for automatically selecting a payment method based on context is disclosed. In one embodiment, such a method includes establishing a user profile for a user. The user profile associates different payment methods with different categories of items. The method detects multiple items in a shopping cart of the user and automatically correlates the items in the shopping cart with the different categories in the user profile. In certain embodiments, automatically correlating includes using machine learning to automatically correlate the items in the shopping car with the different categories in the user profile. The method automatically assigns the different payment methods to the items in accordance with the categories to which the items have been correlated. A corresponding system and computer program product are also disclosed.

Abstract: A computer-implemented method may include: identifying a replacement component for a device component; performing, by a regression model, predictive analysis on user input data to identify a root cause that the device component would need to be replaced; generating, by large language model processing, one or more replacement subsets or one or more replacement supersets, wherein the one or more replacement subsets or one or more replacement supersets each correspond to the replacement component; determining device maintenance metrics over time corresponding to the replacement component and the one or more replacements subsets and the replacement supersets; and in response to determining device maintenance metrics over time, communicating a ranked list of product alternatives.

Abstract: A process is provided for usage-pattern-based generation of a digital model of a product. The process includes obtaining a usage pattern of a user of a product, where the usage pattern includes a set of usage pattern parameters. In addition, the process includes using a machine learning model of an artificial intelligence system to generate a 3-D virtual image of the product aligned with the usage pattern, and analyzing, by the artificial intelligence system, the 3-D virtual image of the product to facilitate identifying a requirement specification for the product based on the usage pattern. Further, the process includes generating a digital model of a new product for the user based on the identified requirement specification.

Abstract: According to one embodiment, a method, computer system, and computer program product for selectively recording meetings is provided. The embodiment may include initiating a digital meeting. The embodiment may also include determining two or more rule sets, including one or more local rule sets, for selectively recording the digital meeting, wherein each rule set includes one or more rules. The embodiment may further include recording one or more recordings of the digital meeting according to each local rule set.

Abstract: Embodiments receive an opt-in to give consent to capture information which determines whether a plurality of secondary users encounters an error, determine that a primary user, different from the secondary users, encounters the error, engage the primary user with the secondary users through a user interface (UI) to ask whether the secondary users have also encountered the error that the primary user encountered, generate a robot process automation (RPA) bot which compiles and runs a plurality of tests for executing an automated workflow verification process for each of the secondary users, perform global tracking for similar automated workflow verification processes as the RPA bot in multiple environments, and train an artificial intelligence (AI) model based on success metrics of the RPA bot and the similar automated workflow verification processes in multiple environments.

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.