Abstract: Techniques for using machine learning to model climatic data are disclosed. In one example, a computer implemented method comprises receiving climate data comprising a plurality of spatial components and a plurality of temporal components, and masking a portion of the climate data. A machine learning model is trained, wherein the training is based at least in part on the masked portion of the climate data. A vector representation of the climate data is generated via the machine learning model.

Abstract: Classification and visualization of user interactions with an interactive computing platform is provided by building and providing a graphical user interface (GUI) of graphical elements for display on a display device. The graphical elements present (i) visualizations of user interactions between users and an interactive computing platform in progression of the users through tasks based on the user interactions and (ii) identified features of the user interactions relative to the tasks and progression of the users therethrough. The building and providing includes, for each of the tasks, building and providing a respective task element that reflects identified features of a set of user interactions of one or more users in progressing through that task. Building a task element for a task includes building the task element as a polygon having vertices corresponding to subtasks of the task and edges between the vertices.

Abstract: One embodiment provides a method, including: obtaining information related to farming activities of a farmer; predicting an annotation category for the information, wherein the annotation category identifies a topic of the information; selecting an annotator for annotating the information based upon the annotation category, wherein the selecting comprises utilizing (i) a social proximity constraint identifying a social connection between the farmer and another farmer and (ii) a farm signature constraint identifying a similarity of the farmer to another farmer; assigning the annotator to annotate the obtained information; and receiving annotations for the information.

Abstract: An approach for modifying in real-time by removing or reinforcing stroboscopic effect from images associated with a viewing experience is disclosed. The approach includes identifying video clips, detecting environmental parameters and calculating display setting. The approach also analyzes display setting using recommendation from GAN, output displaying setting on an AR display and receiving feedback from user.

Abstract: Provided are a computer program product, system, and method for generating constraints based on reported crop arrivals to marketplaces and remote sensed data to estimate farm yields of farm fields. A total arrival quantity constraint for a marketplace is generated comprising a function of a reported arrival of crops at a marketplace and an estimated arrival to the marketplace comprising a sum of farm yields for the farm fields, wherein the farm yields comprise variables to optimize. Sensed data of the farm fields is used to determine relationships of the farm yields for the farm fields. Comparative farm yield constraints are generated comprising relationships of the farm yields for different pairs of farms. A constraint optimization problem is solved based on the total arrival quantity constraint and comparative farm yield constraints to calculate the farm yields of the farm fields.

Abstract: Methods, systems, and computer program products for customizing agricultural practices to maximize crop yield are provided herein. A computer-implemented method includes obtaining data pertaining to (i) a geographical area comprising a plurality of regions and (ii) one or more agricultural practices applied to the geographical area; assigning each of the plurality of regions to a respective cluster of a set clusters, based at least in part on comparing features identified in the data, wherein similar ones of said regions are assigned to the same cluster; generating instructions that are specific to a given cluster in the set, wherein the instructions relate to agricultural tasks to be performed on the regions assigned to the given cluster; and triggering, based on said instructions, one or more automated farming processing devices, thereby carrying out at least a portion of said agricultural tasks.

Abstract: A computer implemented method for forecasting sensor states in a sensor network is provided. A number of processor units identifies weights and changes in the weights over time in a temporal graph of the sensor network using a temporal graph neural network trained using the temporal graph. The number of processor units determines patterns of the weights based on the weights and the changes in the weights over time. The number of processor units trains a regression machine learning model using the patterns of the weights to forecast sensor states for sensors in the sensor network.

Abstract: A computer hardware system having an image modification platform includes a hardware processor configured to initiate the following executable operations. A digital image is received by the image modification platform. A plurality of objects within the digital image are detected using an object segmentation engine of the image modification platform. The plurality of objects are classified using an object classification engine of the image modification platform and involve associating a stakeholder, respectively, to at least two of the plurality of objects. A knowledge map for the digital image is generated based upon the classifying and the stakeholders using a knowledge map engine. A set of constraints are identified for each of the plurality of objects based upon the knowledge map and the plurality of stakeholders. A modified digital image to be published is generated by an interactive image generation engine using the sets of constraints.

Abstract: A method, the method comprising retrieving a sound intensity map for a venue, wherein the sound intensity map is divided up into a plurality of regions, wherein the sound intensity map predicts a sound quality for each region during a current event. Receiving data from a plurality of IOT enabled operated aerial vehicles, where each IOT enabled operated aerial vehicle of the plurality of IOT enabled operated aerial vehicles travels around different regions of the plurality of regions, wherein each IOT enabled operated aerial vehicle collects data during the event. Comparing the received data to the sound intensity map to determine the region where an audio component of a venue audio needs to be adjusted. Determining the adjustment required for the audio component and adjusting the audio equipment.

Abstract: Media is streamed in accordance with carbon footprint considerations. A streaming history of a streaming plan is analyzed to determine a historical carbon footprint. At least one streaming plan is presented that includes target carbon footprint relative to the historical carbon footprint. A selection is received for one of the at least one streaming plan, wherein the streaming performance on the streaming plan is tracked for post streaming plan carbon emissions. The user streaming performance is modified for the post streaming plan carbon emissions to substantially match the target carbon footprint of the at least one streaming plan.

Abstract: In an approach to jointly learning uncertainty-aware trend-informed neural network for a demand forecasting model, a machine learning model is trained to capture uncertainty in input forecasts. The uncertainty in a latent space is represented using an auto-encoder based neural architecture. The uncertainty-aware latent space is modeled and optimized to generate an embedding space. A time-series regressor model is learned from the embedding space. A machine learning model is trained for trend-aware demand forecasting based on said time-series regressor model.

Abstract: Smart meal preparation can be performed where a consumer order for a food item and a consumer request related to the food item are received. One or more recipes for the food item are received. The one or more recipes are analyzed to determine a conflict with the consumer request. The consumer request and the determined conflict are transmitted to a wearable device worn by a cook who will prepare the food item. Alerts related to the conflict are provided to the cook via the wearable device during preparation of the food item. Preparation of the food item is monitored for compliance with the consumer request.

Abstract: Targeted acquisition of data for model training includes automatically generating metadata describing samples, of an initial dataset, in neighborhoods of an embedding space in which the samples are embedded. The samples described by the automatically generated metadata are classified by a classification model, and include both correctly classified samples in the neighborhoods and incorrectly classified samples in the neighborhoods. Additionally, attributes of one or more correctly classified samples of the collection of samples and one or more incorrectly classified samples of the collection of samples are identified, and queries are generated based on the identified attributes, the queries tailored, based on the attributes, to retrieve additional training data for training the classification model to more accurately classify samples and avoid incorrect sample classification.

Abstract: A computer-implemented method applies insights from a variety of data sources to each of the data sources. The method includes identifying a set of data sources, wherein each of the data sources are associated with a domain. The method includes analyzing documentation for each of the data sources. The method further includes extracting a set of attributes for each data source, and determining a data schema associated with each data source. The method includes mapping each data schema to a common domain schema. The method also includes linking, based on the mapping and on the set of attributes for each data source, common features across each data source. The method includes generating, in response to the linking, a knowledge graph. The method further includes preparing a visual display for a set of domain insights; and forking the set of domain insights into a first data source.

Abstract: Targeted acquisition of data for model training includes automatically generating metadata describing samples, of an initial dataset, in neighborhoods of an embedding space in which the samples are embedded. The samples described by the automatically generated metadata are classified by a classification model, and include both correctly classified samples in the neighborhoods and incorrectly classified samples in the neighborhoods. Additionally, attributes of one or more correctly classified samples of the collection of samples and one or more incorrectly classified samples of the collection of samples are identified, and queries are generated based on the identified attributes, the queries tailored, based on the attributes, to retrieve additional training data for training the classification model to more accurately classify samples and avoid incorrect sample classification.

Abstract: The present disclosure relates to a method comprising receiving an electronic message of an electronic communication system. An image based noisy content may be generated. The image based noisy content is different from a content of the received electronic message and relates to the message intents of the received message. The electronic communication system may be configured to provide the image based noisy to the recipient instead of the received electronic message or provide the received electronic message to the recipient in addition to the image based noisy.

Abstract: Embodiments of the present invention provide a computer system, a computer program product, and a method that comprises identifying a plurality of code datasets prior to a data migration; analyzing the identified code datasets for a plurality of parameters; dynamically predicting a carbon footprint associated with the analyzed code datasets based on the plurality of parameters for each analyzed code dataset; and automatically optimizing the analyzed code datasets based on the predicted carbon footprint for data migration.

Abstract: A first user access, directed to a first network document, of a first user is detected. One or more first network document concepts in the first network document are detected. A first user profile of the first user is retrieved based on the first user access. A first knowledge gap of the first user is identified based on the first network document concepts and based on the first user profile. One or more additional network documents are obtained based on the one or more first network document concepts. The additional network documents include at least a first concept of the one or more first network document concepts. The first network document is modified to include at least a portion of the first concept from the one or more additional network documents. The modification is based on the first knowledge gap and on the one or more additional network documents.

Abstract: Classification and visualization of user interactions with an interactive computing platform is provided by applying machine learning (ML) model(s) to user transcripts, the ML model(s) trained to classify interactions with an interactive computing platform, the user transcripts including user interactions between users and the interactive computing platform in progression of the users through tasks based on the user interactions, where the applying classifies the user interactions and identifies features of the user interactions, and building and providing a graphical user interface (GUI) of graphical elements for display on a display device, the graphical elements presenting visualizations of the user interactions and the identified features thereof relative to the tasks and progression of the users therethrough, the GUI including, for each of the tasks, a respective task element that reflects identified features of a set of user interactions of user(s) in progressing through that task.

Abstract: An approach is provided in which the approach analyzes a visual scene corresponding to a position of a user within a mixed reality environment that includes a set of visible objects in a physical world. The approach determines a set of spatial relationships and a set of temporal relationships between the set of visible objects, and generates an advertisement based on the determined set of spatial relationships and the set of temporal relationships. The approach overlays the advertisement in the mixed reality environment on at least one of the set of visible objects.