Abstract: Recommending machine learning models is provided. The method comprises training machine learning models, wherein each machine learning model is trained with a unique respective dataset. Metadata associated with each machine learning model is extracted, wherein the metadata includes properties of the respective dataset used to train the machine learning model. The machine learning models and metadata are stored in a model catalog. Upon receiving a new dataset, similarity scores are calculated between the new dataset and the machine learning models in the model catalog according to the properties of the datasets in the metadata of the machine learning models. A closest match machine learning model is identified from the model catalog for the new dataset according to similarity score. Responsive to a determination that the closest match machine learning model exceeds a similarity threshold, predictions for the new dataset are generated with the closest match machine learning model.

Abstract: A system and method are used to record and present a training video for a software component. The method displays the video recording. The video recording comprises: a display element of the software component, and first recorder activity data (FRAD) of recorder user interface (UI) input activity data (RUIIAD) at a FRAD time with video frame data of the video recording. The method further comprises receiving, during the displaying, viewer UI actions from a UI of the viewing device and converting them to viewer UI input activity data based on viewer interactions with the display element of the software component. Then, first viewer activity data (FVAD) is determined at a FVAD time of the video frame that corresponds to the FRAD, where the FRAD time and the FVAD time differ. Responsive to the determining, the method moves the video recording from the FVAD time to the FRAD time.

Abstract: Assigning wireless resources to a device based on usage context and predicting future demand for wireless resources. Connections to available wireless resources are made according to usage context and managed according to predicted future resource needs. An artificial intelligence prediction module and a supervised learning engine establish a proposed wireless resource usage plan. The available wireless resources are evaluated with reference to the usage plan and connections are made according to the evaluations.

Abstract: A method, computer program, and computer system are provided for data facet generation. Data associated with a dataset is received. The received data includes one or more data entries having one or more elements. The one or more elements are associated with one or more data types. One or more data facets are generated for each of the data entries with the received data based on the associated data type. One or more transformations are generated for the data facet corresponding to a machine learning task associated with the dataset. A recommendation is provided to a user based on the generated transformation. The provided recommendation includes generated computer code corresponding to an optimal transformation associated with the machine learning task.

Abstract: A method for providing personalized reading assistance using visual modifications. The method includes forecasting a complexity of a text corpus for a user. The text corpus includes a plurality of words. Visual modifications may be provided to at least some words of the plurality of words in the text corpus based on the forecast complexity of the text corpus for the user.

Abstract: In an approach to improve knowledge content query generation, embodiments identify one or more drift points based on implicit user feedback-based assessment and a user experience to generate one or more knowledge content queries of different complexity from a multi-media file. Further, embodiments adjust a threshold for the one or more drift points to generate different variations of the one or more knowledge content queries and perform iterative refinement on the one or more knowledge queries based on one or more previous iteration evaluations. Additionally, embodiments generate the one or more knowledge content queries in different modalities based on the multi-media file and the threshold for the one or more drift points.

Abstract: A system may include a memory and a processor in communication with the memory. The processor may be configured to perform operations. The operations may include extracting a first multiple dimensional dataset and a second multiple dimensional dataset and encoding the first multiple dimensional dataset into a first encoded dataset and the second multiple dimensional dataset into a second encoded dataset. The operations may include combining the first encoded dataset and the second encoded dataset into a combined encoded dataset and conveying the combined encoded dataset to a user.

Abstract: Generating a composite screenshot including contextually relevant screenshots and displayed information. The composite screenshot includes one or more screenshots captured from a display screen of a requesting user. Contextual relevance is determined according to collected user communications data such as IoT interactions, social media posts, and chat session transcripts.

Abstract: A method, system, and computer program product for simulating potential consequences and possible solutions due to a release of stored energy using augmented reality. The method may include aggregating IoT feeds from devices in a surrounding area. The method may also include calculating amounts of stored energy in the surrounding area based on the IoT feeds. The method may also include predicting contextual situations that could result due to a release of the stored energy in the surrounding area. The method may also include determining one or more consequences for each of the contextual situations. The method may also include calculating a degree of severity of the one or more consequences for each contextual situation. The method may also include determining one or more proposed solutions based on the degree of severity. The method may also include transmitting a recommendation of at least one proposed solution for implementation.

Abstract: A system for manipulating a 3D object for a computing device is provided. A computer device identifies one or more sensors worn by a user, wherein the one or more sensors track the user's hand and finger movements. The computing device monitors during production of a three dimensional (3D) object, the one or more sensors for modification to a holographic projection representing the 3D object, wherein the user's hand and finger movements modify the holographic projection representing the 3D object. A computing device identifies a modification to the 3D object represented by the holographic projection. A computing device generates a modification request based, at least in part, on the identified modification. The computing device updates production of the 3D object, based, at least in part, on the modification request.

Abstract: Mechanisms are provided for automatically generating a user interactable what-is-new feature for an updated application. The mechanisms obtain footprint data from electronic documents generating during a software development lifecycle to generate an update to an application, where the footprint data specifies development requirements and actions performed by at least application developers during the software development lifecycle. The mechanisms further obtain test case generation data from a recording of test case generation interactions with a graphical user interface and execute computer natural language processing on the footprint data and analysis of the recording of the test case generation interactions to automatically generate a what-is-new container for the update to the application.

Abstract: In an approach to governance of document printing, one or more computer processors receive a request from a user to print a document. One or more computer processors identify a content and a context of the document. Based on the content and the context of the document, one or more computer processors determine whether the request complies with one or more printing policies. Responsive to determining the request complies with the one or more printing policies, one or more computer processors determine one or more format attributes of the document. Based on the one or more format attributes, one or more computer processors determine whether the document format is optimized, where optimization minimizes use of one or more printing supplies. Responsive to determining the document format is not optimized, one or more computer processors provide a recommendation for optimizing the document format to the user.

Abstract: An indication is received. The indication is for a wireless communication being made by a computing device. Responsive to receiving the indication, an optimized location to make the wireless communication is determined based on a location of the computing device and an optimized locations list. The optimized location is communicated to the computing device making the wireless communication.

Abstract: A set of one or more media items is identified by a first computer system configured to host media items for various users. The set of media items has a first relationship. A content analysis is performed on the set of one or more media items. The content analysis is based on a first machine-learning model. A first content pattern contained within the set of media items is determined based on the content analysis. A first set of one or more altered media items is generated in response to the first content pattern.

Abstract: A computer system implemented method for a user includes gathering, by a computer system, records of sources that each have no website and that the computer system, therefore, deems to be nondigital sources. The gathering includes the computer system receiving identifications of the nondigital sources based on transactions of other users with the respective sources. The computer system selects candidate sources from among the nondigital sources based on a geolocation of the first user and geolocations of the sources. The computer system selects a final source from among the candidate sources, wherein the computer system selecting the final source includes an artificial intelligence model fitting attributes of the candidate sources to attributes valued by the first user. The computer system presents the final source to the first user as a recommendation by the computer system.

Abstract: A computer system performs corrective operations during a three-dimensional (3D) printing job. The 3D printing job is initiated to produce a physical object. Printing of the physical object is monitored to identify a defect of the physical object during the printing. The defect is analyzed to determine one or more corrective operations to apply to the physical object. The one or more corrective operations are performed during the printing. Embodiments of the present invention further include a method and program product for performing corrective operations during a 3D printing job model in substantially the same manner described above.

Abstract: The present specification describes a computer-implemented method. According to the method, an electronic file to be stored is received and a first context determined for the electronic file. The electronic file is split into multiple segments such that the first context of the electronic file is not discernible from any segment. The multiple segments are then distributed amongst multiple remote storage devices.

Abstract: According to one embodiment, a method, computer system, and computer program product for cloud service brokerage. The embodiment may include receiving a data set and user defined contextual parameters relating to a machine learning (ML) problem of a user to be performed on the data set. The embodiment may include identifying a resource requirement of the ML problem and available resources. The embodiment may include enabling user configuration of the contextual parameters in an interactive augmented reality (AR) view. The embodiment may include identifying a set of clusters upon which to execute computing tasks of the ML problem. The set of clusters is identified out of the available resources. The embodiment may include implementing a ML evaluation process to determine an optimized load distribution model for execution of the computing tasks within the set of clusters. The embodiment may include implementing the optimized load distribution model.

Abstract: Aspects of the present invention disclose a method for safeguarding an audio device by enabling the audio device to perform harm prevention and retrieval related task based on detecting a set of conditions within an operating environment of the audio device. The method includes one or more processors determining a freefall event of an audio device is occurring. The method further includes one or more processors determining an operating environment type of a surface the audio device contacts after the freefall event ends, wherein the operating environment type of the surface indicates liquid is present. The method further includes responsive to the indication that water is present one or more processors taking a liquid mitigation action.

Abstract: A method can include obtaining personal data of a user and generating, based at least in part on the personal data of the user, a taste profile of the user. The taste profile can include a set of food characteristics that corresponds to one or more food preferences of the user. The method can include obtaining contextual data that corresponds to a location of the user. The method can include generating, based at least in part on the taste profile and the contextual data, a food recommendation. The food recommendation can include a predicted food preference of the user. The method can include transmitting the food recommendation to the user.