Abstract: Techniques for machine learning analysis are provided. A machine learning (ML) model is trained to identify appropriate documents based on lexical knowledge of target groups. A lexical knowledge of a set of users is determined. Additionally, a first document of a plurality of documents is selected by processing the determined level of lexical knowledge using the ML model. The first document is presented to the set of users. A level of engagement of the set of users is then determined. Upon determining that the level of engagement is below a predefined threshold, a second document of the plurality of documents is selected using the ML model.

Abstract: Modifying a recipe for food based on a geographic origin of produce in the recipe is disclosed. A recipe is identified by a data processing system for a food dish, produce procured for the recipe from source(s), and determines a geographic origin where the produce was grown and an estimated transport time of the produce from the geographic origin to the source(s). A data processing system analyzes the recipe against ingredient(s) available to the user and aspect(s) of the produce, resulting in analysis result(s), modifies the recipe, resulting in a modified recipe based on the geographic origin, the transport time (may be estimated) and the analysis result(s), and makes available to the user the modified recipe.

Abstract: Embodiments of the present invention describe creating a monitoring package by integrating a personal information management system (PIMS) and a social network profile for a user. Embodiments comprise continuously monitoring e-mail messages, calendar entries, social media posts, and social media comments of the integrated PIMS and social network profile, collecting participant preference information and meeting data from the monitoring package, identifying an upcoming meeting for the user according to the meeting data, and determining a list of participants for the upcoming meeting based on conversation threads associated with the upcoming meeting collected from the monitoring package.

Abstract: A question is analyzed to determine a set of categories, a category corresponding to a type of possible answers responsive to the question. A set of opinions is extracted from social media data. Each opinion is from a corresponding responder on a social media platform to which the question is sent. An opinion from the set of opinions is categorized into a category from the set of categories. A strength of the opinion is computed using a subset of a set of strength parameters. The opinion is ranked according to the strength of the opinion. The opinion, the strength, and the category are presented in a filterable presentation.

Abstract: Methods, systems, and computer program products for electronic communication operations are provide. Aspects include receiving communication data associated with an electronic communication, wherein the communication data includes textual data, analyzing attachment data associated with an attachment to the electronic communication, comparing the communication data to the attachment data to determine a relevancy score, and based on the relevancy score, enacting an action for the electronic communication.

Abstract: A method, computer program product, and system includes a processor(s) obtaining content displayed to the user on a computing device communicatively coupled to an input device(s) and/or output device(s). The processor(s) analyzes the content, to identify portions that represent types of content and tags the identified portions with one or more generic tags. The processor(s) generates a sentiment profile of the user by identifying interests of the user and identifying subjects of past electronic communications by the user. The processor(s) maps sentiments in the sentiment profile, to topics, based on the interests and the subjects identified. The processor(s) identifies an identified portion of the content relevant to a topic(s) mapped to a sentiment. The processor(s) generates and implements a physical change to the input device(s) or output device(s) to elicit a sentiment from the user, where the sentiment is mapped to the topic(s).

Abstract: A working animal reaction handling method monitors reactions of a working animal in an environment to conditions of that environment. The method uses an imaging device to obtain images of the environment, the images providing at least a portion of environmental data about the environment. The method ascertains, based at least in part on the environmental data, a cause of a reaction, of the monitored reactions, of the working animal, electronically signals a handler of the working animal with an alert about the ascertained cause of the reaction of the working animal.

Abstract: A method can include obtaining vehicle data corresponding to an autonomous vehicle. The autonomous vehicle can include a vehicle controller configured to control a set of driving maneuvers of the autonomous vehicle. The method can further include obtaining accident condition data based, at least in part, on the vehicle data. The accident condition data can include a first set of driving conditions under which a vehicle accident occurred. The method can further include generating an accident condition record based, at least in part, on the accident condition data. The method can further include determining that the vehicle controller is not trained to control the set of driving maneuvers of the autonomous vehicle under the first set of driving conditions. The method can further include generating, in response to the determining, a pending test case based, at least in part, on the accident condition record.

Abstract: Techniques for machine learning analysis are provided. A machine learning (ML) model is trained to identify appropriate documents based on lexical knowledge of target groups. A lexical knowledge of a set of users is determined. Additionally, a first document of a plurality of documents is selected by processing the determined level of lexical knowledge using the ML model. The first document is presented to the set of users. A level of engagement of the set of users is then determined. Upon determining that the level of engagement is below a predefined threshold, a second document of the plurality of documents is selected using the ML model.

Abstract: Embodiments of the present invention describe creating a monitoring package by integrating a personal information management system (PIMS) and a social network profile for a user. Embodiments comprise continuously monitoring e-mail messages, calendar entries, social media posts, and social media comments of the integrated PIMS and social network profile, collecting participant preference information and meeting data from the monitoring package, identifying an upcoming meeting for the user according to the meeting data, and determining a list of participants for the upcoming meeting based on conversation threads associated with the upcoming meeting collected from the monitoring package.

Abstract: A method, computer program product, and system for enhancing dynamically enhancing query results. A processor(s) monitors computing activities performed by a user, via a client. The processor(s) analyzes the computing activities to identify data comprising a user profile: elements relevant to the user and relationships. The processor(s) generates a knowledge graph with the data. The processor(s) obtains input from the user. The processor(s) executes one or more queries on one or more data sources to retrieve standard search results. The processor(s) analyzes the input and the search results to identify segments of the input and the search results relevant to items comprising the data in the graph and defines links between the items, the data, and the segments. The processor(s) generates contextual search results that include the standard search results, a portion of the items, and a portion of the links and transmits the contextual search results, to the client.

Abstract: A question is analyzed to determine a set of categories, a category corresponding to a type of possible answers responsive to the question. A set of opinions is extracted from social media data. Each opinion is from a corresponding responder on a social media platform to which the question is sent. An opinion from the set of opinions is categorized into a category from the set of categories. A strength of the opinion is computed using a subset of a set of strength parameters. The opinion is ranked according to the strength of the opinion. The opinion, the strength, and the category are presented in a filterable presentation.

Abstract: Provided is a method, computer program product, and virtual conferencing system for providing virtual conferencing assistance to a user. A processor may monitor a virtual conferencing session between a plurality of users. The processor may detect a command to interact with a user interface (UI) element has been given. The processor may analyze UI action data in locating the UI element for the plurality of users. The processor may compare the UI action data for the plurality of users to a data threshold. The processor may determine, in response to the data threshold being met, that a user is experiencing difficulty in locating the UI element. The processor may provide, in response to determining that the user is experiencing difficulty, assistance to the user in locating the UI element.

Abstract: Presenting a coloring pattern by a case of a mobile device according to received content includes receiving, by the mobile device, content; determining, based on the content, a coloring pattern; generating, based on the coloring pattern, a heat pattern in an inductive mesh of the mobile device; and presenting, by a thermally activated material included in the case of the mobile device and in thermal contact with the inductive mesh, the coloring pattern.

Abstract: Provided is a method, computer program product, and virtual conferencing system for providing virtual conferencing assistance to a user. A processor may monitor a virtual conferencing session between a plurality of users. The processor may detect a command to interact with a user interface (UI) element has been given. The processor may analyze UI action data in locating the UI element for the plurality of users. The processor may compare the UI action data for the plurality of users to a data threshold. The processor may determine, in response to the data threshold being met, that a user is experiencing difficulty in locating the UI element. The processor may provide, in response to determining that the user is experiencing difficulty, assistance to the user in locating the UI element.

Abstract: In an approach to cognitive message response, a computer receives an electronic message from a message initiator. The computer creates a custom knowledge base associated with a recipient of the electronic message. The computer determines whether the recipient is available to respond to the received electronic message. The computer determines a status of the recipient based on the custom knowledge base and one or more types of unstructured data. The computer determines a status of the message initiator. Based on the received electronic message, the computer determines a content of the message. Based on the status of the recipient, the status of the initiator, and the content of the electronic message, the computer predicts a response of the recipient to the received electronic message. The computer creates a response to the received electronic message based on the predicted response. The computer transmits the response to the message initiator.

Abstract: In an approach to cognitive message response, a computer receives an electronic message from a message initiator. The computer creates a custom knowledge base associated with a recipient of the electronic message. The computer determines whether the recipient is available to respond to the received electronic message. The computer determines a status of the recipient based on the custom knowledge base and one or more types of unstructured data. The computer determines a status of the message initiator. Based on the received electronic message, the computer determines a content of the message. Based on the status of the recipient, the status of the initiator, and the content of the electronic message, the computer predicts a response of the recipient to the received electronic message. The computer creates a response to the received electronic message based on the predicted response. The computer transmits the response to the message initiator.

Abstract: Methods, systems, and computer program products for electronic communication operations are provide. Aspects include receiving communication data associated with an electronic communication, wherein the communication data includes textual data, analyzing attachment data associated with an attachment to the electronic communication, comparing the communication data to the attachment data to determine a relevancy score, and based on the relevancy score, enacting an action for the electronic communication.

Abstract: Presenting a coloring pattern by a case of a mobile device according to received content includes receiving, by the mobile device, content; determining, based on the content, a coloring pattern; generating, based on the coloring pattern, a heat pattern in an inductive mesh of the mobile device; and presenting, by a thermally activated material included in the case of the mobile device and in thermal contact with the inductive mesh, the coloring pattern.

Abstract: Advertisement delivery using audience grouping and image object recognition may include: receiving image data; identifying, based on the image data, a plurality of image objects associated with a plurality of persons; generating, from the plurality of persons and based on the plurality of image objects, a plurality of clusters, each of the plurality of clusters comprising one or more persons of the plurality of persons; determining a classification for a cluster of the plurality of clusters; determining an advertisement associated with the classification; and sending the advertisement to an advertising platform based on a location of the cluster.