Abstract: According to a computer-implemented method, an available amount of each of multiple computing resources is determined by machine logic over a period of time at a computing device. The machine logic also determines an expected usage of each computing resource to execute each workflow in a queue. The machine logic also determines a time of execution of each workflow in the queue based on the available amount of each of the multiple computing resources over time and the expected usage of each computing resource to execute each workflow in the queue.

Abstract: Mechanisms are provided for rendering content in a compacted view. A machine learning computer model is trained by a machine learning process to predict a user attention score for segments of content based on features of the content and historical user attention data. The trained machine learning computer model processes new content to associate with each segment, in a plurality of segments, of the new content, a corresponding user attention score. The segments, in the plurality of segments, of the new content are ranked relative to one another based on the corresponding user attention scores of the segments. A compacted view of the new content is rendered based on the ranking of the segments. A first number of segments are rendered in the compacted view and a second number of segments are not rendered in the compacted view, and are replaced with an inserted user selectable expansion element.

Abstract: Targeted acquisition of data for model training includes identifying attributes of classified samples of a collection of samples classified by a classification model, and generating at least one query based on the identified attributes, the at least one query 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 database of deployed configurations, as well as attempted configurations that failed is maintained and used as reference to compare against configurations of attempted software deployments. Upon detecting a failed deployment, disclosed embodiments search the database for working configurations that most closely resemble the failed configuration, and rank the configurations based on various criteria. Disclosed embodiments may then automatically select a highest ranked working configuration, and perform an automatic upgrade of the necessary components to create a working configuration.

Abstract: A computer-implemented method, system and computer program product for optimizing a flow of data within transform, load (ETL) data processing pipelines. Database columns from a source database that are to be transformed as well as not transformed in the data processing stages of a processing segment of an ETL data processing pipeline are identified. Those database columns that are to be transformed are grouped into a processing schema. Transformations are then performed on the database columns of the processing schema. Those database columns that are not to be transformed are grouped into a non-processing schema. A large object data type (LOB) is then created to reference the non-processing schema. An identifier is created and inserted in the data processing stages to identify the LOB in replace of the database columns that are not to be transformed thereby avoiding the copying of the database columns that are not to be transformed.

Abstract: Targeted acquisition of data for model training includes identifying attributes of classified samples of a collection of samples classified by a classification model, and generating at least one query based on the identified attributes, the at least one query 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: An embodiment for monitoring machine learning models to detect and rectify model drift using governance. The embodiment may receive a plurality of machine learning models and register the plurality of machine learning models to a governance dashboard. The embodiment may automatically monitor the received plurality of machine learning models to identify factors used by each of the received plurality of machine learning models and generate corresponding clusters of similar machine learning models. The embodiment may automatically detect an incorrect decision made by a target machine learning model and then automatically calculate a correlation score between the target machine learning model and machine learning models within an associated corresponding cluster of similar machine learning models. The embodiment may, in response to detecting a correlation score above a threshold, automatically determine and output a cluster reinforcement recommendation.

Abstract: A system and method are disclosed that utilizes an artificial intelligence based virtual proxy node. The virtual proxy node includes an intent resolution model and communicates between a smart audio device and at least one secondary device, wherein the at least one secondary device is configured to be controlled by a smart audio device or smart hub. The virtual proxy node tracks interactions between the smart audio device and the at least one secondary device to derive historical and context data from the tracking interactions. The virtual proxy node uses the historical and context data to predict which secondary device will be successful in responding to the user input command and broadcasts the input command to the virtual proxy node associated with one of the at least one secondary device. The virtual proxy node includes an intent resolution model trained by historical and context data.

Abstract: An embodiment for managing data using machine learning models and information governance. The embodiment may automatically detect a data analysis request made within a system and identify subject datasets. The embodiment may automatically conduct shallow term assignments on each row and column of data in the subject datasets and automatically match the shallow term assignments for each row and column with a stored set of ranked terms, and automatically flag rows or columns matching with ranked terms above a predetermined threshold ranking for further analysis. The embodiment may automatically and continuously monitor and detect irrelevant metadata types to prevent subsequent analysis and storage of data including the irrelevant metadata types. The embodiment may automatically generate a criticality ranking for stored analysis datasets.

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 method includes determining a presence of one or more people around a user and determining, based on the presence of the one or more people, an exposure level. The method also includes receiving a message for the user, the message comprising a first portion and a second portion and removing, based on the exposure level, the first and second portions from the message to produce a first message. The method further includes removing, based on the exposure level, the first portion from the message to produce a second message and presenting, based on the exposure level, the first message to the user. The method also includes receiving, from the user, feedback about the first message and presenting the second message to the user in response to the feedback.

Abstract: Embodiments of the present invention provide computer-implemented methods, computer program products and computer systems. Embodiments of the present invention can dynamically predict a user need based, at least in part, on context associated with an activity. Embodiments of the present invention can then execute a remedial action that satisfies the predicted user need.

Abstract: An approach for a computer application prediction program on a computing device to receive a user input to the computer user interface. The approach includes the computer program retrieving a status of one or more connected computer devices. The approach includes the computer program performing a contextual analysis of the user input. Furthermore, the approach includes the computer program retrieving, from a database, at least one similar user input and at least one target application for each of the at least one similar user inputs. The approach includes the computer program predicting at least one target application for the user input.

Abstract: A method, computer system, and a computer program product for data pipeline prioritization is provided. Embodiments may include receiving, by a cognitive rules engine, one or more data pipelines. Embodiments may then include analyzing, using a computational method of the cognitive rules engine, the one or more data pipelines. Embodiments may lastly include prioritizing the one or more data pipelines based on a result of the computational method of the cognitive rules engine.

Abstract: A method provides for classifying data fields of a dataset. A classifier configured for determining confidence values for a plurality of data classes for the data fields may be applied. Using the confidence values, data class candidates may be identified. Data fields may be determined for which a plurality of data class candidates is identifiable. Using previous user-selected data class assignments, a probability may be determined for the data class candidates that the respective data class candidate is a data class to which the respective data field is to be assigned. The data fields may be classified using the probabilities to select for the data fields a data class from the data class candidates. The dataset may be provided with metadata identifying for the data fields the data classes to which the respective data fields are assigned.

Abstract: A system and method are disclosed that utilizes an artificial intelligence based virtual proxy node. The virtual proxy node includes an intent resolution model and communicates between a smart audio device and at least one secondary device, wherein the at least one secondary device is configured to be controlled by a smart audio device or smart hub. The virtual proxy node tracks interactions between the smart audio device and the at least one secondary device to derive historical and context data from the tracking interactions. The virtual proxy node uses the historical and context data to predict which secondary device will be successful in responding to the user input command and broadcasts the input command to the virtual proxy node associated with one of the at least one secondary device. The virtual proxy node includes an intent resolution model trained by historical and context data.

Abstract: Data-deduplicating includes comparing a first record of a data-store with a second record of the data-store but instead of using a static weight for a field, the present data-deduplicating dynamically assigns a first weight for the first score to generate a first weighted score, wherein the first weight is based on one or both of the first value or the second value; and assigns a second weight for the second score to generate a second weighted score. A composite score is calculated based on the first weighted score and the second weighted score; and it is determined whether or not the first record and the second record are duplicate records, based on the composite score.

Abstract: An approach is provided for completing a pipeline graph. Using a deep learning based sequence model, an initial data pipeline having a sequence of nodes is generated. Mismatch(es) between data formats required by input and output in the sequence of nodes is identified. Virtual gap node(s) that correct the mismatch(es) are added to the initial data pipeline. For a given virtual gap node, tentative graph structures are determined using knowledge graphs and a crowd sourced validation system. Reuse forecast scores and performance scores for the tentative graph structures are calculated. Based on the reuse forecast scores and the performance scores, a final graph structure for implementing the given virtual gap node is determined.

Abstract: An approach is disclosed that receives a code segment from a source code file. The received code segment is inputted to a trained knowledge manager from which one or more comparable code segments are received. Each of the comparable code segments are identified as having similar functionality as the selected code segment and each of the comparable code segments has a set of metadata. The metadata is used to select one of the comparable code segments and this selected code segment is used to replace the original code segment in the source code file.

Abstract: A computer-implemented method, system and computer program product for optimizing a flow of data within transform, load (ETL) data processing pipelines. Database columns from a source database that are to be transformed as well as not transformed in the data processing stages of a processing segment of an ETL data processing pipeline are identified. Those database columns that are to be transformed are grouped into a processing schema. Transformations are then performed on the database columns of the processing schema. Those database columns that are not to be transformed are grouped into a non-processing schema. A large object data type (LOB) is then created to reference the non-processing schema. An identifier is created and inserted in the data processing stages to identify the LOB in replace of the database columns that are not to be transformed thereby avoiding the copying of the database columns that are not to be transformed.