Abstract: A processor set is configured to receive tabular data records and generate a plurality of clusters, associated with specific real-world entities, within the received tabular data records, wherein each cluster is associated with a specific real-world entity. The processor set may further identify informative features within a first cluster and mask a subset of the informative features. Based on the masked subset of informative features and using self-supervision techniques, the processor set may train a tabular foundation model.

Abstract: An embodiment includes computing a computational cost of a job, using a first machine learning algorithm. The job may include an original amount of an egress of data from a cloud computing environment. The embodiment includes determining, using a second machine learning algorithm, the amount of the egress of data corresponding to the job has a computer business criticality that exceeds a threshold level of business criticality. The embodiment includes analyzing a current egress plan used in computing the computational cost of the job. The embodiment includes reconfiguring the current plan to a second egress plan to reduce the computational cost of the job. The embodiment includes implementing a second plan such responsive to execution of the job, the second plan causes data egress behavior to change from the original egress of data behavior. A modified egress behavior causes an effective reduction in the egress cost of the job.

Abstract: A method including employing corporate data to generate an acyclic graph representing a reporting hierarchy, identifying relevant key performance indicators (KPIs) for a plurality of users, training a classifier to predict whether a KPI is relevant to a user of the plurality of users, generating interest-persona maps correlating KPI impact alerts with a subset of the plurality of users based on detected information technology (IT) event streams, business streams, KPI streams, and environmental metrics streams, assigning a score to each KPI impact alert based on compatibility between user features and KPI features, and notifying, by consulting the interest-persona maps, one or more targeted users of the KPI impact alerts relevant to them.

Abstract: Mechanisms are provided for forecasting information technology (IT) and environmental impacts on key performance indicators (KPIs). Machine learning (ML) computer model(s) are trained on historical data representing events and KPIs of organizational processes (OPs). The ML computer model(s) forecast KPI impact given events. Correlation graph data structure(s) are generated that map at least one of events to IT computing resources, or KPI impacts to OPs. A unified model is trained to model OPs and IT resources. The trained ML computer model(s) and unified model process input data to generate a forecast output that specifies at least one of a forecasted IT event or a KPI impact. The forecasted output is correlated with at least one of IT computing resource(s) or OP(s), at least by applying the correlation graph data structure(s) to the forecast output to generate a correlation output.

Abstract: Records can be matched by a graph neural network model performing entity resolution on the records, and representing each record as a respective node in a graph. Record matching explanations can be generated, each record matching explanation indicating a first set of attributes, and a first set of corresponding values, used for the matching at least two of the records. Nodes can be clustered into a plurality of clusters by aggregating the record matching explanations and, based on the record matching explanations, determining which of the records have high importance values, in the first set of values, that match. At least one cluster explanation can be generated, the cluster explanation indicating a second set of attributes, and a second set of values corresponding to the second set of attributes, used for the clustering the nodes. The record matching explanation and the cluster explanation can be output.

Abstract: A computer implemented method classifies records. A number of processor units creates a training dataset comprising subgraphs of matched records matched to an entity and identifying an importance of attributes in the matched records. The matched records in a subgraph are related to each other by a subset of the attributes. The number of processor units trains a graph neural network using the training dataset. The graph neural network classifies the records as belonging to the entity.

Abstract: Mechanisms are provided for forecasting information technology (IT) and environmental impacts on key performance indicators (KPIs). Machine learning (ML) computer model(s) are trained on historical data representing IT events and KPIs of organizational processes (OPs). The ML computer model(s) forecast IT events given KPIs, or KPI impact given IT events. Correlation graph data structure(s) are generated that map at least one of IT events to IT computing resources, or KPI impacts to OPs. The trained ML computer model(s) process input data to generate a forecast output that specifies at least one of a forecasted IT event or a KPI impact. The forecasted output is correlated with at least one of IT computing resource(s) or OP(s), at least by applying the correlation graph data structure(s) to the forecast output to generate a correlation output. A remedial action recommendation is generated based on the forecast output and correlation output.

Abstract: Mechanisms are provided for automated generation of an electronic form for an electronic messaging subsystem. Historical conversation logs are obtained from the electronic messaging subsystem which comprise a plurality of communication sequences. Communication sequences within the historical conversation logs are clustered according to similarity of features. For each cluster the following operations are performed: identifying, within the cluster, sequences of normalized utterances that are repeated across communication sequences of the cluster; categorizing each sequence, in a set of the sequences of normalized utterances, as to whether the sequence can be represented as one or more electronic forms; and extracting, for each communication sequence in the cluster, attributes and corresponding attribute values. One or more electronic form data structures are generated based on the attributes and corresponding attribute values extracted for each communication sequence in the cluster.

Abstract: Predicting the impact of an information technology (IT) failure includes detecting a computer-generated indication of the failure. Responsive to determining that the IT failure is a previously unseen IT failure, operations of a computer-implemented unseen event handler can be invoked. The unseen event handler can map the previously unseen IT failure to a previously seen IT failure based on a similarity score generated by a computer-implemented similarity scorer, wherein the similarity score is based on a unified process-IT topology. A machine learning model can generate an IT failure impact prediction and recommendation based on the mapping, wherein the machine learning model also is based on the unified process-IT topology. An output of the IT failure prediction and recommendation can be generated.

Abstract: Scheduling IT maintenance based on key performance indicators (KPIs) includes receiving current information technology (IT) data corresponding to an IT system and current process data corresponding to a process implemented with the IT system. A suffix and time prediction model generates a suffix prediction of a likely sequence of suffixes given a prefix of the process and a time prediction of time remaining to complete the likely sequence of suffixes, wherein the suffix and time predictions are based on based on the IT data and current process data; A load on each of step of the process is determined based on the suffix and time prediction; One or more KPI metrics of the process is determined based on the load of each step. An IT maintenance schedule output is generated based on the one or more KPI metrics.

Abstract: Records can be matched by a graph neural network model performing entity resolution on the records, and representing each record as a respective node in a graph. Record matching explanations can be generated, each record matching explanation indicating a first set of attributes, and a first set of corresponding values, used for the matching at least two of the records. Nodes can be clustered into a plurality of clusters by aggregating the record matching explanations and, based on the record matching explanations, determining which of the records have high importance values, in the first set of values, that match. At least one cluster explanation can be generated, the cluster explanation indicating a second set of attributes, and a second set of values corresponding to the second set of attributes, used for the clustering the nodes. The record matching explanation and the cluster explanation can be output.

Abstract: Mechanisms are provided for forecasting information technology (IT) and environmental impacts on key performance indicators (KPIs). Machine learning (ML) computer model(s) are trained on historical data representing IT events and KPIs of organizational processes (OPs). The ML computer model(s) forecast IT events given KPIs, or KPI impact given IT events. Correlation graph data structure(s) are generated that map at least one of IT events to IT computing resources, or KPI impacts to OPs. The trained ML computer model(s) process input data to generate a forecast output that specifies at least one of a forecasted IT event or a KPI impact. The forecasted output is correlated with at least one of IT computing resource(s) or OP(s), at least by applying the correlation graph data structure(s) to the forecast output to generate a correlation output. A remedial action recommendation is generated based on the forecast output and correlation output.

Abstract: Mechanisms are provided for dynamic re-resolution of entities in a knowledge graph (KG) based on streaming updates. The KG and corresponding initial clusters associated with first entities are received along with a dynamic data stream having second documents referencing second entities. Clustering on the second documents based on the set of initial clusters, and document features of the second documents, is performed to provide a set of second document clusters. For second document clusters that should be modified based on entities associated with the second document cluster, a cluster modification operation is performed. Updated clusters are generated based on the clustering and modification of clusters. Entity re-resolution is dynamically performed on the entities in the KG based on the second entities associated with the updated clusters to generate an updated knowledge graph data structure.

Abstract: An embodiment establishes a designated attribute value as a semantic criterion for grouping records in a bucket, identifies a first set of records having attribute values that satisfy the semantic criterion, and adds the first set of records to the bucket. The embodiment detects that the first set of records represent a first series of events that occurred in succession at respective times. The embodiment derives a temporal attribute value representative of a time pattern formed by the times of the first series of events and designates the temporal attribute value as a temporal criterion for grouping records in the bucket. The embodiment identifies a second set of records that represent a second series of events and satisfy the temporal criterion and adds the second set of records to the bucket based at least in part on the second set of records satisfying the temporal criterion.

Abstract: An embodiment establishes a designated attribute value as a semantic criterion for grouping records in a bucket, identifies a first set of records having attribute values that satisfy the semantic criterion, and adds the first set of records to the bucket. The embodiment detects that the first set of records represent a first series of events that occurred in succession at respective times. The embodiment derives a temporal attribute value representative of a time pattern formed by the times of the first series of events and designates the temporal attribute value as a temporal criterion for grouping records in the bucket. The embodiment identifies a second set of records that represent a second series of events and satisfy the temporal criterion and adds the second set of records to the bucket based at least in part on the second set of records satisfying the temporal criterion.