Abstract: Revising data management indices by defining a plurality of data entity buckets, assigning bucket hash values to each data record according to the plurality of data entity buckets, generating a record index including data record bucket membership data, determining a combined record score for each data record according to the bucket membership data, revising data record relationship indices of a data management system according to the combined record score of a data record, and responding to a user query according to the record index including the revised record relationships.

Abstract: Described are techniques for disintegrating an entity into smaller entities. A graph (“first graph”) for the entity of records to be disintegrated is constructed, where each vertex of the first graph represents a record in the entity of records to be disintegrated. The edges in the first graph connecting records in the entity of records represent matching links between the records, where each edge is associated with a weight corresponding to a similarity score. Furthermore, two or more additional graphs representing two or more sub-entities of the entity of records to be disintegrated are constructed. Such graphs are constructed based on selecting edges with a maximum weight out of the edges connected between each pair of records in the first graph or based on the number of connections each record has with other records in the first graph exceeding a threshold value.

Abstract: A computer-implemented method for improving entity matching in a probabilistic matching engine can train a graph neural network (GNN) model on an output of a probabilistic matching engine to perform entity matching and determine counterfactual explanations for non-matches of entities. A list of data transformations can be identified by actionable recourse using the GNN model. The list of data transformations can be ranked, using the GNN model, based on computational overhead and an estimated improvement in entity matching within the probabilistic matching engine.

Abstract: Described are techniques for disintegrating an entity into smaller entities. A graph (“first graph”) for the entity of records to be disintegrated is constructed, where each vertex of the first graph represents a record in the entity of records to be disintegrated. The edges in the first graph connecting records in the entity of records represent matching links between the records, where each edge is associated with a weight corresponding to a similarity score. Furthermore, two or more additional graphs representing two or more sub-entities of the entity of records to be disintegrated are constructed. Such graphs are constructed based on selecting edges with a maximum weight out of the edges connected between each pair of records in the first graph or based on the number of connections each record has with other records in the first graph exceeding a threshold value.

Abstract: One or more trained embedding generation artificial intelligence models are executed to generate a plurality of record attribute embeddings. The plurality of record attribute embeddings represents a plurality of attributes of data of a plurality of records. Grouping of the plurality of record attribute embeddings is performed. The grouping of a record attribute embedding includes grouping attribute values of the record attribute embedding into one or more groups of attribute values. The performing grouping provides a plurality of groups of attribute values for the plurality of record attribute embeddings. Selected records are compared to provide a set of matched records. The comparing, based on a group of attribute values, includes comparing records that include one or more attribute values grouped in the group of attribute values providing a subset of matched records of the set of matched records. The set of matched records is stored in an accessible computer location.

Abstract: A system for searching and/or linking corresponding electronic data records includes: choosing, for a target electronic data record, a group of candidate electronic data record(s) for comparison; selecting a first set of one or more attributes for comparison between the target record and the group of candidate record(s), and choosing a threshold for the comparisons. The system further includes comparing attribute(s) of the target electronic data record to attribute(s) of the group of candidate records; and scoring the comparison of the attribute(s) of the target electronic data record to the attribute(s) of the candidate records. The system further can include configuring a sieve to apply to electronic records, wherein the sieve has a second set of attribute(s) and a corresponding threshold for each of the second set of attribute(s), and applying the sieve to one or more electronic data records to identify records.

Abstract: An approach for accessing multi-attribute data records of a master data management system. The method comprises: enhancing the master data management system with one or more search engines for enabling data record access. A request of data may be received at the master data management system. A set of one or more of the multiple attributes, referenced in the received request, may be identified. A combination of one or more of the search engines of the master data management system, whose performances for searching values of at least part of the set of attributes fulfil a current selection rule may be selected. And, the request may be processed using the combination of search engines. At least part of the results of the processing may be provided, and the selection rule may be updated based on user operations on the provided results, the updated selection rule becoming the current selection rule.

Abstract: A network system to provide real-time integration and processing of user data with infrastructure data to generate solutions to user pain points. Real-time user data, including feedback and interactions, is generally not uniform and overwhelmingly large. The system provides solutions to user pain-points at scale, which, in some instances, may be unknown to the service provider. The system does so by contextually linking user data and categorizing it into standardized taxonomies. The infrastructure data is then analyzed against the taxonomies by the system's AI/ML network. The system then provides one or more pain point identifications and solutions. The system may also provide an interface to visualize the taxonomies, pain points, and trend analysis of the pain points.

Abstract: A method, system, and computer program product are configured to create a tuned data record matching model by adjusting values of one or more parameters in a data record matching model based on a second training data set labeled at a data record level, wherein the data record matching model is initially trained using a first training data set labeled at an attribute level.

Abstract: A method, a structure, and a computer system for mapping data fields. The exemplary embodiments may include, based on determining that a first data set and a second data set contain homogenous data, mapping at least one column of the first data set to at least one column of the second data set based on comparing at least one of relative column position and unique value sets. Based on determining that the first data set and the second data set contain heterogeneous data, the exemplary embodiments may include mapping the at least one column of the first data set to the at least one column of the second data set based on a difference between distribution signatures of unique value sets within each of the first data set and the second data set being less than a threshold.

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: A computer-implemented method increases the efficiency of matching records from two sources. The method includes identifying a first source and a second source wherein each of the sources include one or more records and each record includes one or more attributes. The method further includes determining, based on a corpus, the one or more attributes and generating, based on the attributes, a set of feature vectors which vectors represent the one or more attributes. The method includes comparing each record in the first source against each record in the second source. The method further includes generating, in response to the comparing, a link confidence. The method also includes linking, in response to the link confidence being above a linking threshold, the associated records. The method includes determining a first feature vector of the set of feature vectors used in the linking, and outputting a set of results.

Abstract: A method, apparatus, computer system, and computer program product for managing information. A set of bucket hashes and comparison information for a data record are identified by a computer system. The set of bucket hashes is generated from the comparison information, wherein the set of bucket hashes and the comparison information form a metadata record. A number of candidate metadata records in a metadata database is identified by the computer system using the set of bucket hashes, wherein the number of candidate metadata records comprises a set of candidate bucket hashes and candidate comparison information. An entity membership is identified by the computer system for the data record from a comparison of the comparison information in the metadata record with the candidate comparison information in the number of candidate metadata records.

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: Computer system for performing biometric matching in a way that balances accuracy level required in the biometric matching against computing resources (for example, processor cycles) that will be needed to match authentication requesters with profiles of authorized users. In some embodiments, this is achieved by controlling the number of clusters and/or the number of clusters to be searched pursuant to an authentication request.

Abstract: Managing record hierarchies and record groups in a data management system is provided. A root record node is identified for a record hierarchy. A probabilistic search of a graph of the record hierarchy is performed to identify record nodes related to the root record node based on record relationships data. Identified record nodes related to the root record node are positioned as a level under the root record node in the record hierarchy. Any record nodes that are not related to the root record node but match a definition of the record hierarchy are identified. It is determined whether a set of record nodes unrelated to the root record node was identified. In response to determining that a set of record nodes unrelated to the root record node was not identified, it is determined that records matching the definition of the record hierarchy are positioned in the record hierarchy.

Abstract: A computer-implemented method, system and computer program product for managing relationships between entities in master data management based systems. Record level relationships at an entity level are resolved. Furthermore, a unified view of relationships between entities is determined using composite rules on the underlying resolved record level relationships. Additionally, an anchor member for each linked entity is determined, where the entities are linked together based on the determined unified view of relationships between entities. Furthermore, a record transaction involving creating, updating or deleting a record of an entity linked with another entity is received. The relationship between the linked entities is then validated or invalidated based on an impact of the record transaction with the anchor member of one of the linked entities.

Abstract: Records linking is provided. Two records are selected from a plurality of records corresponding to a customer for pair-wise record comparison. It is determined whether the two records are included in different entities. A local auto-link-threshold value of the different entities is identified in response to determining that the two records are included in different entities. An attribute comparison is performed between the two records. A comparison score is generated based on the attribute comparison between the two records. It is determined whether the comparison score is greater than the local auto-link-threshold value of the different entities. The two records are linked in response to determining that the comparison score is greater than the local auto-link-threshold value of the different entities.

Abstract: A voice command can be received from a user. One or more voice command devices (VCDs) that the voice command is targeting can be determined. A visual indicator of each of the one or more targeted VCDs can be displayed on an XR device worn by the user, wherein each visual indicator visually indicates a respective targeted VCD the voice command is directed to on the XR device.