Abstract: Some embodiments provide a novel method for deploying different virtual networks over several public cloud datacenters for different entities. For each entity, the method (1) identifies a set of public cloud datacenters of one or more public cloud providers to connect a set of machines of the entity, (2) deploys managed forwarding nodes (MFNs) for the entity in the identified set of public cloud datacenters, and then (3) configures the MFNs to implement a virtual network that connects the entity's set of machines across its identified set of public cloud datacenters. In some embodiments, the method identifies the set of public cloud datacenters for an entity by receiving input from the entity's network administrator. In some embodiments, this input specifies the public cloud providers to use and/or the public cloud regions in which the virtual network should be defined. Conjunctively, or alternatively, this input in some embodiments specifies actual public cloud datacenters to use.

Abstract: A large collection of sample genomes containing misclassified k-mers and metadata errors from a reference taxonomy was converted to a self-consistent k-mer database comprising a self-consistent taxonomy. The self-consistent taxonomy was based on genetic distances calculated using the MinHash method or the Meier-Koltoff method. An agglomerative clustering algorithm was used to calculate the self-consistent taxonomy. Each k-mer of the sample genomes was assigned to only one node of the self-consistent taxonomy. In another step, each node of the self-consistent taxonomy was mapped to the reference taxonomy, thereby preserving in the self-consistent taxonomy links to the reference taxonomy while correcting for the misclassification errors therein. The self-consistent k-mer database can be used to taxonomically profile sequenced nucleic acids with greater specificity compared to systems relying on the reference taxonomy.

Abstract: An information management security health monitoring system is provided herein that can monitor the security health of one or more information management systems, identify any security issues that are occurring within an information management system, and automatically, or in response to a user input, transmit an instruction to the information management system to execute a workflow to resolve the security issue(s). For example, the information management cell health monitoring system receives security data and security policies, and/or similar data from an information management cell(s) via a network. The information management security health monitoring system analyzes the received information to identify any issues. If an issue is detected, the information management security health monitoring system retrieves workflows and determines whether any of the workflows can be used to resolve the detected issue.

Abstract: A method performed by a computing system comprises receiving, by a hierarchical document classification system (HDCS), a document that includes image data and textual data, wherein the document is associated with a document type. The HDCS determines a first machine learning (ML) classifier of the HDCS associated with the document type. The first ML classifier corresponds to one of an image-based ML classifier of the HDCS and a text-based ML classifier of the HDCS. The determined first ML classifier classifies the document as being associated with a particular class of document of a plurality of classes of documents. A second ML classifier associated with the class of document classifies the document as being associated with a particular subclass of a plurality of subclasses of the class of document. The HDCS communicates information that specifies the class and the subclass associated with the document to a terminal.

Abstract: System and method for processing time-related geospatial data from one or more data sources. For example, a system includes an application server; and a storage. The application server is configured to: receive data including temporal information and geospatial information for each data object of one or more data objects, send the data to a client device to display the data on a map, and generate one or more first multi-dimensional tiles based at least in part on the temporal information and the geospatial information. The one or more first multi-dimensional tiles correspond to a temporal dimension associated with a first temporal width. The application server is further configured to send the one or more first multi-dimensional tiles to store in the storage for retrieval by the client device.

Abstract: The technology disclosed relates to manipulating and visualizing database objects that represent layers and parts of layers in insurance or other risk transfer programs. Disclosed is graphically creating placement objects, maintaining a database of objects that can be linked in trees, with policy objects linked to respective proposal objects, and coverage objects linked to the proposal objects and the policy objects. A GUI represents a placement using objects, with metadata that describes objects' statuses, with groups of objects properly nested and linked in the database of objects. User selection of a displayed first object and mode control initiates creation of second objects linked to the first object and, opening of a panel that scales the selected first object to fit the panel and allows the user to add nested second objects, by drawing an added object or entering metadata for the added object, with consistent metadata constrained with the nesting.

Abstract: Provided is a sensor network capable of acquiring, from a data center, learning data composed of a pair of sensor data (input) and a classification label (output) necessary for adding/changing a classification label for updating a classifier while reducing the volume of communication between a sensor and the center. One aspect of the present invention relates to a sensor network, including one or more sensor nodes and a data center. The sensor node includes: an encoding unit for encoding sensor data by an encoder part of an autoencoder; and a transmission unit for transmitting the encoded data. The data center includes: a reception unit for receiving data encoded from sensor data by the encoder part of the autoencoder; a decoding unit for decoding the encoded data by a decoder part of the autoencoder; and a storage unit for storing the decoded data therein.

Abstract: Systems, devices, and methods are provided for efficient query execution on distributed data sets, such as in the context of data lakes. In at least one embodiment, indexing information is used to identify candidate and non-candidate portions of a data set. Non-candidate portions may be irrelevant to the query. Indexing information can be encoded using Bloom filters.

Abstract: A computer-implemented method includes a first module of a computing system receiving application-specific data for determining utilization attributes of multiple application programs. A second module of the system receives data for determining patient archetypes. A third module of the system determines a clinical benefit of each application program based on analysis of the application-specific data. The system matches an application program with a particular patient archetype based on analysis of each patient archetype and the determined clinical benefit of the application program. The system selects a particular application program for use by a patient of a first patient group based on: i) a characteristic of the first patient group that is included in a first patient archetype; and ii) the clinical benefit of the particular application exceeding a threshold clinical benefit.

Abstract: A dynamically generated search query is generated based on rarity scores associated with raw-level computer events. Event data is pre-processed using historical information about the frequency, or rarity, of instances of individual events. Each event is assigned one or more labels that identify the event based on the historical information. The rarity scores represent probabilities of events occurring with the same labels. The rarity scores are associated with n-grams of the labels (e.g., a combination of two labels, three labels, etc.). A label n-gram score is calculated based on newly observed events and the rarity scores corresponding to the label n-grams. The search query is generated based on the label n-gram score. The search query is executed against a database to retrieve information, such as diagnostics, used to alert an administrator to events that are potentially anomalous.

Abstract: An artificial intelligence (AI) search system can build or include one or more knowledge graphs and databases of indexes. The knowledge graphs can encode information and relationships from a variety of domains. The databases of indexes can store pointers to public and/or private data and raw content in those domains. An example of a domain includes fields requiring or handling geospatial data. The AI search system can receive a user query and process the query with AI models, including language models and knowledge graphs to identify data matching the intention and context of the user query. The AI search system can present relevant results in an assortment of user interfaces.

Abstract: A method for issuing control signals in response to sentiment. In some embodiments, the method includes: assigning, to each of a plurality of comments, a respective topic vector of length k; determining whether a largest element of the topic vector of a first comment of the plurality of comments exceeds a weight threshold; in response to the determining that the largest element exceeds the weight threshold, classifying the first comment into a first topic, of k topics, the first topic corresponding to the position, in the topic vector, of the largest element of the topic vector; calculating a first sentiment score; calculating an average sentiment score, based in part on the first sentiment score; determining whether the average sentiment score meets a criterion; and in response to the determining that the criterion is met, generating a control signal, the control signal including a message related to the first topic.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform obtaining a set of items that have been grouped together as matching items in a group; performing an ensemble mismatch detection; performing multiple detection models on the set of items to generate respective outputs regarding mismatches; combining the respective outputs to determine whether a quantity of detected mismatches is at least a predetermined threshold; when the quantity of detected mismatches is at least the predetermined threshold, the acts also can include separating at least one of the set of items from the group; and when the quantity of detected mismatches is not at least the predetermined threshold, the acts additionally can include maintaining each item of the set of items in the group. Other embodiments are disclosed.

Abstract: One or more computing devices, systems, and/or methods for constructing and implementing a network management plan are provided. Baseline classifications are assigned to a set of locations. Sub-locations at the set of locations are evaluated to determine whether to override any of the baseline classifications. A model is used to evaluate images depicting the set of locations to generate predicted classifications for the set of locations. Classifications are assigned to the set of locations by implementing conflict resolution rules to selectively retain or replace baseline classifications with predicted classifications. The classifications are used to construct and/or implement a network management plan.

Abstract: Systems and methods for managing access to data stored in a distributed manner across different zones of a multi-zone computing platform involve maintaining a global representation of all data stored within the multi-zone computing platform. After receiving a request on behalf of a user to access given data, a first zone of the multi-zone computing platform (i) obtains, from the global representation, information indicating a particular zone where the given data is stored, (ii) validates the request, and (iii) if the validation is successful, processes the request. If the given data is stored at the first zone, the first zone retrieves the given data for provision to the user. If the given data is stored at a second zone, the first zone issues a request to the second zone to validate the request, and if the validation is successful, retrieve the given data for provision to the user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing filters from resource content. In one aspect, a system receives data identifying a set of resources that are determined to be responsive to a search query and extracts a set of keywords from the contents of the resources and related queries. The keywords are processed according to candidate selection criteria, and a set of candidate query filters are determined. The candidate filters may be used to filter the resources that are responsive to the query.

Abstract: An information processing device includes: a memory; and a processor coupled to the memory, the processor being configured to: sort stream data buffered in units of wraps of a sequential recording medium, in a column order and a time order of the stream data, as primary data to be written into a primary wrap of the sequential recording medium; and control writing of the sorted primary data into the primary wrap, wherein the sorting of the stream data is configured to sort secondary data to be written into a secondary wrap that follows the primary wrap, in a reverse order of the column order and in the time order, and wherein the controlling of the primary data is configured to control writing of the sorted secondary data into the secondary wrap.

Abstract: Embodiments described herein provide natural language processing (NLP) systems and methods that provide a customized summarization of scientific or technical articles, which disentangles background information from new contributions, and summarizes the background information or the new information (or both) based on a user's preference. Specifically, the systems and methods utilize machine learning classifiers to classify portions of sentences within the article as containing background information or as containing a new contribution attributable to the article. The systems and methods then incorporate the background information in the summary or incorporate the new contribution in the summary and output the summary. In this way, the systems and methods can provide summaries of scientific literatures, which largely accelerates literature review in scientific fields.

Abstract: An improved hash table structure compatible with in-memory processing for increasing cache efficiency during hash join processing of a small and large table in a relational database system. The hash table, residing in processor memory, includes a first partition containing a join condition column providing best selectivity for joining the small table with the large table, at least one additional partition containing additional join condition columns for joining the small table with the large table; and an array of hash values, the array of hash values providing an index into the hash table partitions.

Abstract: In a method for managing data in a partition table, a server, and a storage medium, a management node obtains a creation duration range of data of a target partition in the partition table, further obtains a geographical area to which the target partition belongs, historical access information of the target partition, or a service type of the target partition; and then, when the creation duration range of the data of the target partition meets a predetermined time range of a first storage stage and a preset condition is met, the management node alters the data that is of the target partition and that has been stored based on a storage policy corresponding to a second storage stage to be stored based on a storage policy corresponding to the first storage stage.

Abstract: Embodiments for automatic extraction of a compliance profile for an organization by a processor. Text data may be extracted from one or more data sources representing one or more objects describing a compliance named entity of type organization expected to perform an obligation. A compliance profile of type organization may be determined for the compliance named entity of type organization according to the extracted text data.

Abstract: A method of persisting and performing a clustering analysis through use of a large data electronic file system includes generating a job identifier and linking the job identifier with a configuration identifier, a plurality of model identifiers and a plurality of data regularization identifiers. Each of the configuration identifier, model identifiers and data regularization identifiers are stored in respective management tables of the file system along with meta-data indicating a physical location of an analysis configuration, a physical location of a data regularizer and a physical location of a clustering model, respectively. The method further includes specifying the job identifier to a clustering analysis application causing the analysis configuration, the clustering models and the data regularizers to load into the clustering analysis application and receiving a plurality of scores resulting from a cluster analysis performed by the clustering analysis application based on the job identifier.

Abstract: An exemplary query parsing system accesses user input text representative of a search query. The system generates a plurality of candidate parses of the user input text. Each of the plurality of candidate parses includes a candidate what portion and a candidate where portion determined from the user input text. The system generates a score for each candidate parse in the plurality of candidate parses based on whether one or more of a plurality of predefined parsing features are found in the respective candidate parse and on a machine learning mode. The system selects, from the plurality of candidate parses based on the scores for the plurality of candidate parses, the candidate what portion and the candidate where portion of a particular candidate parse included in the plurality of candidate parses as search parameters for the search query. Corresponding methods and systems are also disclosed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing filters from resource content. In one aspect, a system receives data identifying a set of resources that are determined to be responsive to a search query and extracts a set of keywords from the contents of the resources and related queries. The keywords are processed according to candidate selection criteria, and a set of candidate query filters are determined. The candidate filters may be used to filter the resources that are responsive to the query.

Abstract: A URL categorization query is received. The URL categorization query includes at least one URL. The URL is used to determine a set of keys. A database is queried using the determined set of keys. Categorization information is returned. The returned URL categorization information can be used to enforce policies.

Abstract: Methods, computer program products, computer systems, and the like providing for executing orchestration operations across data center infrastructures are disclosed. In one embodiment, the method includes analyzing a property graph to determine whether a node representing at least one entity in a first data center infrastructure has a contact point with a node representing one or more entities representing one or more core physical or hardware-based resources in a second data center infrastructure. If a contact point exist between nodes of associated with the first and second data centers, the orchestration operation is executed on the at least one entity in the first data center and a corresponding orchestration operation is executed on at least another entity in the second data center infrastructure represented at a contact point in the dependency relationships of the property graph.

Abstract: Provided is a method for managing source identifiers of clustered records, which may include receiving a plurality of records, each having identification data associated with a source thereof. The records may be sorted into a first plurality of clusters. A first source identifier may be generated for each cluster. Update data associated with an update to the records may be received. At least some of the records may be sorted into a second plurality of clusters. A first cluster of the second plurality of clusters may have first identification data different from at least one cluster of the first plurality of clusters, and a second cluster of the second plurality of clusters may have identification data that substantially matches the at least one cluster. Whether to assign the first source identifier to the first or second cluster may be determined. A system and computer program product are also disclosed.

Abstract: A method, system and computer program product for facilitating risk mitigation of information security threats. Data obtained from at least one tracked data source is analyzed for identifying at least one event related to a threat, to be stored in a database comprising date and time of each event identified, enabling generation of threat timeline comprising temporally ordered sequence of each event related to respective threat identified. Features selected using correlation between features from threat timelines in the database and labeling assigned using records of threat usage incidents are extracted from events in threat timeline for the threat which the at least one event related thereto being identified and based thereon a dynamic score indicating an estimated level of risk posed by the threat is calculated using at least one machine learning model for predicting threat usage during a time window defined, enabling risk mitigation based on outputted indication thereof.

Abstract: This disclosure relates generally to retrieval of prior court cases using witness testimonies. Conventional state-of-the-art methods use supervised techniques for answering basic questions in legal domain using numerous features and do not address interpretability of results and the performance and precision of retrieving prior court cases for these methods are less. Embodiments of the present disclosure obtains an embedded representation for an event structure of a user query and testimony sentences identified from prior court cases using a trained Bi-LSTM classifier and a set of linguistic rules. A similarity is estimated between the embedded representation for the event structure of the user query and the event structure of each testimony sentence from the prior court cases. Further a relevance score is assigned in accordance with the estimated similarity to retrieve the relevant prior court cases. The disclosed method is used to retrieve the relevant prior court cases using witness testimonies.

Abstract: A method is provided to predict a disaster for a computer system based on logs. The method includes representing existing logs as first vectors by tokenizing the existing logs and partitioning the first vectors into clusters. The clusters represent disaster types. The method further includes selecting representative vectors for the clusters, representing a new log of the computer system as a second vector by tokenizing the new log, matching the second vector to a cluster by comparing the second vector and the representative vectors, and categorizing the new log as a disaster type represented by the cluster.

Abstract: A method includes generating a base model by training with a first dataset of data pairs and generating an adapted model by training the base model on a second dataset of data pairs. The method also includes determining a contrastive score for each data pair of a third dataset of data pairs using the base model and the adapted model. The contrastive score is indicative of a probability of quality of the respective data pair. The method also includes training a target model using the data pairs of the third dataset and the contrastive scores.

Abstract: Disclosed are methods, systems, and computer-readable medium for data wrangling. For instance, the method may include: determining one or more data operations to be performed on a multi-dimensional dataset; determining at least one operation type associated with the one or more data operations to be performed; dividing the multi-dimensional dataset into a plurality of sub-tables based on the operation type; providing the plurality of sub-tables to a data wrangling process, the one or more data operations being performed on each of the plurality of sub-tables in parallel to determine a processed plurality of sub-tables; and determining, based on the processed plurality of sub-tables, a finished multi-dimensional dataset. The data wrangling process may process sub-tables (e.g., change-only columns/rows) of tables of the multi-dimensional dataset and/or process sub-tables in parallel for data operations that are not related.

Abstract: A testing device includes a first interface device configured to enable communication with a first component of a vehicle and a second interface device configured to enable communication with a second component of the vehicle. The testing device includes a test module storage configured to store one or more test modules and also includes a user interface and a test controller. The test controller is responsive to the user interface to select a test module from the test module storage and to cause the first interface device to communicate first test data to the first component of the vehicle and to cause the second interface device to communicate second test data to the second component of the vehicle to perform a vulnerability test associated with the selected test module.

Abstract: A server, method, and non-transitory computer readable medium for classifying queries based on contextual information are provided. The server includes a network interface, a memory storage unit and a processor. The method involves receiving a plurality of queries, analyzing the queries and determining a likelihood divergence and selecting a domain. The non-transitory computer readable medium is encoded with programming instructions to direct a processor to carry out the method.

Abstract: The disclosed embodiments provide a system for streamlining machine learning. During operation, the system determines a resource overhead for a baseline version of a machine learning model that uses a set of features to produce entity rankings and a number of features to be removed to lower the resource overhead to a target resource overhead. Next, the system calculates importance scores for the features, wherein each importance score represents an impact of a corresponding feature on the entity rankings. The system then identifies a first subset of the features to be removed as the number of features with lowest importance scores and trains a simplified version of the machine learning model using a second subset of the features that excludes the first subset of the features. Finally, the system executes the simplified version to produce new entity rankings.

Abstract: Certain aspects involve updating data structures to indicate relationships between attribute trends and response variables used for training automated modeling systems. For example, a data structure stores data for training an automated modeling algorithm. The training data includes attribute values for multiple entities over a time period. A computing system generates, for each entity, at least one trend attribute that is a function of a respective time series of attribute values. The computing system modifies the data structure to include the generated trend attributes and updates the training data to include trend attribute values for the trend attributes. The computing system trains the automated modeling algorithm with the trend attribute values from the data structure. In some aspects, trend attributes are generated by applying a frequency transform to a time series of attribute values and selecting, as trend attributes, some of the coefficients generated by the frequency transform.

Abstract: Domain name system (DNS) configuration during virtual private network (VPN) connection includes, by a VPN entry server, receiving from a client device, via a VPN tunnel between the entry server and the client device, a first request for first content that identifies a first external source for the first content, receiving, from an operative DNS server configured for the tunnel, an Internet Protocol (IP) address of a first VPN system exit server, in response to determining that the first content is unavailable via the first system exit server, identifying a second DNS server, such that a second VPN system exit server for obtaining the first content is available using the second DNS server, obtaining, from the second system exit server, the first content, wherein the second system exit server obtained the first content from the first external source, and transmitting, to the client device, via the tunnel, the first content.

Abstract: Domain name system (DNS) configuration during virtual private network (VPN) connection includes transmitting, to a VPN entry server, from a client device, via a VPN tunnel between the VPN entry server and the client device, a first request for first content, wherein the first request identifies a first external source for the first content, and wherein a first DNS server is configured as an operative DNS server for the VPN tunnel, and receiving, by the client device, from the VPN entry server, via the VPN tunnel, the first content, wherein the VPN entry server obtained the first content from the first VPN system exit server identified by the VPN entry server using a second DNS server as the operative DNS server for the VPN tunnel, and the first VPN system exit server obtained the first content from the first external source.

Abstract: Domain name system (DNS) configuration during virtual private network (VPN) connection includes establishing a VPN tunnel between a client device and a VPN system entry server, which includes configuring a first DNS server as an operative DNS server for the VPN tunnel, and obtaining first content by transmitting to the VPN entry server, a first request that identifies a first external source for the first content, receiving from the VPN entry server a DNS configuration message indicating a second DNS server, configuring the second DNS server as the operative DNS server, and receiving from the VPN entry server, via the VPN tunnel, the first content, wherein the VPN entry server obtained the first content from the first VPN system exit server identified by the VPN entry server using the second DNS server, and the first VPN system exit server obtained the first content from the first external source.

Abstract: Devices and techniques are generally described for search using a micro-partitioned catalog. In some examples, a search system can determine a plurality of items in an online catalog. The search system can generate a similarity graph based on vector representations of the plurality of items and determine an item as a representative item for a cluster of items. The search system can receive a query and determine the cluster of items as a candidate cluster based on a comparison between the query and the representative item. The search system can determine a candidate item of the cluster of items based on a comparison between the query and the cluster of items.

Abstract: In some aspects, a computing device receives a scan of a code displayed on an order post located near a restaurant, determines that the code is associated with the restaurant, and automatically opens a software application and navigates the software application to an ordering page associated with the restaurant. The computing device initiates receiving, via the software application, input associated with an order, sends the input to a machine learning based software agent executing on a server, receives a predicted response to the input, provides the predicted response as audio output and/or displays the predicted response on the touchscreen display device. After the order is complete, the computing device sends order data associated with the order to the restaurant. After receiving an indication from the restaurant that the order is ready, the computing device indicates that the order is ready to be picked up.

Abstract: Methods, systems, and computer programs are presented for displaying a customized news stream. One method includes an operation for identifying dwell times spent by users while accessing a first plurality of items, each item belonging to one media type from a plurality of media types (e.g., news articles, videos, slide shows, etc.). In addition, the method includes operations for determining statistical parameters for each media type based on the identified dwell times, and for detecting a news corpus having a second plurality of items. A priority for each item in the news corpus is determined based on the media type of the item, the corresponding statistical parameters for the media type of the item, and the profile of a user. The news stream is sent to the user for presentation on a display, the news stream being sorted based on the priority of the items in the news corpus.

Abstract: An apparatus includes a processor operatively coupled to a memory. The processor detects a software application installed on a client computing device, and/or usage data. Detected usage data is associated with a current user of the client computing device and with the software application. The processor identifies a user role for the current user based on the software application and/or usage data. The processor applies a security configuration to the client computing device based on the user role. The security configuration limits access by the current user to a portion of the software application. The processor sends an identifier of the user role to an administrative server for storage in an Active Directory (AD) database.

Abstract: A system and method of providing user customizable web advertisements are disclosed. In some embodiments, the system may include a non-transitory, computer-readable medium storing computer-executable instructions and one or more processors. When the one or more processors execute the computer-executable instructions, the processors may be configured to receive a customizable advertisement, the customizable advertisement having at least one customizable portion. When a customizable portion is selected, the one or more processors may display a plurality of customization options for the at least one customizable portion. The one or more processors may be further configured to receive a selection of a customization option from the plurality of customization options. The one or more processors may then apply the selected customization option to the customizable portion of the customizable advertisement.

Abstract: Using a natural language (NL) latent presentation in the automated conversion of source code from a base programming language (e.g., C++) to a target programming language (e.g., Python). A base-to-NL model can be used to generate an NL latent representation by processing a base source code snippet in the base programming language. Further, an NL-to-target model can be used to generate a target source code snippet in the target programming language (that is functionally equivalent to the base source code snippet), by processing the NL latent representation. In some implementations, output(s) from the NL-to-target model indicate canonical representation(s) of variables, and in generating the target source code snippet, technique(s) are used to match those canonical representation(s) to variable(s) of the base source code snippet. In some implementations, multiple candidate target source code snippets are generated, and a subset (e.g., one) is selected based on evaluation(s).

Abstract: As described herein, a system, method, and computer program are provided for normalizing a JSON structure. In use, input defining at least one entity type of a target data structure is received. A source JSON structure is identified. The source JSON structure is traversed for a particular JSON data type to map values in the source JSON structure to corresponding entities of the target data structure based on the at least one entity type defined for the target data structure, where each entity of the target data structure is defined using a relative path between nodes of the source JSON structure.

Abstract: The present disclosure relates to a retrieval method including: generating a graph representing a set of users, items, and queries; generating clusters from the media items; generating embeddings for each cluster from embeddings of the items within the corresponding cluster; generating augmented query embeddings for each cluster from the embedding of the corresponding cluster and query embeddings of the queries; inputting the cluster embeddings and the augmented query embeddings to a layer of a graph convolutional network (GCN) to determine user embeddings of the users; inputting the embedding of the given user and a query embedding of the given query to a layer of the GCN to determine a user-specific query embedding; generating a score for each of the items based on the item embeddings and the user-specific query embedding; and presenting the items having the score exceeding a threshold.

Abstract: Examples of the present disclosure describe systems and methods for determining relationships between content items to create a visualization associated with the various content items. The visualization may provide information regarding what various individuals in a group, team, or organization have been working on (e.g., content, documents, projects).

Abstract: Methods, systems, and computer-readable media are disclosed herein to provide rule-based reconciliation of records. Specifically, rules are utilized to reconcile one or more records and identify duplicates therein. Once duplicate records are identified, one or more ranking sets can be utilized to identify which of the duplicate records to write to the system.

Abstract: The present disclosure is directed to systems and methods for reference source matching. Specifically, the systems and methods disclosed enable matching among tokens using a reference source. In one example, a Consolidation Platform may receive tokens from a customer environment and tokens from a reference source environment. The customer tokens may be compared to each other using AB matching. If a match does not occur, the customer tokens may further be compared to the reference source tokens via transitive matching. If a match does occur, then the customer tokens may be denoted as a match. In further example aspects, the reference source may be a universal reference token repository that comprises unique tokens. If, after a match is indicated, the matched token(s) may be compared to the universal reference token repository. If the matched token(s) does not exist, it may be added to the repository for future use.