Abstract: Methods, systems, and computer-readable storage media for improving the accuracy of embedding vector generation for domain-specific text for purposes of semantic search and generative AI. A dictionary of domain-specific terms can be built to include embedding vectors generated for respective domain-specific terms using a pre-trained large language model. A list of domain-specific terms pertaining to a particular domain can be identified and textual content comprising a description or a definition of a respective domain-specific term can be obtained. A domain-adapted embedding vector for each domain-specific term can be generated based on the textual content for that term. The domain-specific dictionary can be built to include a combination of a domain-specific term, a corresponding domain-adapted embedding vector and the domain-specific term definition or description.

Abstract: One or more systems, devices, computer program products and/or computer-implemented methods provided herein facilitate generation of an essentialized document that aids a technician in solving a problem with an asset. In an embodiment, the essentialized document is customizable based on technician skill. In another embodiment, the essentialized document is customizable based on the asset and asset components at issue.

Abstract: A computerized system and method may automatically generate a hierarchical, multi-tiered taxonomy based on measuring and/or quantifying degrees of generality for a plurality of input entities. In some embodiments of the invention, a computerized system comprising a processor, and a memory including a plurality of entities such as documents or text files, may be used for extracting words from a plurality of documents; calculating generality scores for the extracted words; selecting some of the extracted words as exemplars based on the scores; and clustering unselected words under appropriate exemplars to produce or output a corresponding taxonomy. Some embodiments of the invention may allow categorizing interactions among remotely connected computers using a domain taxonomy, and routing interactions between remotely connected computer systems based on the taxonomy.

Abstract: The disclosure is directed to a document management system having a label management user interface. The document management system may be configured to display document content in either a document view mode or a document edit mode in which document content is displayed in a content panel of the graphical user interface. When in the document view mode and an authenticated user has edit permissions with respect to a current document, the graphical user interface is configured to display a label management user interface including an array of user-selectable label graphical objects and a list of recommended labels selected in accordance with a label score. The label score for each recommended label is determined with respect to the current document based on a set of multiple heuristics including proximity of the document with respect to other labeled documents in a hierarchical element tree.

Abstract: A computer-readable recording medium stores an information verification program for causing a computer to execute a process. The process includes: acquiring posted information that includes first information which is text information, second information which is at least one of image information and audio information, and metadata of the second information; specifying an event that occurs, based on the first information; specifying an influence range that indicates a range of at least one of a geographical influence and a temporal influence by the event; and calculating a first reliability level of the posted information based on the influence range and the metadata.

Abstract: Methods and systems for site prospecting includes the operations of: receiving a site request indicating a required use for a site; generating a plurality of capacity scores corresponding to a plurality of land parcels using a first machine learning model; filtering the plurality of land parcels into a subset of land parcels based on the plurality of capacity scores; and for at least one land parcel in the subset of land parcels: generating a parcel potential description using a first language model based at least in part on geographic information associated with the at least one land parcel; generating a parcel potential score using a second machine learning model based at least in part on the parcel potential description; and presenting the parcel potential description and the parcel potential score.

Abstract: A computer system associated with a media-providing service obtains a set of text descriptors from a plurality of media items in a pool of media items personalized for a first user. The system selects, from the set of text descriptors, a first text descriptor using first criteria and a second text descriptor using second criteria. The system causes an application associated with the media-providing service, executing on a client device of the first user, to concurrently display a first option for playing back a first playlist of media items associated with the first text descriptor and a second option for playing back a second playlist of media items associated with the second text descriptor. In response to receiving selection by the first user of the first option, the system provides, to the application, media items from the first playlist of media items associated with the first text descriptor.

Abstract: One embodiment of the present invention sets forth a technique for processing a geometric query. The technique includes determining a tree structure and a set of configurable parameters specified in the geometric query, wherein the set of configurable parameters includes one or more query primitives and one or more search routines. The technique also includes performing one or more operations that traverse the tree structure and execute the one or more search routines to match the one or more query primitives to a first set of geometric objects included in the tree structure. The technique further includes generating a response to the geometric query, wherein the response includes the first set of geometric objects.

Abstract: A method for correcting translation errors when performing GUI globalization testing is disclosed. In one embodiment, such a method takes a screenshot of a graphical user interface (GUI). The method further enables a user to select text in the screenshot. In certain embodiments, one or more of the text and coordinates associated with the text are stored in picture attributes associated with the screenshot. In response to the selection, the method automatically retrieves a file in which at least some portion of the text is contained. The method automatically locates, within the file, the portion, and enables the user to update the portion in the file in order to update corresponding text in the graphical user interface. In certain embodiments, the method provides a probability percentage associated with the file that indicates a probability that the text in the file links to the selected text in the graphical user interface.

Abstract: Approaches presented herein provide for the generation and maintenance of a minimally-sized octree (or other spatial representation) for an arbitrary or dynamic dataset. An octree representation allows for efficient real-time querying of dynamic content, where such a spatial database can scale from a single machine to multiple machines. Objects in a scene can be stored to the spatial database using a set of spatial primitives. When a first object (or set of objects) in a scene is determined, an initial octree and root can be determined based at least in part upon the size and location of the object(s). An additional object may be added that is outside the existing root of the octree. A new bounding volume is generated that surrounds this new object, and the bounding volume is grown in alternating directions until the bounding volume also includes the other objects in the environment, and this new volume boundary becomes the new root of an updated octree for this environment.

Abstract: A query acquisition unit (610) acquires pose information (hereinafter, described as query pose information) that is information to be a query and indicates a pose of a person. A search information acquisition unit (620) acquires a plurality of pieces of search pose information. The search pose information is pose information about an image (hereinafter, described as a target image) to be a search target, and is stored in a database by a plurality of target images. A selection unit (630) selects, from among the plurality of pieces of search pose information, two or more pieces of the search pose information whose degree of similarity to the query pose information satisfies a reference. The selection processing becomes substantially processing of selecting a target image similar to an image (hereinafter, described as a query image) associated with the query pose information.

Abstract: An image retrieval system with high retrieval accuracy is provided. The image retrieval system includes a database and a processing portion. The database has a function of storing a plurality of pieces of database image data, and a database tag is linked to each of the plurality of pieces of database image data. The processing portion has a function of obtaining database image feature value data representing a feature value of the database image data for each piece of the database image data. The processing portion has a function of obtaining query image feature value data representing a feature value of the query image data. The processing portion has a function of calculating first similarity of the database image data to the query image data for each piece of the database image data. The processing portion has a function of obtaining a query tag linked to the query image data using some of the database tags.

Abstract: Example techniques relate to a playback device that facilitates queue identification. In an example implementation, a playback device receives, from a first controller, instructions representing a command to populate a playback queue with one or more first media items; the instructions are associated with a first application identifier corresponding to the first controller. Based on the received instructions, the playback device populates the playback queue with the one or more first media items and forms an association between the playback queue and the first application identifier. The playback device receives, from a second controller, instructions representing one or more first commands to access the playback queue; the instructions are associated with a second application identifier corresponding to the second controller.

Abstract: Example systems and methods for automated cover song identification are disclosed. An example apparatus includes at least one memory, machine-readable instructions, and one or more processors to execute the machine-readable instructions to at least execute a constant Q transform on time slices of first audio data to output constant Q transformed time slices, binarize the constant Q transformed time slices to output binarized and constant Q transformed time slices, execute a two-dimensional Fourier transform on time windows within the binarized and constant Q transformed time slices to output two-dimensional Fourier transforms of the time windows, generate a reference data structure based on a sequential order of the two-dimensional Fourier transforms, store the reference data structure in a database, and identify a query data structure associated with query audio data as a cover rendition of the audio data based on a comparison of the query and reference data structures.

Abstract: Systems and methods for processing and indexing media content for summarization and generating insights. A media source provides audio/video files to a media pre-processor to enhance the quality of an audio-video stream. A media processor comprises a video feature extractor, an audio feature extractor and a text feature extractor to extract base features from the audio-video stream at various time intervals. Higher-level features are extracted by executing Artificial Intelligence algorithms over the base features. A feature combining and indexing unit combines the extracted base features and higher-level features for time-stamping and indexing of the audio-video stream. Insightful output is generated based on the input content. The audio-video stream indexed by time and features are stored in a database along with the generated insights. A data explorer allows efficient searching, editing and compression of the audio/video files.

Abstract: In order to implement the foregoing object, an exemplary embodiment of the present disclosure discloses a video retrieval method performed by a computing device. The video retrieval method may include: receiving retrieval query data for one or more video data; generating a retrieval query vector based on the received retrieval query data; comparing the retrieval query vector with a plurality of video retrieval vectors each representing target retrieval video data segmented based on key frame information; and selecting one or more target retrieval video data by the result of the comparison between the retrieval query vector and the plurality of video retrieval vectors.

Abstract: A data construction system and a data construction method. The data construction system includes a construction process management unit and a data construction unit. The data construction unit is configured to receive an initiation construction request. The data construction unit obtains corresponding construction rule data from the construction process management unit according to the initiation construction request, and sends an obtaining source data request to the construction process management unit. The construction process management unit receives the obtaining source data request, and calculates a new access location according to a source data identification number and access location information in the obtaining source data request. The construction process management unit obtains corresponding source data according to the new access location, and updates the access location information according to the new access location.

Abstract: A method may include obtaining a dataset of graph-structured data, the dataset including one or more subgraphs. The method may include determining a correlation between analyses of the dataset using graph explainable artificial intelligence (GXAI) techniques and using a plurality of graph analysis algorithms. A ranked list of the plurality of graph analysis algorithms may be generated based on the correlation. The method may further include determining general characteristic of the dataset of graph-structured data, the general characteristic indicative of similarities among the one or more subgraphs. A threshold number of the graph analysis algorithms may be assigned to the general characteristic based on the ranked list of the plurality of graph analysis algorithms. An assignment table may be generated including the general characteristic and the threshold number of the graph analysis algorithms. A display may be generated within a graphical user interface (GUI) that visualizes the assignment table.

Abstract: According to one aspect, causal graph chain reasoning predictions may be implemented by generating a causal graph of one or more participants within an operating environment including an ego-vehicle, one or more agents, and one or more potential obstacles, generating a prediction for each participant within the operating environment based on the causal graph, and generating an action for the ego-vehicle based on the prediction for each participant within the operating environment. Nodes of the causal graph may represent the ego-vehicle or one or more of the agents. Edges of the causal graph may represent a causal relationship between two nodes of the causal graph. The causal relationship may be a leader-follower relationship, a trajectory-dependency relationship, or a collision relationship.

Abstract: Systems, apparatuses and methods may provide for technology that determines a plurality of means based on a plurality of vectors, wherein each mean in the plurality of means corresponds to center of a cluster, assigns each vector in a plurality of vectors to a mean in the plurality of means, and conducts a compression of the plurality of vectors based on the plurality of means. The technology may also build a directed graph based on the compressed plurality of vectors and update the directed graph. Updating the graph may involve determining a plurality of modified means, detecting that a change in one or more modified means in the plurality of modified means exceeds a threshold, conducting an update of the modified mean(s), and bypassing the update for one or more remaining means in the plurality of modified means.

Abstract: A hypergraph workload manager in a server is configured for failure tolerant and explainable state machine driven hypergraph execution. The hypergraph executor comprises a query optimizer, a hypergraph enlister, a pipeline analyzer, and a state machine generator. The query optimizer translates a user query into a query operator graph. The hypergraph enlister enlists the query operator graph into a hypergraph containing a set of query operator graphs representative of already submitted user queries. The enlistment is configured to join query operator graphs where it makes sense to optimize query executions. Updates to the hypergraph based on the enlistment results in a set of disconnected graphs. The pipeline analyzer performs an analysis of all operators of all queries in the hypergraph to find an optimal sequencing of execution. The state machine generator is configured to generate a hierarchical state machine for all operators of a disconnected graph of the hypergraph.

Abstract: A method may include obtaining a graph including components of a facility and connections among the components, receiving a query corresponding to a subset of the components, and generating, based on the graph and according to the query, a visual representation comprising the subset of the components and connections among the subset of the components.

Abstract: A related-information display device includes: a knowledge graph DB that stores a knowledge graph holding knowledge information with a plurality of nodes and links connecting the nodes; a related-information inference unit that infers related information related to a keyword from the knowledge graph; a relatedness calculation unit calculates relatedness between the keyword and the related information; and a display-data generation unit that generates display data for displaying a flow rate diagram indicating a relationship between the keyword and the related information related to the keyword by connecting the keyword and the related information with a band, wherein the width of the band increases as the relatedness increases.

Abstract: In certain aspects, a computer-implemented method includes receiving an incoming element; determining whether a distance of the incoming element to the nearest existing element that is found is above a clustering threshold; creating, based on determining the distance is above the clustering threshold, a new cluster and associating the new cluster to the incoming element; determining, based on determining the distance is not above the clustering threshold, whether distances from the incoming element to contents of the nearest existing element that is found are all below the clustering threshold; associating, based on determining the distances from the incoming element to the contents of the nearest existing element that is found are all below the clustering threshold, associating the incoming element to the cluster of the nearest existing element that is found; processing, based on determining the distances from the incoming element to the contents of the nearest existing element that is found are not all belo

Abstract: A system includes a core flood apparatus, a laboratory computing device configured to collect laboratory data when performing a core flooding test using the core flood apparatus, and a data control and data monitor server communicatively interfacing with the laboratory computing device and configured to receive the laboratory data, and send instructions to the laboratory computing device. The system further includes a data management and analysis facility that is physically separate from the laboratory computing device and the core flood apparatus, and configured to receive the laboratory data from the data control and data monitor server and store the laboratory data.

Abstract: Embodiments are provided for summarization and recommendation of content. In disclosed embodiments, a summarization engine scores constituent parts of content, and generates a plurality of summaries from a plurality of points of view for the content based at least in part on the scores of constituent parts. The summaries may be formed with constituent parts extracted from the contents. A recommendation engine provides recommendations to a user based on rankings of the summaries generated by the summarization engine. Other embodiments may be described and/or claimed.

Abstract: Aspects described herein may allow for generating a filter list to facilitate users to refine their online search. Data items displayed on a web page may be determined based on the user's input data, for example, based on a selection command selecting the data items. Each data item corresponds to a feature of a product. Filters corresponding to the data items may be determined and the weight of each filter may be updated. A filter list may be generated and displayed to facilitate the user to refine his search for a class of products. The filters on the filter list may be automatically ordered based on the weight of each filter. In this way, a user may be able to find filters that are associated with features that he cares about the most and therefore use the filter list to obtain more satisfying results.

Abstract: Systems and methods are provided that implement spatio-temporal query of perception data streams. A system is distinctly designed to implement efficient and flexible spatio-temporal queries, which conduct pattern matching over perception data streams for automotive applications, particularly autonomous vehicles. The system is also designed to enable the spatio-temporal queries to be expressed in SpREs (Spatial Regular Expressions), which is implemented as a querying language that combines the ease of REs (regular expressions) with the enhanced capabilities of spatial logic. Additionally, a method includes receiving a command associated with a spatio-temporal query, wherein the expression comprises spatial regular expressions (SpREs). Thereafter, performing the spatio-temporal query of a plurality of perception data streams using the SpREs, where the SpREs describe spatio-temporal patterns to be searched within the plurality of perception data streams.

Abstract: Methods, systems, and apparatus including computer programs encoded on a computer storage medium, for retrieving image search results using embedding neural network models. In one aspect, an image search query is received. A respective pair numeric embedding for each of a plurality of image-landing page pairs is determined. Each pair numeric embedding is a numeric representation in an embedding space. An image search query embedding neural network processes features of the image search query and generates a query numeric embedding. The query numeric embedding is a numeric representation of the image search query in the same embedding space. A subset of the image-landing page pairs having pair numeric embeddings that are closest to the query numeric embedding of the image search query in the embedding space are identified as first candidate image search results.

Abstract: Provided are systems and methods for providing insights to users of a mobile application based on triggers that may be detected from within the data or externally. In one example, a method may include receiving, via a message transmitted to a uniform resource locator (URL) of a host platform, information about a user of a software application, reading a plurality of records associated with the user from the received information, the plurality of records comprising data and a plurality of timestamps, respectively, generating derived data based on the data included in the plurality of records, converting the plurality of timestamps into a single timing identifier that represents all of the plurality of timestamps, and transmitting, to a computing system, a notification comprising the derived data and the new timing identifier.

Abstract: Aspects of the disclosure relate to validation of a request for an asset based on information in the request. Validation of the request can be processed by a network node in a network repository that contains the asset. In one aspect, validation of the request can comprise determining if the request is legitimate based at least on processing at least a portion of the information in the request. In response to a legitimate request, the network node can process the request and, as a result, a data object associated with the asset can be supplied to a device originating the request. In the alternative, the network node can yield an exception.

Abstract: With conventional technologies, it is impossible to accumulate and use management information, which is information based on user operations on web pages, in a terminal device, in association with the web pages. With a terminal device 1 including: an operation acceptance unit 121 that accepts one or more operations performed on a web page; a management information acquisition unit 133 that acquires management information that is a set of pieces of operation information regarding the one or more operations, or a dynamic attribute value that is based on the set of pieces of operation information; and a management information accumulation unit 134 that accumulates the management information in association with a page identifier that identifies the web page, it is possible to accumulate and use management information that is based on user operations on web pages, in a terminal device, in association with the web pages.

Abstract: An example methodology includes, by a computing device, analyzing an application to identify one or more elements in the application and, for each element of the one or more elements, an attribution of the element. The method also includes, for each element of the one or more elements, by the computing device, determining an attribute of the element which is a non-duplicate attribute, creating a first accessibility to the element by the non-duplicate attribute, creating a second accessibility to the element by XPath, and storing the first accessibility and the second accessibility created for the element within a central repository. The method may also include, by the computing device, generating a handler class for an element of the one or more elements in the application, wherein the handler class contains code for accessing the element, and providing the generated handler class to another computing device.

Abstract: The present disclosure describes how to capture events (e.g., changes and user interactions) of a document and combine those changes with the original tree data structure displayed to accurately and efficiently enable a replay engine to redisplay the tree data structure, changes, and user interactions which occurred at the client device. The data collected from a client-side capture engine can be combined with a minimal amount of contextual information to a replay engine so as to accurately and efficiently replay a session of a plurality of documents.

Abstract: One method includes receiving a request concerning resolution of a QUBO (quadratic unconstrained binary optimization) problem, randomly generating benchmark QUBOs, obtaining information, including telemetry and solutions, concerning execution of the benchmark QUBOs on each annealer in a group of annealers, and comparing respective performances of each of the annealers. Each of the QUBOs is associated with a respective QUBO matrix, and each QUBO matrix having the same size and density.

Abstract: An orchestration is configured to identify a compilation job and an annealing job of a quantum job. The compilation job and the annealing job may be orchestrated based on estimated costs, resources availability, and service level objectives. The compilation job and the annealing job may be orchestrated in the context of orchestrating multiple quantum jobs. The orchestration engine may orchestrate the compilation and annealing jobs in the same or different systems.

Abstract: Systems and methods for solving multivariate state-estimation problems in accordance with numerous embodiments of the invention are illustrated. One embodiment includes a method that develops a characteristic function associated with a state-estimation problem. The method collects, for a set including at least one time point, from a sensory instrument, a set of measurements reflecting the state corresponding to the most recent time point. The method derives, from the set of measurements, new parameters corresponding to the state of the system at the most recent time point. The method updates a hyperplane arrangement based on the new parameters. The method runs a cell-enumeration algorithm based on the hyperplane arrangement to obtain sign vectors for each cell of the hyperplane arrangement. The method determines, based at least in part on the sign vectors, function parameters for the updated characteristic function, wherein the updated characteristic function includes a linear combination of sign vectors.

Abstract: A calculation device that simulates spin states of a finite-temperature Ising model by combining characteristics of a coherent Ising machine and an electronic calculator is provided. The calculation device includes a coherent Ising machine including a sampling unit that samples a spin state, the sampling unit introducing a check spin for extracting a sample in addition to a target spin for solving a target problem for which a solution of the Ising model is to be obtained, and an electronic calculator including a temperature estimation unit that calculates a temperature or an inverse temperature of a sample in a spin state extracted by the sampling unit, and an integration processing unit that integrates a plurality of samples and performs statistical processing.

Abstract: An acquisition section (101) acquires an actual measurement value of a measured change in a specific environment of a data assimilation target. A computation section (102) uses a preliminary initial state and a preliminary value of an unknown parameter that are related to the data assimilation target to perform a numerical computation of a change in the specific environment of the data assimilation target. An update section (103) computes a value of an evaluation function representing errors between the actual measurement value and a value obtained from a result of the numerical computation and corresponding to the actual measurement value, finds an acquisition function from plural combinations of values of the initial state and the unknown parameter combined with the evaluation function value, and updates values of the initial state and the unknown parameter so as to minimize a value of the evaluation function based on a value of the acquisition function.

Abstract: This document relates to solving challenges associated with solving partial differential equations (PDEs) via numerical simulations relating to natural or physical systems. One example obtains input data relating to a physical system and partitions tensors of a neural network across multiple parallel processors. The example distributes the input data across multiple parallel cloud processing resources for numerical simulations involving partial differential equations to produce corresponding output data. The example trains the neural network across the tensors of the multiple parallel processors with the input data and the output data to produce a surrogate model of the partial differential equations. The example can receive subsequent input data and generate corresponding subsequent output data utilizing the surrogate model.

Abstract: A compute express link (CXL) memory device includes a memory device storing data, and a controller configured to read the data from the memory device based on a first command received through a first protocol, select a calculation engine based on a second command received through a second protocol different from the first protocol, and control the calculation engine to perform a calculation on the read data.

Abstract: A computer implemented method for detecting the existence of a condition indicated by a signature vector sequence of events in an input vector sequence of events, each of the signature and input vector sequences being constituted by an ordered sequence of vectors, can include converting the signature vector sequence into an signature ordered numerical sequence in which each vector in the signature vector sequence is converted to a number indicative of a magnitude of the vector such that the signature numerical sequence is a sequence of magnitudes in the order of the signature vector sequence; converting the input vector sequence into an input ordered numerical sequence in which each vector in the input vector sequence is converted to a number indicative of a magnitude of the vector such that the input numerical sequence is a sequence of magnitudes in the order of the input vector sequence; and determining a degree of similarity of the signature numerical sequence and the input numerical sequence to detect the

Abstract: A method, system, and compute program product are configured to: receive a dataset comprising a multivariate time series that includes plural channels; generate an original forecast of the multivariate time series using a channel-independent backbone and a prediction head; and generate a revised forecast of the multivariate time series using a cross-channel reconciliation head with the original forecast, wherein the cross-channel reconciliation head generates the revised forecast based on correlations between the channels of the multivariate time series.

Abstract: One example method includes receiving a quadratic unconstrained binary optimization (QUBO) matrix associated with a QUBO problem, determining a classical computing infrastructure on which to execute a simulated first process to solve the QUBO problem, causing execution of the simulated first process to solve the QUBO problem, wherein the execution of the simulated first process is performed on the classical computing infrastructure, and generates a result, and causing execution of an actual, or simulated, second process to solve the QUBO problem, wherein execution of the simulated second process is performed on a quantum computing infrastructure, and the result is used as a starting point for performance of the second process.

Abstract: Cutting matrix configurations in order to execute complex quantum jobs is disclosed. A combinatorial optimization problem may be received and converted into a matrix configuration that can be executed by a quantum computing system. The matrix configuration is cut into k smaller matrix configurations that can be executed on different quantum computing systems or on the same quantum computing system. The smaller matrix configurations are executed and the solutions of the smaller matrix configurations are combined to generate a solution to the original matrix configuration.

Abstract: Systems and methods for predicting yearly performance improvement rates for nearly all definable technologies for the first time are provided. In one embodiment, a correspondence of all patents within the U.S. patent system to a set of technology domains is created. From the identified patent sets, the invention may calculate average centrality of the patents in each domain to predict improvement rates, following a patent network-based methodology. Also disclosed is a system to intake a user technology search query and match user intent with the technology domain as well as the corresponding improvement rate.

Abstract: In various examples, sets of testing data may be selected and applied to an MLM such that differences in performance of the MLM in the testing between the sets indicates and may be used to determine whether and/or an extent by which the MLM is trained to rely on artifacts. Training data for the MLM may be generated using a first value of a parameter that defines a value of a characteristic of the training data. For testing, first testing data may be selected that corresponds to a second value of the parameter that shifts the value in a first direction and second testing data may be selected that corresponds to a third value of the parameter that shifts the value in a second direction (e.g., opposite the first direction). Various possible actions may be taken based on results of analyzing the differences in performance.

Abstract: In some aspects, techniques for creating representative and informative training datasets for the training of machine-learning models are provided. For example, a risk assessment system can receive a risk assessment query for a target entity. The risk assessment system can compute an output risk indicator for the target entity by applying a machine learning model to values of informative attributes associated with the target entity. The machine learning model may be trained using training samples selected from a representative and informative (RAI) dataset. The RAI dataset can be created by determining the informative attributes based on attributes used by a set of models and further extracting representative data records from an initial training dataset based on the determined informative attributes. The risk assessment system can transmit a responsive message including the output risk indicator for use in controlling access of the target entity to an interactive computing environment.

Abstract: The disclosure relates to methods and systems of generating geodata match classifications based on an ensemble of weak supervision-based match labels and geo-similarity models. The system may generate at least two match classifiers that are ensembled together. The match classifiers may include a location name classifier and a geo-similarity classifier. The location name classifier may generate a match classification that is based on a similarity of characters of the location names. The geo-similarity classifier may generate a match classification that is based on a proximity of the geolocations defined by the location names. The match classifications may be aggregated together to generate a geodata match classification.

Abstract: A learning device includes a learning means for learning a discriminative model that discriminates a class to which second data belongs, the second data being data corresponding to an unknown object, by using first training data that includes a group including a plurality of pieces of first data corresponding to the same object, and a first data label with respect to the group. The learning means computes a discrimination score with respect to the first data by using the discriminative model, and learns the discriminative model by using a loss weighted by a weight that depends on a relative height of the discrimination score in the group.