Abstract: Contextual data may be generated from assets in asset portfolios using metadata enrichment services. A recommendation engine may generate a set of recommended assets for presentation in a content stream based on the contextual data. Brand safety may be implemented using a brand safety policy that uses the contextual data as indicators of potentially offensive content. Advertisements included in the content stream may also be targeted based on the contextual data.

Abstract: A search request for content can be initiated by a user, and the content that is relevant to the search request can be identified and presented in a manner that indicates the content's trustworthiness or relevancy. The identified content can be ranked based on the number of times the content has been referenced as well as by source that referenced the content. The relevant identified content can then be displayed in an ordered list that is ordered based on the number of times the content has been referenced. In some cases, the order may be modified by the authority of the source.

Abstract: Example embodiments that analyze images to characterize aspects of the images rely on a same neural network to characterize multiple aspects in parallel. Because additional neural networks are not required for additional aspects, such an approach scales with increased aspects.

Abstract: The disclosed technologies are capable of selectively using data structure operations for path query evaluation. One technique involves reading a query that traverses at least two nodes and at least one edge of a graph in a graph database; compiling the query into a set of variables and a set of constraints, where the set of variables and the set of constraints correspond to the two nodes and the one edge of the graph; creating an in-memory data structure that comprises a table; using the set of variables and the set of constraints to determine an operation that is performable using the in-memory data structure; checking for an existence of a condition relating to the in-memory data structure or the operation; skipping the operation if the condition exists or executing the operation if the condition does not exist; and storing a set of intermediate query results in the table.

Abstract: Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform acts of: providing a search engine that includes, or communicates with, a recall personalization model configured to generate personalized recall sets of search results for users; receiving, at the search engine, a search query submitted by a user; generating, using the recall personalization module, a feature vector for the user that includes contextual features associated with the user; generating, using the recall personalization model, a simulated narrowing query that includes the search query submitted by the user and the feature vector; generating, using the search engine, a recall set of search results based, at least in part, on the simulated narrowing query. Other embodiments are disclosed herein.

Abstract: A quality assessment system models product or service quality based on contextual relevance. A neural network generates a contextual relevance model that differentiates descriptive characteristics based on a modeled relevance of each descriptive characteristic to the product or service. The system filters reviews based on the contextual relevance model to retain filtered reviews that reference any of the first set of descriptive characteristics. The system generates theme clusters with an encoder. Each theme cluster groups a different subset of the filtered reviews based on an amount of semantic similarity between the different subset of reviews and the theme cluster. The system presents an interface with a first visualization and a second visualization. The first visualization graphically represents a sentiment expressed in reviews grouped to the first theme cluster, and the second visualization graphically represents a sentiment expressed in reviews groups to the second theme cluster.

Abstract: According to an embodiment, a relay apparatus corresponds to a node included in nodes of a wireless network system. The relay apparatus includes wireless communication circuitry, and processor circuitry. The wireless communication circuitry includes ports capable of transmitting a transmission frame or receiving a reception frame via a wireless communication line respectively. The processor circuitry is configured to select a first port connected to a first apparatus from the ports, if the wireless communication circuitry receives a first frame and if the first frame is to be relayed to the first apparatus via a wireless communication line. The wireless communication circuitry is further configured to transmit the first frame through the first port.

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy.

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy.

Abstract: Methods, systems and computer-readable media enable various techniques related to semantic navigation. One aspect is a technique for displaying semantically derived facets in the search engine interface. Each of the facets comprises faceted search results. Each of the faceted search results is displayed in association with user interface elements for including or excluding the faceted search result as additional search terms to subsequently refine the search query. Another aspect automatically infers new metadata from the content and from existing metadata and then automatically annotates the content with the new metadata to improve recall and navigation. Another aspect identifies semantic annotations by determining semantic connections between the semantic annotations and then dynamically generating a topic page based on the semantic connections.

Abstract: Systems and methods for generating a playlist on a device for a user of a music service are provided. In embodiments, a method includes initiating an online mode of the music service, where the device is connected to a music service server, and storing, in a cache memory of the device, at least one media object accessed by the user using the music service. The method further includes identifying the at least one media object as satisfying a pro-designated condition, and in response, automatically adding the at least one media object to a playlist stored in the cache memory. The method also includes receiving an indication, via a user interface of the device, to initiate an offline mode of the music service, where the device is not connected to the music service server, and upon initialing the offline mode, enabling playback of the playlist from the cache memory.

Abstract: Embodiments described herein provide methods and systems for summarizing multiple documents. A system receives a plurality of documents and generates embeddings of the sentences from the plurality of documents. The embedded sentences are clustered in a representation space. Sentences from a reference summary are embedded and aligned with the closest cluster. Sentences from each cluster are summarized with the aligned reference sentences as a target. A loss is computed based on the summarized sentences and the aligned references, and the natural language processing model is updated based on the loss. Sentences may be masked from being used in the summarization by identifying sentences that are contradicted by other sentences within the plurality of documents.

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy.

Abstract: Systems and methods for developing a classification model for classifying medical reports, such as radiology reports. One method includes selecting, from a corpus of reports, a training set and a testing set, assigning labels of a modality and an anatomical focus to the reports in both sets, and extracting a sparse representation matrix for each set based on features in the training set. The method also includes learning, with one or more electronic processors, a correlation between the features of the training set and the corresponding labels using a machine learning classifier, thereby building a classification model and testing the classification model on the reports in the testing set for accuracy using the sparse representation matrix of the testing set. The method further includes predicting, with the classification model, labels of an anatomical focus and a modality for remaining reports in the corpus not included in the sets.

Abstract: Generally, the present disclosure relates to solving issues with inefficient management of data within an organization by building a central repository of events within the data ecosystem. The data event management system, disclosed herein, offers a platform agnostic ledger of the data ecosystem that enables data transparency and identifies patterns/bottlenecks to properly adjust to meeting the needs of the data consumer. With this information, data driven decisions can be made to prioritize workload and ensure funding is supporting the consumer prioritized assets.

Abstract: To obtain an image desired by a user by performing an intuitive operation. An image display system includes a first display control unit that displays a reference image on a display surface, and a second display control unit that selectably displays a plurality of candidate images each having image information different from the reference image in the periphery of a display area of the reference image on the display surface, wherein each image data of the plurality of candidate images is included in a search area on a predetermined space determined based on the image data of the reference image.

Abstract: Method, apparatus and computer program product for generating a channel calendar in a group-based communication system are described herein. The apparatus is configured to at least generate a channel calendar identifier associated with a group-based communication channel in the group-based communication system, receive a calendar sharing request associated with the channel calendar identifier from a client device, retrieve the calendar object and calendar object metadata set based on the channel calendar identifier, and cause rendering of the calendar object for display via a user interface of the client device.

Abstract: Disclosed is a method for controlling an intelligent device, the method including: acquiring source data for generating a profile of a user; generating a cluster related to recommendation of a service by using the source data; generating the profile of the user using the cluster; and transmitting the profile of the user to a server, wherein the server recommends the service to the user based on the profile of the user. Accordingly, the present disclosure may automatically recommend an appropriate service to a user. The intelligent device of the present disclosure can be associated with artificial intelligence modules, drones (unmanned aerial vehicles (UAVs)), robots, augmented reality (AR) devices, virtual reality (VR) devices, devices related to 5G service, etc.

Abstract: Systems implement a software platform that creates dynamic user device pools that facilitate in-person engagement between and among platform users as well as the participation in local events and activities. The system includes a profile server that communicates with one or more user computing devices, such as smartphones. The provider server determines a pool area that can be predefined or determined based on a geometric area surrounding a user computing device. The provider server determines the user devices within the pool area and obtains profile data associated with the user devices where the profile data represents information personal to the platform users. The provider server applies user relationship information and privacy settings to filter the profile data before transmitting the profile data to user computing devices for display. Platform users communicate with other users in the pool area according to permissions settings that safely limit the types of permissible communications.

Abstract: In one embodiment, a method includes receiving a user request from a client system associated with a user, generating a response to the user request which references one or more entities, generating a personalized recommendation based on the user request and the response, wherein the personalized recommendation references one or more of the entities of the response, and sending instructions for presenting the response and the personalized recommendation to the client system.

Abstract: The disclosed systems and methods are directed to detecting sensitive data on a computing device. This includes matching predetermined keywords in input data, to determine data in vicinities of matched keywords in the input data in which sensitive data is likely to be found, and matching predefined patterns associated with sensitive data to the data in vicinities of matched keywords to detect sensitive data. Matching the predetermined keywords occurs prior to matching the predefined patterns, and the data in vicinities of matched keywords is substantially shorter than the input data.

Abstract: In an example embodiment, a self-supervised learning task is used for training commonsense-aware representations in a minimally supervised fashion and a pair level mutual-exclusive loss is used to enforce commonsense knowledge during representation learning. This helps to exploit the mutual-exclusive nature of the training samples of commonsense reasoning corpora. Given two pieces of input where the only difference between them are trigger pieces of data, it may be postulated that the pairwise pronoun disambiguation is mutually exclusive. This idea is formulated using a contrastive loss and then this is used to update the language model.

Abstract: Methods and systems are described that detail the rapid validation of scientific products and scientist consumer information derived from multiple science/research based sources, such as peer-reviewed scientific publications and online resources, for example, as well as any other available tasked forum or venue. The refined information will facilitate unique product evaluation opportunities via an accessible website/forum, trade shows, etc., and provide previously unknowable sales leads, market intelligence, and targeted advertising opportunities for product manufacturers.

Abstract: A computer-implemented, cloud-based productivity tool incorporates a user interface configured to: (a) prompt the user to input a set of unstructured text strings; (b) allow the user to categorize the set of unstructured text strings in accordance with a set of categories provided by the user; (c) sequence the categorized text strings; (d) summarize the categorized text strings; and (e) produce an output based on the foregoing. In this way—by using a form of “reverse mind-mapping”—a user can quickly enter his or her thoughts in a free-form way and then subsequently organize that information in accordance with emergent categories that become apparent to the user through the process provided by the tool. A social aspect is provided to allow shared thinking between subscribers through text communication. In addition, a journaling module is provided for extracting insights from writing that accumulates over time.

Abstract: The present disclosure generally relates to a feedback processing service that can receive customer input, as customer feedback corresponds to a service context. The feedback processing service aggregates semantically similar feedback as a cluster. Then, the feedback processing service can prioritize each of the clusters by ranking each of the clusters.

Abstract: Customized recommendations of multimedia content streams are provided herein. A method can include facilitating, by a network device of a wireless network and comprising a processor, an output of a recommendation of a first multimedia content stream at a communications device based on a monitored interaction determined to have been performed by an identified entity at the communications device. The method can also include facilitating, by the network device, a first transmission of the first multimedia content stream to the communications device in response to an acceptance of the recommendation of the first multimedia content stream received from the communications device. The first multimedia content stream can be renderable for consumption at the communications device.

Abstract: Creating and executing flow plans by performing at least the following: obtaining a run-time flow plan that comprises a trigger, a first operation, and a second operation, wherein the first operation precedes the second operation within the run-time flow plan and one or more input values of the second operation are linked to the first operation, determining whether one or more conditions of the trigger are met, execute the first operation based at least on the determination that the one or more conditions of the trigger are met, monitoring whether the second operation is ready for execution based at least on a determination that the one or more input values of a second action operation are ready, and executing the second action operation when the second action operation has been identified as ready for execution.

Abstract: Techniques are described for fast approximate conditional sampling by randomly sampling a dataset and then performing a nearest neighbor search on the pre-sampled dataset to reduce the data over which the nearest neighbor search must be performed and, according to an embodiment, to effectively reduce the number of nearest neighbors that are to be found within the random sample. Furthermore, KD-Tree-based stratified sampling is used to generate a representative sample of a dataset. KD-Tree-based stratified sampling may be used to identify the random sample for fast approximate conditional sampling, which reduces variance in the resulting data sample. As such, using KD-Tree-based stratified sampling to generate the random sample for fast approximate conditional sampling ensures that any nearest neighbor selected, for a target data instance, from the random sample is likely to be among the nearest neighbors of the target data instance within the unsampled dataset.

Abstract: Systems and methods of the present disclosure may read source data corresponding to a source variable and apply a transformation to the source variable to generate an output variable. Accordingly, one such method comprises receiving a request to generate an output variable by applying one or more data transformations to a source variable; before applying an actual data transformation to the source variable, checking a database of prior data transformations that have been applied to source variables to generate existing output variables for a match with the requested one or more data transformations to the source variable; determining that a match exists in the database, wherein the match is associated with an existing output variable; and/or responding to the request by providing the existing output variable associated with the match that exists in the database of prior data transformations.

Abstract: An embodiment calculates a point of view (POV) value for a subportion of an item of media content, including evaluating user inputs related to an automatically detected POV of the subportion of the item. The embodiment also calculates a non-factual cost for the subportion of the item indicative of an amount of the subportion of the item that lacks support in a fact-based corpora. The embodiment also performs a cascaded summarization process comprising generating a summary of the item using the subportion of the item and a summarization technique, analyzing the summary using a fact-checking algorithm to determine whether the summary satisfies a factual score threshold, and performing a next iteration of the cascaded summarization process while the summary fails to satisfy the factual score threshold. The embodiment also communicates a summary satisfying the factual score threshold to a user device.

Abstract: A method for matching supplied organizational data with trade directory information includes: receiving a data file including a plurality of organizational entries, each entry including an entity name and a geographic location; normalizing the entity name in each of the organizational entries; identifying a plurality of matching entries for each organizational entries, each matching entry including a matching name and location, and where each matching entry is identified based on a first correspondence between the matching name and the entity name and a second correspondence between the matching location and the geographic location; determining a confidence level for each of the organizational entries based on the correspondence between the first correspondence and the second correspondence for at least one of the identified matching entries; and transmitting the plurality of matching entries and determined confidence level for each of the plurality of organizational entries.

Abstract: Organizing and/or aligning fragments of text that are included in a set of physical and/or digital documents so that the arrangement of the text fragments is in a readily understandable and meaningful format for a given reader. This organization and/or alignment uses a relation model of the various text fragments to correlate a meaning between and amongst the various text fragments to ultimately determine the final alignment and/or arrangement of those text fragments.

Abstract: Described herein is a computer implemented method for inserting a new item to a tree-structured dataset. The method comprises: calculating a new item data summary; generating a new leaf node for storing the new item and the new item data summary; adding the new leaf node to the dataset; and recalculating data summaries for all internal dataset nodes in an update path starting at a parent of the new leaf node and ending at a root node of the dataset. The data summary for a given internal node in the update path is calculated based on data summaries for each of the given internal node's children nodes.

Abstract: The processor(s) may be configured to electronically process a computer readable set of user data records to generate media consumption data. The processor(s) may be configured to electronically process the computer readable set of user data records to generate social media interaction data. In some implementations, the processor(s) may be configured to electronically process the computer readable set of user data records to generate gaming interaction data. In yet some implementations, the processor(s) may be configured to electronically process the computer readable set of user data records to generate a computer readable virtual wallet with kinetic avatar data associated with at least one of the user data records. The processor(s) may be configured to electronically process the virtual wallet including a digital virtual credit card attribute data. The processor(s) may be configured to be housed on a substrate body.

Abstract: A hierarchical representation of an input data set comprising similarity scores for respective entity pairs is generated iteratively. In a particular iteration, clusters are obtained from a subset of the iteration's input entity pairs which satisfy a similarity criterion, and then spanning trees are generated for at least some of the clusters. An indication of at least a representative pair of one or more of the clusters is added to the hierarchical representation in the iteration. The hierarchical representation is used to respond to clustering requests.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for re-ranking a collection of documents according to a first metric and subject to a constraint on a function of one or more second metrics. One of the methods includes: obtaining, for each document in the first collection of documents, a respective first metric value corresponding to the first metric and respective one or more second metric values corresponding to the one or more second metrics; re-ranking the first collection of documents, comprising: determining the constraint on the function of one or more second metrics by computing a first threshold value using a variable threshold function that takes as input second metric values for the documents in the first collection of documents; and determining the re-ranking for the first collection of documents by solving a constrained optimization for the first metric constrained by the first threshold value.

Abstract: Apparatuses and methods for training users of ambulatory medical devices. The methods relate to improving user interactions with the touchscreens of devices. In one embodiment there is an operating mode that records all user interactions along with various device parameters and allows the clinician to review the patient's performance for the initial use period. Automated analysis software may be employed to analyze the data generated by the device. The results of the analysis may be used by the clinician to improve the patient and device interaction.

Abstract: There is provided an information processing device including an analysis unit configured to analyze a character string indicating contents of utterance obtained as a result of speech recognition, and a display control unit configured to display the character string indicating the contents of the utterance and an analysis result on a display screen.

Abstract: In some implementations, an analytics platform may process a data lineage configuration to identify one or more data sources storing one or more data sets associated with a data analytics use case and identify one or more data relations associated with the one or more data sets. The analytics platform may parse a source code template defining one or more functions to build a data mart and a data pipeline that enables the data analytics use case. The source code template may include one or more tokens to specify elements that define the one or more data sources, data sets, and data relations identified in the data lineage configuration. The analytics platform may automatically generate source code that is executable to build a data mart and a data pipeline to enable the data analytics use case based on the data lineage configuration and the source code template.

Abstract: Provided is an information providing device capable of outputting provided information reflecting user's needs even if the user conducts information search on the basis of a query which is on the basis of an unspecific desire. The information providing device is configured with a computer system and outputs provided information in response to input of a query. The information providing device includes storage means that stores the query being composed of at least one property and provided information, rating means that rates the provided information on the basis of the query, and output means that outputs the provided information rated by the rating means in response to the query inputted.

Abstract: Systems and methods are provided for automatic resource replacement in web pages. In one embodiment, a method comprises determining a link to a first resource in a web page is broken, automatically selecting a second resource from a resource database to replace the first resource in the web page, the second resource of a same type as the first resource, updating the web page with a link to the second resource in place of the link to the first resource, and displaying the updated web page to a user. In this way, resource errors may be mitigated in real-time and user experience of accessing web pages is improved.

Abstract: Techniques in this disclosure may provide a user interface that concurrently displays multiple panels which provide visualization of emergency call data of a law enforcement agency. The user interface can provide a high-level overview of emergency calls in a geographical area. Each panel in the user interface can provide visualization of the emergency calls and/or statistics relating to the calls. A user can customize which panels to include in the user interface and/or customize setting for each panel. The user may apply various types of filters to the data displayed in the user interface, and the panels can update the visualizations according to the filters. The user interface can also provide the ability to show data at various levels of detail within the same user interface or panel. The techniques in the disclosure can provide a convenient, digestible overview of tactical and/or strategic data in a single user interface.

Abstract: Embodiments of the present invention provide a method, computer program-product, and system for generating predicate-weighted histograms in a database management system. Further, the methods, computer program-products and systems in accordance with the present invention generate histograms that are biased towards the predicate literals of the queries that are submitted to the database management system. The resulting histograms will improve query performance by generating histograms with greater resolution near predicate literals that represent the queries submitted to the database management system.

Abstract: An approach is disclosed for mapping an intent of a query. The approach receives query data for one or more queries. The approach determines product type data of the one or more queries. The approach determines attribute data for the determined product type data. The attribute data includes sequence attribute data and sequence value data. The approach retrieves catalog data for one or more items that include product type data that matches the determined product type data. The catalog data includes catalog attribute data and catalog value data. The approach determines engagement data for the one or more received queries. The engagement data includes a number of times customers interacted with a retrieved item for a respective query. The approach determines cumulative engagement data based on the retrieved catalog data and the determined engagement data. The approach generates a probability and an evaluation score based on the cumulative engagement data.

Abstract: Embodiments of the present application disclose a content generation method and apparatus. The method includes: acquiring product description information; selecting, by using a deep neural network model component, a content phrase matched with the product description information, wherein the deep neural network model component is obtained by training according to a plurality of pieces of historical product description information and historical content of the historical product description information; and generating content corresponding to the product description information based on the selected content phrase.

Abstract: Disclosed embodiments relate to systems and methods for performing privileged computing actions. Techniques include receiving an indication of a request by an entity to perform a privileged computing operation; determining whether the entity is associated with a privilege needed to perform the privileged computing operation; accessing historical data indicating at least one series of events leading to the entity being associated with the privilege; determining whether the entity is properly associated with the privilege; causing the privileged computing operation to be performed by a secured processing element; and causing a result of the privileged computing operation performed by the secured processing element to be provided to the entity.

Abstract: A reader platform system generates a display of a document that is formed based on a tree structure from a structured document template with an idea type attribute and content of a supporting idea type attribute that is based on the idea type attribute. The idea type attribute and the content of the supporting idea type attribute are validated based on taxonomy rules for the structured document template. The reader platform system receives a reader feedback of the document via a feedback user interface of the display. The profile of a reader of the document and the profile of an author of the document are updated based on the reader feedback. The feed of another reader related to the reader of the document is updated based on the updated profile of the reader of the document.

Abstract: A dynamic browser bookmark system configured to present a user's bookmarks in a hierarchical manner based on webpage contextual data is disclosed. The system may aggregate browser bookmarks associated with a user device. The system may determine a first webpage being viewed on the user device. The system may dynamically filter the browser bookmarks to determine a first plurality of browser bookmarks associated with the first webpage. The system may generate and transmit a graphical user interface including a hierarchical view of the browser bookmarks. The system may determine a second webpage being viewed on the user device and dynamically filter the aggregated browser bookmarks to determine a second plurality of browser bookmarks relevant to the second webpage, which may include a browser bookmark of the first plurality of browser bookmarks, and generate a second graphical user interface including a hierarchical view of the second plurality of browser bookmarks.

Abstract: The present invention provides a similarity search engine for a digital visual object (i.e., a digital image that represents a design, graphics, logo, symbols, words, or any combination thereof). The similarity search engine is based on a method that consists of conducting several independent search queries, thus each query examining a different aspect of similarity.

Abstract: A method for ranking relevance of documents includes using a set of queries, searching a corpus of documents for a set of candidate documents with information relevant to the set of queries. The method further includes ranking the set of candidate documents by a deep learning processing system according to relevance to respective ones of the set of queries. The method additionally includes responsive to user input, revising the ranked set of candidate documents to produce a revised ranked set of candidate documents. The method further includes using the revised ranked set of candidate documents to retrain the deep learning processing system. The method still further includes performing a categorization of the set of candidate documents by the retrained deep learning processing system.