Abstract: Technologies for on-device experimentation include embodiments that receive a request to provide digital content for display in a slot of a user interface display screen of a first device. By a secure execution environment, an identifier received with the request is anonymized. The request is determined to be associated with a content distribution test. The anonymized identifier is assigned to a test group associated with the content distribution test. The secure execution environment receives user interface event data generated by the first device in response to the content distribution test. The received user interface event data is attributed to the test group. An association of the user interface event data with the test group and the content distribution test is provided by the secure execution environment to a second device different than the first device while the identifier and the anonymized identifier are not provided to the second device.

Abstract: An auto-styler device may provide a style-based or outfit-driven shopping experience. The auto-styler device may select a style definition that defines a style-conforming outfit based on rules that apply to a combination of a first item type and a second item type of the style-conforming outfit, and that defines a customized presentation for the style-conforming outfit. The auto-styler device may generate a style-conforming outfit with a first item of the first item type and a second item of the second item type in response to a collective style produced by the combination satisfying the rules. The auto-styler device may position and size a first image of the first item relative to a second image of the second item in a single interface based on the specified the customized presentation of the style definition, and may present or publish the resulting single interface on a merchant site.

Abstract: To facilitate efficient entity resolution, systems and methods include a first dataset is received from a first database associated with a first entity and a second dataset is received from a second database associated with a second entity. A geo-grid is mapped to a geographical area covering the first entity data records and the second entity data records. A grid matching area in the geo-grid is generated for each first entity data record based on latitude data and longitude data of each first entity data record. Candidate matching records are determined from the second entity data records based on respective grid matching areas. Actual matching records are determined from the candidate matching records based on a threshold for a trigram similarity between each candidate matching record and the respective first entity data record. The actual matching records are associated with the respective first entity data record in the first database.

Abstract: Embodiments include a computer-implemented method for storing comments received at a collaboration page of a content collaboration system in a hierarchical relationship that corresponds to the comments position in a comments field. The method can include receiving a comment at a user interface displaying the collaboration page and creating a thread identification string for the comment having a set of string portions. The method can also include obtaining a comment index file comprising an ordered list of previously-entered comments having respective thread identification strings, adding the comment and thread identification string to the ordered list of previously-entered comments at a location indicated by the thread identification string, designating a pagination of the ordered list of previously-entered comments in the comment index file in accordance with a fixed number of simultaneously viewable lines, and displaying a subset of the ordered list of previously-entered comments in the comment field.

Abstract: A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with “ground truth” for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Abstract: A system, including a memory storing instructions and a processor configured to execute the instructions is provided. The instructions executed by the processor cause the system to retrieve a diagnostic information for a patient, to retrieve a physician information for a physician in charge of the patient, and to determine an anticipated medication prescription for the patient based on the diagnostic information, the physician information, and a medication prescribing pattern stored in the memory. A method for using the system and a non-transitory, computer readable medium including the instructions are also provided.

Abstract: Technologies relating to model interpretation are described herein. A text classifier is provided with input text and assigns a class to the input text from amongst several possible classes. Based upon the class assigned to the input text by the text classifier, a class profile of a centroidal classifier is selected, where the class profile is constructed based upon numerous input texts to which the text classifier has previously assigned the class. Based upon the selection of the class profile, information that is indicative of operation of the text classifier with respect to the text classifier assigning the class to the input text is output, where the information includes an exemplar text sequence.

Abstract: A data classification method and a device thereof for hierarchical multi-label classification are proposed. The data classification device includes an encoder configured to vectorize data, a decoder configured to receive the vectorized data from the encoder, and classification data stored in the decoder, wherein the decoder starts classification of the vectorized data on the basis of a first node in the classification data, allocates a plurality of nodes included in the classification data to the vectorized data, and forms partial layers based on the allocated nodes with respect to the vectorized data, and when terminal nodes or temporary nodes are allocated among the nodes, the decoder ends the allocation to the partial layers to which the terminal nodes or the temporary nodes are allocated and allocates final partial layers formed on the basis of the partial layers to the data.

Abstract: A method for communicating a playback order for a plurality of media content items to a user device operating in an online mode, the method performed at a server system and comprising receiving an indication that the user device will enter an offline mode, generating a playback order for the plurality of media content items, and transmitting the generated playback order to the user device before the user device enters the offline mode.

Abstract: One or more computer processors receive a pattern string comprising a pointer to a pattern search point and a string comprising a pointer to a search point; determine a plausible search point within the string for the pattern character; execute a search string (SRST) instruction with the determined search point and the pattern character; compute a pattern offset comprising a difference between a first occurrence of the determined pattern character located within the pattern string but not in the correct position and a current pattern character position; execute a subsequent SRST instruction with the current pattern character position and the computed offset set as a new search point; and responsive to identifying all pattern characters in the correct position within the string based on the executed subsequent SRST instruction, transmit memory address associated with a substring comprised of the string and corresponding pattern characters in the correct position.

Abstract: The application discloses a device and a method for picking up top k values from N values. The method comprises: A) controlling a buffer to receive values into a data pool until the number of values in the data pool reaches the predetermined memory size; B) dividing the values in the data pool into a first portion and a second portion; C) discarding the values in the second portion and controlling the buffer to continue to receive values into the data pool; D) repeating steps B to C until the buffer has received all the N values; E) dividing the values in the data pool into the first portion and the second portion until the number of values in the first portion reaches k; and F) controlling the buffer to output the k values in the first portion as the top k values.

Abstract: Described are systems and methods directed to a multi-channel communication system that enables real-time or near real-time communication between two or more individuals, such as a patient and a clinician in which at least one of the individuals uses one device for audio/video communication and another device for exchange of data, such as annotated images. For example, a patient may utilize a patient device to communicate with a clinician via a first clinician device. In addition to the first clinician device, the clinician may also use a second clinician device to annotate and share visual data, such as templates or patient images, with the patient, without disrupting the audio/video communication between the patient device and the first clinician device.

Abstract: Systems, methods, and non-transitory computer readable media are provided for mapping data based on an ontology of a platform. A data set may be obtained. Data within the data set may be for use by an operation platform based on an operation ontology. The operation ontology may define an operation data structure for the operation platform. The data may be shaped based on a target ontology. The target ontology may define a target data structure for a target platform. The data may be shaped such that the data is mapped to the target data structure.

Abstract: A technique for data reconstruction includes determining a plurality of types of data stored in a storage disk; determining an order in which data reconstruction is to be performed on the plurality of types of data based on corresponding likelihoods that the plurality of types of data are expected to be accessed; and performing the data reconstruction on the plurality of types of data in the determined order. Such a technique can improve system performance during data reconstruction and reduce the access delay.

Abstract: In general, embodiments of the present invention provide systems, methods and computer readable media to forecast demand by implementing an online demand prediction framework that includes a hierarchical temporal memory network (HTM) configured to learn temporal patterns representing sequences of states of time-series data collected from a set of one or more data sources representing demand and input to the HTM. In some embodiments, the HTM learns the temporal patterns using a Cortical Learning Algorithm.

Abstract: A method for pre-positioning data based on a user attribute is disclosed. The method includes retrieving audit information that corresponds to a user, the audit information including the user attribute; determining, by using a model, a user characteristic for the user based on the audit information, the user characteristic including a probability value and a geographical location; comparing the user characteristic with a predetermined threshold; identifying raw data from a networked platform, the raw data corresponding to the user; generating a structured data set for the user based on the user characteristic; and persisting the structured data set in an edge computing repository based on a result of the comparing, the edge computing repository corresponding to the geographical location.

Abstract: A facility for managing the displayed documents by an application is described. The facility displays a plurality of tabs each corresponding to a document loaded by the application. The facility receives a document stowing command in connection with a distinguished document to which one of the displayed tabs corresponds. For each of the documents to which a displayed tab corresponds, based on the contents of the document, the facility semantically determines one or more subjects of the distinguished document. At least in part in response to receiving the document stowing command in connection with the distinguished document, for each of two or more subjects determined for the distinguished document, the facility displays a visual indication of the determined subject. The facility receives user input selecting one of the displayed visual indications. The facility hides the tabs corresponding to the documents for which the subject indicated by the selected visual indication was semantically determined.

Abstract: A network service maintains a recently used resource list for a user that includes a list of references (e.g., hyperlinks) to resources (e.g., files) recently accessed by the user. An application, such as a web browser application or a web browser extension, periodically retrieves the recently used resource list for the user from the network service. When the application is utilized to display a text entry field, a user interface (“UI”) can be provided for inserting references to recently used resources identified in the recently used resource list into the text entry field. Through the UI, the user can select a resource identified in the recently used resource list using appropriate user input (e.g., selection of a resource using a mouse cursor, touch, or keyboard input). In response thereto, a reference (e.g., a hyperlink) to a selected resource can be inserted into the text entry field.

Abstract: Systems, methods, computing platforms, and storage media for comparing non-adjacent data subsets are disclosed. Exemplary implementations may: receive an input data set, the input data set including information to be analyzed; generate at least one list of data subsets of the input data set; determine whether at least one data subset of the at least one list of data subsets contains a notable characteristic; identify at least one data pattern in the at least one list of data subsets of the input data set; sort data subsets of the at least one list of data subsets of the input data set; and display, via a user interface, the data subsets of the at least one list of data subsets, based on the sorting.

Abstract: The invention provides a comprehensive system, method and computer program product for data structure clustering based on variation in data attribute performance. In general, the invention is configured to link a sub-set of data structures out of a global set of data structures based on identifying stimulus type attributes that are structured to produce a functional performance parameter of the associated data structure, particularly in the field of hospital data analysis. Typically, the invention is configured to receive a request for construction of a peer data cluster for a first functional performance parameter of the first entity data structure. The invention may then determine at least one second entity data structure of a plurality of entity data structures that is a peer entity data structure, and linking the first entity data structure and the second entity data structure to form the peer data cluster.

Abstract: A system and method for optimizing a display layout of multiple video streams at a sink device is provided. The display layout of multiple streams may be dynamically optimized based on a number of different variables, including characteristics of the sink device, total number of active incoming streams, active audio, and other characteristics of the source material or device. The source of an incoming media stream may contain useful characteristics for optimizing the display layout of multiple media streams. One such characteristic of a source device may include the device type, such as laptop, PC, phone, or tablet. Information may be extracted from each incoming stream in order to predict a source device type from which the incoming media stream originates.

Abstract: To facilitate efficient entity resolution, systems and methods include a first dataset is received from a first database associated with a first entity and a second dataset is received from a second database associated with a second entity. A geo-grid is mapped to a geographical area covering the first entity data records and the second entity data records. A grid matching area in the geo-grid is generated for each first entity data record based on latitude data and longitude data of each first entity data record. Candidate matching records are determined from the second entity data records based on respective grid matching areas. Actual matching records are determined from the candidate matching records based on a threshold for a trigram similarity between each candidate matching record and the respective first entity data record. The actual matching records are associated with the respective first entity data record in the first database.

Abstract: Disclosed herein are an artificial intelligence device including a memory configured to store user interest data, a processor configured to generate a keyword combination including at least one of a time keyword, a place keyword, an object keyword or an application type keyword based on the user interest data, and a display configured to display at least one of a time keyword, a place keyword, an object keyword or an application type keyword included in the keyword combination.

Abstract: A system, including a memory storing instructions and a processor configured to execute the instructions is provided. The instructions executed by the processor cause the system to retrieve a diagnostic information for a patient, to retrieve a physician information for a physician in charge of the patient, and to determine an anticipated medication prescription for the patient based on the diagnostic information, the physician information, and a medication prescribing pattern stored in the memory. A method for using the system and a non-transitory, computer readable medium including the instructions are also provided.

Abstract: A system is configured for automatic recognition of data store architecture and tracking dynamic changes and evolution in data store architecture. The system ef is a complementary system, which can be added onto an existing data store system using the existing interfaces or can be integrated with a data store system. The system comprises three main components that are configured to compose an approximation of the data store architecture. The first of these components is adapted to execute an analysis of the architecture of the data store; the second of the components is adapted to collect and compile statistics from said data store; and the third of the components is adapted to compose an approximation of the architecture of said data store.

Abstract: Techniques are provided for lazy push optimization, allowing for constant time push operations. A d-heap is used as the underlying data structure for indexing values being inserted. The d-heap is vectorized by storing values in a contiguous memory array. Heapify operations are delayed until a retrieve operation occurs, improving insert performance of vectorized d-heaps that use horizontal aggregation SIMD instructions at the cost of slightly lower retrieve performance.

Abstract: A method for communicating a playback order for a plurality of media content items to a user device operating in an online mode, the method performed at a server system and comprising receiving an indication that the user device will enter an offline mode, generating a playback order for the plurality of media content items, and transmitting the generated playback order to the user device before the user device enters the offline mode.

Abstract: A computer-implemented method includes providing a user interface that allows selection of a product attribute and in the alternative allows selection of a previously defined rule. When a product attribute is selected in the user interface, a rule under construction that describes what products are to appear on a retail webpage is altered based on the selected product attribute. When a previously-defined rule is selected in the user interface, the rule under construction is modified so that the rule under construction uses the selected rule.

Abstract: Provide is a visualization system that enables generation of a “dashboard” of individual visualizations. In further embodiments, the system enables users to quickly and easily generate these visualizations and integrate complex filters, queries, aggregations, etc., with simple UI input. The visualizations can be provided as a service that requests information from an underlying database. The database itself may also be hosted as a service, permitting granular and native database functions layered with the visualization architecture. The system can support additional functionality and access management to generate visualizations that can be shared with other users and/or integrated into websites, blogs, etc. The system can handle the complex logic, data interactions, dynamic data transformation, dynamic authorization, etc., needed to manage data rules (e.g., access rules layered over database permission based control, summarization/aggregation requirements, etc.

Abstract: The present disclosure provides a method for ordering comments, and apparatus, device and computer storage medium thereof, which relates to the technical field of 5 big data. A specific implementation solution is as follows: obtaining comments data, ordering the comments based on basic attributes of the comment to obtain a first ordering result, and establishing an index for the first ordering result; in response to obtaining an event of the user triggering display of the comments, obtaining the first ordering result through the index, and performing ordering adjustment on the first 10 ordering result based on the user's feature data to obtain a second ordering result which serves as a basis for displaying the comments to the user. The present disclosure can meet the user's personalized needs for comment ordering.

Abstract: Disclosed in a fast, compact, efficient, hardware agnostic, stateful data store that can universally store and track millions of stateful data objects, such as networking routing data (or any stateful data) in files for use in, for example, embedded applications. Stateful data objects can be tracked by a modified tree for updating and insert data objects into the stateful data store. The stateful data object can also be allocated additional space in memory for potential future data updates.

Abstract: Techniques for implementing an improved user interface for efficiently browsing through and comparing digital content items are disclosed. In some embodiments, a method comprises: receiving a request for one or more additional graphical widgets indicating at least one of a plurality of graphical widgets selected by a user for concurrent display with the additional graphical widget(s), the request also indicating at least one attribute value for each one of the at least one of the plurality of graphical widgets, with at least one attribute value having been selected by the user for use in determining the additional graphical widget(s). The additional graphical widget(s) may be determined from amongst a plurality of additional graphical widgets based on the attribute value(s) indicated by the request, and then displayed concurrently on the computing device along with the graphical widget(s) selected by the user for concurrent display with the additional graphical widget(s).

Abstract: Techniques are presented herein for efficient query processing and data change propagation at a secondary database system. The techniques involve determining execution costs for executing a query at a primary DBMS and for executing the query at an offload DBMS. The cost for executing the query at the offload DBMS includes the cost of propagating changes to database objects required by the query to the offload DBMS. Based on the execution cost, the query is sent to either the primary DBMS or the offload DBMS.

Abstract: A computer joining an online chat service, based on unstructured text input, can be matched automatically under computer control to one of multiple different online chat conversations using a trained transformer-based machine learning model, training techniques, and similarity assessment techniques. Computer analysis in this manner improves the likelihood that the unstructured text input results in assigning the computer to a relevant chat conversation. Additionally, or alternatively, a dense passage retrieval machine learning model having a first encoder for resources and a second encoder for messages can automatically match relevant resources to computers or sessions based on analysis of a series of messages of an online chat conversation. In either approach, continuous re-training is supported based on feedback from a moderator computer and/or user computers.

Abstract: To facilitate efficient entity resolution, systems and methods include a first dataset is received from a first database associated with a first entity and a second dataset is received from a second database associated with a second entity. A geo-grid is mapped to a geographical area covering the first entity data records and the second entity data records. A grid matching area in the geo-grid is generated for each first entity data record based on latitude data and longitude data of each first entity data record. Candidate matching records are determined from the second entity data records based on respective grid matching areas. Actual matching records are determined from the candidate matching records based on a threshold for a trigram similarity between each candidate matching record and the respective first entity data record. The actual matching records are associated with the respective first entity data record in the first database.

Abstract: An evolving system of computers linked into a neural network continuously scans and gathers information from, understands, and interacts with, an environment; and a client computer program interactively executes software instructions using a subject matter data warehouse to transform input into a search pattern. The evolving system server supercomputer program executes multivariant big data indexing to cherry pick the optimal input and output using page, site and supersite probabilities. The client computer program analyzes the optimal output given a search pattern in order to interact and engage scripted communication with the end user.

Abstract: A Codex system of computers linked into a neural network continuously scans and gathers information from, understands, and interacts with, an environment, an optimizer software executing software instructions based on rules of grammar and semantics searches a encyclopedia of human knowledge to transform input into a search pattern. Then the Codex monetizes and commercializes each transformed input and corresponding optimal output. An artificial intelligence interaction software, hereinafter referred to as virtual maestro, uses the search pattern and optimal output in order to interact and engage scripted communication with the end user.

Abstract: The present invention provides a method for populating and searching a drug informatics database that includes receiving unprocessed data associated with a chemical compound from one or more data sources. The unprocessed data is parsed into a plurality of data objects based on a categorization associated with each of the data objects. Additional information, such as explanatory notes, is identified and associated with at least one of the data objects. The data objects are stored in entries within a data structure, where the data structure is searchable based on one or more of the data objects. A query for data associated with a chemical compound is received at a drug informatics database. The drug informatics database is then searched for data associated with the chemical compound and the search results are provided to a user.

Abstract: Disclosed herein are system, method, and computer program product embodiments for categorizing customer complaints on social media using a model trained on customer voice calls or chats with agents. Additionally, users interested in monitoring regulatory compliance issues based on customer complaints can receive notifications regarding complaints that are linked to regulatory topic areas, without the need to manually scan vast numbers of social media postings.

Abstract: Systems, devices, computer-implemented methods, and tangible non-transitory computer readable media for performing multilayered generation and processing of computer instructions are provided. For example, a computing device may receive a request with instructions in a first computer language, parse the instructions in the first computer language, analyze the instructions in the first computer language in view of information describing structure of a first application, generate instructions in a second computer language different from the first computer language where the instructions in the second computer language are generated based on the instructions in the first computer language and the information describing structure of the first application, obtain a result from a second application where the result comprises information based on the instructions in the second computing language, and provide the result in response to the request comprising the instructions in the first computer language.

Abstract: Methods and apparatus for associating an entity with at least one search query. Some implementations are directed to methods and apparatus for identifying multiple queries associated with an entity and identifying one or more of the queries as an entity search query that provides desired search results for the entity. Some implementations are directed to methods and apparatus for identifying a particular entity and, in response to identifying the particular entity, identifying an entity search query corresponding to the particular entity.

Abstract: A method, system, and computer program product for identifying and resolving constraint violations in a database replication system determine a predetermined type of constraint violation and, in response thereto, write a record that failed to replicate into an error table on a target database and create a trigger on the error table to automatically and immediately handle the error record, based on why the record failed to replicate.

Abstract: In some examples, candidate projection enumeration based query response generation may include obtaining a query, and identifying, based on the query, a plurality of candidate columns from at least one table to respond to the query. Using a tree-based model, the candidate columns may be analyzed to generate a candidate set of projections to respond to the query. Each candidate projection of the candidate set of projections may include a sort order associated with at least one candidate column of the candidate columns, a segmentation key value associated with at least one candidate column of the candidate columns, and a replication value associated with the candidate projection. Based on the candidate set of projections, a response to the query may be generated.

Abstract: A flux sensing system includes a memory and a processor in communication with the memory and a sensing device, the memory storing a plurality of capabilities and a plurality of semantic fluxes associated with the plurality of capabilities. The computing system is configured to infer a semantic based on received inputs and route the inputs to the semantic fluxes based on semantic drift inference between their associated capabilities and inferred semantic. Further, it positions the sensing device for receiving the input from a user in an optimal manner.

Abstract: Aspects of the disclosure provide a method of determining types of user geographical locations. A target geographical location corresponding to a plurality of uploading times can be obtained. At least one time-location type mapping relationship table can be obtained. Based on the obtained at least one time-location type mapping relationship table, for each of location types in the at least one time-relation type mapping relationship table, a sum of probability values of the respective location type can be calculated to obtain a degree that the target geographical location belongs to each of the location types. Each of the probability values corresponds to one of the uploading times of the target geographical location. Which of the location types corresponds to the target geographical location can be determined according to the degrees that the target geographical location belongs to each of the location types.

Abstract: Systems, methods and apparatus for generating music recommendations based on combining song and user influencers with channel rule characterizations are presented. Such systems and methods output a playlist, which may be delivered as an information stream of audio on a user or client device, such as a telephone or smartphone, tablet, computer or MP3 player, or any consumer device with audio play capabilities. The playlist may comprise various individual audio clips of one genre or type, such as songs, or of multiple types, such as music, talk, sports and comedy. The individual audio clips may be ordered by a sequencer, which, using large amounts of data, generates both (i) user independent and (i) user dependent influencer weightings for each clip, and then combines all of such influencer weightings into a combined play weighting W for a given audio clip, for a given user.

Abstract: Embodiments of the present invention provide a method, system and computer program product for the automated merger of content in progeny of a primary file in a version control system. In one embodiment of the invention, the method can include establishing different segments of content in a primary file in a merger tool executing in memory by a processor of a host computer and associating each of the segments with a classification corresponding to a particular merger rule selected from amongst different merger rules. The method also can include parsing content in different progeny of the primary file and identifying changes in a parsed portion of the progeny corresponding to a specific one of the segments of the primary file. Finally, the method can include applying to the identified changes a merger rule corresponding to a classification associated with the specific one of the segments.

Abstract: Embodiments of the present disclosure may be used to gather, rank, categorize, and perform other processing of various types of content. In some embodiments, content items such as text, images, video, and other content are received from a variety of different sources and are processed to generate an article containing selected content items. While there may be hundreds or thousands of separate articles and stories regarding a particular topic, embodiments of the present disclosure help provide users with a single concise article that contains high-quality content items selected from among a potentially vast number of disparate sources.

Abstract: Personalized network searching, in which a search query is received from a user, and a request is received to personalize a search result. Responsive to the search query and the request to personalize the search result, a personalized search result is generated by searching a personalized search object. Responsive to the search query, a general search result is generated by searching the general search object. The personalized search result and the general search result are provided to a client device, an advertisement is selected based at least in part upon the personalized search object, and the advertisement, the personalized search result, and the general search result are displayed.

Abstract: An apparatus comprises at least one processing device configured to maintain a metadata update journal for at least a first storage node of a plurality of storage nodes of a distributed storage system, and to destage the metadata update journal of the first storage node to one or more storage devices accessible to the first storage node. The metadata update journal of the first storage node utilizes a multi-level bucket aggregation of metadata updates in which metadata pages each comprising a plurality of address-to-hash mapping pairs are arranged in a plurality of first-level buckets containing respective subsets of the metadata pages for respective different logical storage volumes, and a plurality of second-level buckets are each configured to aggregate multiple metadata pages from different ones of the first-level buckets in order to increase a number of metadata updates in each second-level bucket relative to numbers of metadata updates in its corresponding first-level buckets.