99results about "Data merging" patented technology

A method includes generating, by a large language model (LLM), a multitude of search queries from a user input term obtained from a user application. The method further includes, executing, by the LLM, the multitude of search queries to obtain a multitude of verified search results regarding a multitude of candidates. A multitude of web presence schemas corresponding to the multitude of candidates from the multitude of verified search results is generated. The method further includes web-scraping a multitude of websites of a subset of candidates selected from the multitude of candidates to obtain a set of corresponding web-scraping payloads. The method further includes merging the corresponding web-scraping payloads with corresponding web presence schemas of the subset of candidates to obtain a multitude of updated web presence schemas. The method further includes presenting the multitude of updated web presence schemas of the subset of candidates in the user application.

Embodiments of the present application relate to the technical field of artificial intelligence. Disclosed are a method and apparatus for attention calculation in an artificial intelligence model. The method comprises: acquiring a plurality of elements requiring attention calculation; generating a Q matrix, a K matrix, and a V matrix required for attention calculation and corresponding to the plurality of elements; determining a position of valid data in a relation matrix corresponding to the Q matrix and the K matrix; on the basis of the determined position of the valid data in the relation matrix, determining calculation data for matrix multiplication calculation in the Q matrix and the K matrix; performing matrix multiplication calculation on the determined calculation data in the Q matrix and the K matrix to obtain the relation matrix; and performing an operation on the relation matrix and a V matrix to implement attention calculation. By means of the present application, invalid data can be prevented from participating in attention calculation, thereby improving the attention calculation efficiency.

Systems and methods for rendering static and dynamic data in tablature are disclosed. A system may include at least one processor configured to output a signal for rendering a table containing a plurality of cells, wherein at least a first cell may be enabled to contain static data and at least a second cell is enabled to contain dynamic data. The processor may provide access to a menu of values for inclusion in at least the second cell, wherein the menu of values identifies dynamic data values associated with selectable animations. The processor may enable selection of a dynamic data value to specify an associated selectable animation for inclusion in the second cell, and may cause the associated selectable animation in the second cell to dynamically display while the first cell displays static data.

Provided is a data compaction method that includes: determining, by an electronic device, a plurality of Sorted String Table (SSTable) files on which a compaction operation is to be performed among SSTable files stored in a storage device connected to the electronic device, the data in the SSTable files stored in the storage device is stored based on a Log Structured Merge Tree (LSM-Tree) mechanism; based on receiving the plurality of SSTable files from the storage device, performing, by the electronic device, merge sorting on data in data blocks of the plurality of SSTable files based on indexes of the data in the data blocks of the plurality of SSTable files, to obtain reconstructed data blocks; and sending, by the electronic device, the reconstructed data blocks to the storage device, the reconstructed data blocks being used by the storage device to obtain reconstructed SSTable files.

Methods, computer program products, and systems are presented. The methods, computer program products, and systems can include, for example, processing multiple datasets using metadata, wherein the metadata can characterize relationships among datasets. In dependence on such metadata, production datasets can be, e.g., generated, versioned, and / or merged. The resulting production datasets can support subsequent computing uses such as analytics, machine learning, application testing, and / or enterprise processing.

The embodiment of the invention discloses an attention calculation method and device in an artificial intelligence model, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring a plurality of elements which need to be subjected to attention calculation; generating a Q matrix, a K matrix and a V matrix required by attention calculation corresponding to the plurality of elements; and determining the positions of effective data in the relation matrix corresponding to the Q matrix and the K matrix in the relation matrix. And determining calculation data for matrix multiplication calculation in the Q matrix and the K matrix according to the determined position of the effective data in the relation matrix. And performing matrix multiplication calculation on the determined Q matrix and K matrix calculation data to obtain a relation matrix. And carrying out operation on the relation matrix and the V matrix to realize attention calculation. By adopting the data processing method and device, invalid data can be prevented from participating in the attention calculation, and the efficiency of the attention calculation can be improved.

A list sort instruction is provided to perform a sort and / or merge operation. The instruction is an architected machine instruction of an instruction set architecture and is executed by a general-purpose processor of a computing environment. The execution includes sorting a plurality of input lists to obtain one or more sorted output lists as output.

A processor comprising storage, execution circuitry and a handling unit. The handling unit is configured to obtain operation data indicative of an operation to be executed using the execution circuitry. The execution circuitry is configured to operate over a multi-dimensional nested loop defining an operation space. The handling unit obtains a dimensional combination instruction and, in response, combines a plurality of dimensions of the operation space to obtain a local space dimension in an operation-specific local space as part of a procedure to map each operation block of a plurality of operation blocks in the operation space to a different respective local block in the local space dimension. The handling unit dispatches invocation data for the plurality of operation blocks to the execution circuitry. The invocation data for each respective operation block specifies a local space dimension range of a local block to be operated on for the respective operation block.

A hierarchy is a multi-level linked structure of nodes, wherein the hierarchy represents data relating to a set of one or more items to be processed. Where there are multiple input hierarchies, it may improve the efficiency of the processing of the items to merge the input hierarchies to form a merged hierarchy. The hierarchies are merged by identifying two or more sub-hierarchies within the input hierarchies which are to be merged, and determining one or more nodes of the merged hierarchy which reference nodes of the identified sub-hierarchies. The determined nodes of the merged hierarchy are stored and indications of the references between the determined nodes of the merged hierarchy and the referenced nodes of the identified sub-hierarchies are also stored. In this way, the merged hierarchy is formed for use in processing the items.

The present invention provides a data merging method and apparatus of a physical random access channel, and a storage medium. The method comprises: parsing task parameters of this PRACH data merging; reading PRACH data to be merged of an antenna from a PRACH data cache; carrying out, according to the task parameters, PRACH data merging between a plurality of antennas and / or RACH data merging in a PRACH of this antenna; and outputting the merged data to a shared cache.

Disclosed are methods, apparatus, and systems for clustering processes used by a cloud platform. Process clustering may include receiving business data transmitted and received between each pair of processes in a process set. A matrix may be generated based on the business data, the matrix including a row and a column for each process in the process set. Hierarchical clustering may be performed on the matrix based on the business data, the hierarchical clustering outputting a plurality of clusters, each cluster including one or more processes in the process set. The plurality of clusters may then be merged into a set of merged process clusters.

Methods, systems, apparatuses, and non-transitory computer readable media for managing file generation and retention are provided. These systems may be used across a wide range of businesses to significantly automate the generation and storage of files. By utilizing nested file templates as described herein, example systems may increase the ease in which a file template may be generated, stored, retrieved, and retained. Moreover, by using the file retention storage system described below, example systems may simplify the process of ensuring files are retained for an appropriate amount of time and further assuring that files are deleted when they no longer need to be retained.

Performance of a collection of machines, arranged in a linear sequence of machines that form a linear communication orbit (LCO), is monitored. Multiple machines in the LCO receive, via the LCO, a set of rules (or various subsets of the same set of rules), each rule specifying one or a combination of conditions (e.g., a performance metric and corresponding criterion) for satisfying the rule, evaluate those rules with respect to locally occurring events and local processes, and store results of those evaluations in a local database. In response to a query sent to the machines via the LCO, each of the machines returns a report, including information identifying processes whose performance during the specified time period satisfies at least one rule in the set of one or more rules. Those reports are aggregated and used to generate a merged report reflecting performance information with respect to a set of machines.

The present disclosure generally relates to methods, systems, apparatuses, and non-transitory computer readable media for managing generated files. These systems may be of used across a wide range of businesses to significantly automate the generation, storage, and eventual deletion of files. In particular, embodiments of the present disclosure may comprise a server creating a file from a file template in response to a request from a client device. This request or the indicated file template may indicate when the file should be deleted. After the file is created, the server may store the first file and subsequently monitor for the occurrence of a relevant deletion-triggering event. If the server detects the occurrence of a deletion-triggering event, it may then delete the stored file. Thus, files are both retained for an appropriate amount of time and are deleted when they are no longer appropriate to keep.

Storage access for the merge operation is minimized. Multiple records from multiple input lists are merged. The merge includes: determining that one of the input lists has become empty; and, based on this determination, checking the controls dedicated to that input list. These controls determine how to proceed, such as whether to end the merge or continue.

In an environment containing big data, noisy data, and / or unstructured data, it is desirable to identify an entity referenced by input data. The entity can be identified by generating records corresponding to characteristics of the entity based on the input data. These records can be merged when it is determined that more than one record corresponds to the same entity. By doing so it is possible to more easily identify and classify information related to an entity, though such information may have been obtained in a manner that might otherwise be deemed unstructured or noisy. The method can be applied across large sets of data (“big data”) to obtain meaning from data that may otherwise be unclassifiable to a human observer.

A system and method are provided that store electronic data describing events that have occurred in a computing system, index the electronic data to create indexed data records; and store the indexed data records in computer memory as part of a flat data structure.

The invention discloses a bidirectional storage distributed sorting method and system based on an FPGA, and the method comprises the steps: to-be-sorted data is written into a first storage block in a first order through a write address generation module, and the first storage block is RAMX; reading operation is executed from the first storage block RAMX, the first data is written into a second storage block according to a second sequence through bit judgment, and the second storage block is RAMY; reading operation is executed from the second storage block RAMY, and second data are written into the first storage block RAMX according to a third sequence through bit judgment to form bidirectional storage; and repeatedly executing a reading, storing and sorting process until a target number of iterations is reached, and reading a sorted result from the first storage block RAMX or the second storage block RAMY.

A method, system, and computer system for performing an operation with respect to a target table are disclosed. The method includes performing first, second and a third jobs, and obtaining a resulting table based at least in part on the second job resulting file(s) and third job resulting file(s). Performing the first job includes determining a set of matching target table files and storing target table information indicating for each of the set of matching target table files, a particular set of rows having matching rows. Performing the second job includes performing a matching action based on matched rows and obtaining the second job resulting file(s). Performing the third job includes determining unmatched rows for target table files and storing the unmatched rows in third job resulting file(s).

A method for data processing is described. The method may include retrieving, from a cloud storage environment accessible to a data management system (DMS), two or more observability data sets associated with a set of time intervals within a selected time range. The observability data sets may contain a set of time-partitioned data blocks that include tracing data associated with operations performed by one or more nodes of a node cluster of the DMS during the selected time range. The method may further include merging the tracing data by selecting a first subset of overlapping time-partitioned data blocks, omitting a second subset of the overlapping time-partitioned data blocks, and retaining one or more non-overlapping time-partitioned data blocks. The method may further include transmitting data corresponding to a visualization of the merged tracing data.

Embodiments of the present disclosure provide a file merging method for a real-time data lake, a device and a storage medium. The method comprises: obtaining characteristic information of a query task for a real-time data lake table and attribute information of files to be merged in the real-time data lake table; determining, based on the characteristic information of the query task and the attribute information of the files to be merged, whether to initiate a merging task; and in accordance with a determination that the merging task is initiated, merging the files to be merged.

A method of contextual modification of data sharing constraints is disclosed. The method comprises receiving a data sharing request to share a first data model with a database associated with a second data model; generating a shareable version of the first data model in response to the data sharing request; determining a parameter value used to perform a data model merging operation to merge the shareable version of the first data model with the second data model, the parameter value indicating whether to execute or skip a particular process during the data model merging operation; determining context data for the data model merging operation based on the generating; modifying the parameter value based on the context data; performing the data model merging operation using the modified parameter value.

Embodiment of the present disclosure provides method, apparatus, device and storage medium for data query. In the method, first data is obtained by a first device based on a request for a data query, where the request indicates a first data attribute to be classified and aggregated and at least a second data attribute and a third data attribute for classification, and the first data at least includes data associated with a second data attribute; and a third data fragment of a result of the data query is generated based on the first data fragment of the first data and the second data fragment of the second data received from the second device, where the second data at least includes data associated with the third data attribute, and the third data fragment includes an element corresponding to the first data attribute that is classified and aggregated based on at least the second data attribute and the third data attribute.

Techniques for computing matrix operations for arbitrarily large matrices on a limited size hybrid analog-digital matrix processor are described. Techniques for gain adjustment in a limited size hybrid analog-digital matrix processor are described that enable the system to achieve higher energy efficiency, greater physical density, and improved numerical accuracy. In some embodiments, these techniques maximize the prediction accuracy of GEMM-based convolutional neural networks using low-precision data representations.

Methods and apparatus to generate and display trends associated with a process control system are disclosed. An example apparatus includes memory, machine readable instructions, and processor circuitry to execute the instructions to generate a first graphical user interface. The first graphical user interface to include a graphical representation of a component in a process control system. The processor circuitry to generate a second graphical user interface. The second graphical user interface to include a chart region with a trend represented therein. The trend indicative of values of a process parameter of the process control system over a period of time. The processor circuitry to automatically generate the trend in the chart region in response to a graphical element being dragged and dropped from the first graphical user interface to the second graphical user interface.

A digital security system can store data associated with entities in resolver trees. If the digital security system determines that two resolver trees are likely representing the same entity, the digital security system can use a merge operation to merge the resolver trees into a single resolver tree that represents the entity. The single resolver tree can include a merge node indicating a merge identifier of the merge operation. Nodes containing information merged into the resolver tree from another resolver tree during the merge operation can be tagged with the corresponding merge identifier. Accordingly, if the merge operation is to be undone, for instance if subsequent information indicates that the entries are likely separate entities, the resolver tree can be unmerged and the nodes tagged with the merge identifier can be restored to a separate resolver tree.

The present invention provides a loyalty extraction machine, wherein “quadratic multiform separation” (QMS) is modified and executed multiplicatively in an even generalized way. In each execution, the characteristic of one single membership is either enhanced or reduced. This process is performed in turn to each membership. Thus, every sample data (or element) receives multiple classification results. Then, the multiple classification results are collected and analyzed by an “eclectic classifier” to reach a final decision. The combination of the generalized QMS and the eclectic classifier therefore develops the loyalty extraction machine. Moreover, a label called “loyalty type” of the element is introduced to describe the effectiveness of membership recognition with respect to a training set.

Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: processing a plurality of raw datasets for generating index metadata and associating one or more index defining the index metadata to respective ones of the plurality raw datasets, wherein the respective ones of the plurality of raw datasets define respective metadata augmented datasets by the associating of the one or more index thereto; examining augmented datasets of the metadata augmented datasets in dependence on metadata of the index metadata and in dependence on user defined input data; and merging first and second ones of the augmented datasets in dependence on the examining augmented datasets, wherein the merging first and second ones of the augmented datasets in dependence on the examining augmented datasets is performed in support of preparing a production dataset.

A data processing method, apparatus, device, and storage medium are provided. The method includes acquiring a data reading task; determining the target feature data from a data file stored in a target storage system based on the data reading task and metadata corresponding to the target storage system; reading the target feature data from the data file based on operation semantic information corresponding to the target feature data, and determining the read target feature data as the first target feature data; and merging the first target feature data to acquire target merged feature data in the case that the size of data volume of the read first target feature data meets a preset condition, and repeating the process of reading and merging the first target feature data until a completion of reading of the target feature data in the data file.