481results about "Data sorting" patented technology

An information processing method, an electronic device, and a storage medium. The method includes: obtaining a query statement of a user, determining at least one model identifier of at least one candidate service model based on the query statement; generating at least one first prompt word based on the query statement and the at least one model identifier, inputting the at least one first prompt word into a pre-trained target large model, and outputting, by the target large model, at least one screening parameter of the at least one candidate service model based on the at least one first prompt word; determining a target service model from the at least one candidate service model based on the at least one screening parameter; and inputting the query statement into the target service model, and obtaining feedback information corresponding to the query statement.

Various embodiments include systems and methods of anomalous data transfer detection. Hotspots for an asset of an organization may be determined, corresponding to period(s) of time in which outbound data from the asset satisfies a hotspot threshold determined to be indicative of high outbound data traffic activity. Based on the outbound data, a first set of days are identified as “quiet” day(s); a second set of days are identified as “active” day(s); and “quiet” hour(s) of the day, associated with “active” day(s), are identified. The “quiet” day(s) and the “quiet” hour(s) are identified as a warmspot dataset, which may be utilized to detect anomalous data transfer activity associated with the asset. Detecting the anomalous data transfer activity includes computing one or more statistics on the warmspot dataset. Responsive to detecting the anomalous data transfer activity, an alert associated with the asset may be generated.

Embodiments of the present disclosure relate to the technical field of cloud computing, and provide a server failure prediction method and apparatus, an electronic device, and a storage medium. The method comprises: first, acquiring time series data of the operating state of a server; cutting the time series data of the operating state according to a period, and stacking a plurality of pieces of sub-time series data obtained by cutting; extracting time series multi-dimensional features from the stacked data; and performing failure prediction on the server by using the time series multi-dimensional features. In the embodiments, the time series multi-dimensional features are extracted on the basis of the time series data of the operating state for failure prediction, so that a server failure can be predicted in advance. Additionally, the time series data of the operating state are cut according to the period and stacked, and then the time series multi-dimensional features are extracted, so that related information on the fluctuation trend of data over time and different performances of the data before and after the period can be extracted, thereby improving the accuracy and recall rate of failure prediction.

The invention discloses a mathematical algorithm calculation system and optimization method based on big data, and the method comprises the steps: calculating the local entropy density of a data set, constructing an entropy gradient field, and extracting a representative data subset with global sorting information through employing a multi-level recursive strategy; calculating initial partition mapping based on the subset, and introducing an entropy difference driven partition adjustment mechanism to dynamically optimize a data partition boundary; constructing a relative sequence relation of data points in each partition, calculating an optimal exchange sequence by using a minimum cost and maximum flow method, and reducing global adjustment times through a recursive hierarchical structure to achieve local sorting lower than linear time; after sorting, local optimal exchange or global correction is carried out through a dynamic entropy difference adjustment strategy; in a real-time data stream scene, a dynamic sorting topological graph is constructed, only a local area is updated, and global resorting is avoided. Compared with a traditional sorting method, the method has remarkable advantages in the aspects of real-time performance, high efficiency and calculation cost.

The invention discloses a computer system based on a unified information calculation model of electric power Internet of Things terminal equipment, and relates to the field of computer systems, in which a data acquisition module acquires information of a target power grid; the power supply path analysis module determines on-off state change information and a load point path and analyzes to obtain an initial power supply path; a path search module extracts a load point to be transferred and searches for alternative power supply paths to obtain a potential transfer path list; a marking module calls the unified calculation model for analysis, line capacity limitation information and switch operation sequence constraints are obtained, a potential transfer path list is marked through the line capacity limitation information, and a feasible transfer path set is obtained; and the load transfer execution module controls the target power grid to perform load transfer based on the deduced switching operation strategy on the basis of the constraints. According to the method, a data basis is provided for the computer system to execute accurate decision-making of load transfer, and meanwhile, the execution quality and accuracy of the computer system are improved.

A new value is written from a dataset to a data structure comprising a set of sorted values. The new value replaces an oldest value and is inserted in a sorted position. The data structure is modified by subtracting a median value from each value of the set of sorted values to obtain sorted signed deviation values. The sorted signed deviation values are segmented to obtain data substructures comprising subsets of sorted absolute deviation values. A binary search is performed on the data substructures to identify a median absolute deviation value. A difference is computed between a particular value and the median value, and based on whether the difference is less than a threshold value computed from the median absolute deviation value, an outlier decision output is generated indicative of whether the particular value comprises an outlier value.

The invention provides a data sorting method for a tensor processor, a tensor processor system and a medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring input data and the storage capacity of a single tensor processor TPU register, and distributing core id for marking an initial position for each element; when the data size of the input data is smaller than or equal to the storage capacity of the single TPU register, single register level sorting operation is executed, and a double-tone sequence of the single register is obtained; otherwise, performing double-tone sorting recursion on the input data to divide the input data into a plurality of subsequences, executing single-register-level sorting operation on each subsequence, and merging the sorted subsequences to obtain a double-tone sequence of multiple registers; wherein the core id of the element is transmitted when the single register level sorting operation is executed for each sub-sequence and the sub-sequences are combined; and adjusting the sequence of the equal elements according to the core id after the double-tone sequence is obtained. The stability of the data sorting result can be improved.

The invention relates to a webpage search result sorting adjustment method and system based on machine learning. The method comprises the following steps: constructing a data set corresponding to a webpage; updating a data set of the webpage based on the historical operation information; acquiring current search information of a user; obtaining the semantic similarity between the current search information of the user and the webpage data set information, and adjusting the webpage sequence based on the semantic similarity to obtain a first sequence; adjusting the first sequence based on a user model and a public model to obtain a second sequence; and based on the semantic similarity between the historical search information of the user and the current search information, adjusting the second sequence to obtain a third sequence. According to the method, the historical operation condition of the user on the webpage is fully considered, the operation condition of the user is matched with the search information, the webpage data set information is updated based on the search information, and the participation of the user behavior in the search result is fully reflected.

The invention provides a data rearrangement method and device, electronic equipment and a storage medium. The data rearrangement method comprises the steps of obtaining input data comprising sub-data of multiple dimensions; determining a sub-instruction meeting a preset rule from a pre-stored rearrangement instruction; determining first position information of the sub-data in the input data by utilizing the sub-instruction; splitting the rearrangement instruction based on the sub-instruction to obtain a first instruction and a second instruction; executing a first rearrangement operation on the sub-data according to the first position information and the first instruction to obtain cached data; the first instruction indicates a mapping relation between the first position information and second position information of the sub-data in the cache data; executing a second rearrangement operation on the sub-data in the cache data according to the second position information and the second instruction to obtain output data; the second instruction indicates a mapping relation between the second position information and third position information of the sub-data in the output data. The invention relates to the technical field of data rearrangement, and the data rearrangement efficiency can be improved.

The present application relates to a device for selecting top values from a set of raw values, comprising: an output queue, a loop queue, a top value storage module and a control module. The control module is configured to, at a higher priority, merge the intermediate sequence stored in the loop queue with the at most N top values stored in a storage area of the top value storage module, and sort the merged values to generate a merged sequence, until a predetermined number of storage areas in the top value storage module are traversed; wherein the control module is further configured to, when there is no intermediate sequence being stored in the loop queue, merge the output sequence with the at most N top values stored in a storage area of the top value storage module, and sort the merged values to generate a merged sequence; wherein the control module is further configured to provide a first subsequence in the merged sequence which is closer to a top most value of the merged sequence to the top value storage module to update the top value storage module, and provide a second subsequence in the merged sequence which is farther away from the top most value of the merged sequence to the loop queue to generate or update the intermediate sequence.

Methods and systems provide content searching and retrieval using generative artificial intelligence (AI) Models. The system is configured to receive a user search for content, media or item listings. The system receives a natural language-based input associated with a client device of a user. The system generates a search criterion for the received natural language-based input. The system, via the generative AI-bases search and retrieval system, generates a relevancy-ranked output listing of content items. The relevancy-ranked output listing content items responsive to the generated search criterion content items having an associated content identifier and a content description. The system generates a carousel display structure definition of the relevancy-ranked content items. The system transmits the carousel display structure definition of the relevancy-ranked content items and the content items to the client device. The client device renders, via a user interface, at least a portion of the relevancy-ranked content items.

Systems and methods are provided herein for generating a synthetic training data set that can be used to train a machine-learning model to identify when two addresses match (e.g., when a user-defined address and an authoritative address match). The addresses may each be tokenized. Each candidate address can be scored based on a number of common tokens it shares with the user-defined address. The highest-scored candidate address may be selected as a matching address for the user-defined address. In some embodiments, a number of the remaining candidate address can be selected as negative examples (e.g., candidate addresses that do not match the user-defined address) based on, for example, historical delivery information associated with the corresponding addresses. In this manner, an expansive training data set may be generated using addresses associated with user profiles of an online service provider and a set of authoritative addresses obtained from an authoritative source.

Provided are systems, methods, and apparatuses of data processing for machine learning. In one or more examples, the systems, devices, and methods include determining priority values for elements of a weight matrix based on a gradient of a loss function of an AI model and the weight matrix; determining an index value based on a number of elements in the weight importance matrix and a sparsity ratio; determining a threshold based on sorting the elements of the weight importance matrix in sequential order and determining a value of an element of the sorted weight importance matrix based on using the index value as an index of the sorted weight importance matrix; determining a pruned weights matrix based on the threshold; and processing a query using an updated AI model, the updated AI model being based on the pruned weights matrix being implemented in the AI model.

A case management system and method of managing and organizing documents within the case portfolio for efficient review includes data that is dynamically updated to result in a final report that includes real-time updated data. The system collects and displays the data on a secure portal for interactivity with a professional stakeholder reviewing the data. The data is automatically transformed into different formats when received, including a standard text-and-image file, a text-only file, and an image-only file, thereby allowing the reviewing stakeholder to interact with the data in multiple ways. As the system translates the document into different viewable formats, the system automatically scans the data within the document for important information, such as dates, times, names, locations, and other important pieces of information. The system then assigns date-and-time associations with each document, and automatically organizes the documents in a chronological order on a display.

Apparatuses, systems, and techniques to perform non-maximum suppression (NMS) with a bit-reduced radix sort to remove redundant bounding boxes are described. In at least one embodiment, one or more circuits perform i) a bit-reduced radix sort operation to sort a list of confidence scores associated with a set of bounding boxes corresponding to one or more objects within one or more digital images and ii) a non-maximum suppression (NMS) operation on the sorted list to remove one or more redundant bounding boxes from the set.

A data input sub-system of a parallelized database system includes processing core resources operable to obtain divisions of data slabs of a dataset and compress the divisions of data slabs using a null elimination compression scheme to produce divisions of compressed data slabs, A first data slab of a first division of data slabs of the divisions of data slabs is compressed using the null elimination compression scheme to produce a first compressed data slab. The first compressed data slab includes first compressed data and first compression information. The processing core resources are further operable to store a respective division of compressed data slabs of the divisions of compressed data slabs.

A point cloud data transmission method according to embodiments may comprise the steps of: encoding point cloud data; and transmitting a bitstream comprising the point cloud data. A point cloud data reception method according to embodiments may comprise the steps of: receiving a bitstream comprising point cloud data; and decoding the point cloud data.

Techniques for handling block format floating point and / or integer numbers are described. In some examples, circuitry for handling block format floating point and / or integer numbers includes a plurality of multiplexers to select between the output of the mantissa multiplier circuits and outputs of the shift circuits to allow for support for block and non-block numbers.

Provided are a data processing method, an electronic device, a computer-readable storage medium and a computer program product, which relate to the field of databases. The data processing method comprises: determining, from a database, M data groups to be integrated, wherein M is a positive integer greater than 1, each data group to be integrated comprises a plurality of data records having sorting key values and has a corresponding sorting key value interval, a minimum endpoint value and a maximum endpoint value of the sorting key value interval are respectively a minimum sorting key value and a maximum sorting key value in each data record in the corresponding data group, and for any data group to be integrated, there is at least one other data group to be integrated, the sorting key value interval of which overlaps with the sorting key value interval thereof; and reintegrating the M data groups to be integrated into at least one target data group, and adding the at least one reintegrated target data group to the database.

Methods and systems of verifying a hardware design for a sorter are disclosed. The methods include generating a modified version of the hardware design of the sorter accepting extended inputs and performing formal verification comprising: implementing a constraint that the least significant bits of each input in a set of extended inputs represent a unique value; and formally verifying aspects of the modified version of the hardware design of the sorter for the set of extended inputs.

The invention relates to a method for rapidly counting and analyzing massive time sequence violation paths. The method comprises the following steps of: performing read operation on a time sequence report file generated by a static time sequence analysis tool by adopting a storage I / O mapping technology in a shared memory, and separating an original file into a plurality of independent fragments according to a multi-mode character string matching algorithm, starting a plurality of sub-threads and distributing information of corresponding fragments of the original file to the sub-threads; adopting a multi-mode character string matching algorithm to obtain specific attributes of a time sequence violation path according to the keyword information; the sub-threads are used for carrying out attribute extraction and distribution statistics according to specific attributes in sequence, and after the main thread receives that the sub-threads complete attribute extraction and distribution statistics, various kinds of distribution statistics information of the sub-threads is merged into a total statistical table of the main thread according to a rapid sorting algorithm to generate various kinds of distribution statistics sorting information. By adopting the method, rapid statistics and analysis of massive time sequence violation paths can be realized.

Provided are systems, methods, and apparatuses of data processing for machine learning. In one or more examples, the systems, devices, and methods include determining priority values for elements of a weight matrix based on a gradient of a loss function of an AI model and the weight matrix; determining an index value based on a number of elements in the weight importance matrix and a sparsity ratio; determining a threshold based on sorting the elements of the weight importance matrix in sequential order and determining a value of an element of the sorted weight importance matrix based on using the index value as an index of the sorted weight importance matrix; determining a pruned weights matrix based on the threshold; and processing a query using an updated AI model, the updated AI model being based on the pruned weights matrix being implemented in the AI model.

A method for enhancing hybrid-classical algorithms for combinatorial optimization includes executing, on a quantum processor, a variational algorithm, the execution producing a subset of a set of solutions (solution space) of the variational algorithm within a predefined period, the variational algorithm computing a quantum state of a quantum system corresponding to a combinatorial optimization problem, each solution in the subset having a corresponding value; sorting, according to a sorting criterion, the subset of solutions; isolating, based on the sorting, a portion of the subset of solutions wherein a value corresponding to each solution in the portion is within a boundary defined by a threshold value; computing an average value of the solutions from the portion of the subset of solutions; and altering the variational algorithm to produce a second subset of solutions such that the second subset of solutions comprises solutions having values within the boundary.

To provide a suitable processor for alignment processing.SOLUTION: A processor (core A3) includes: a first source operand; a second source operand; a third source operand; a fourth source operand; a fifth source operand; and a sixth source operand. The processor (core A3) also includes: a first adder-subtractor for computing a first intermediate sum of the element of the first source operand and the element of the second source operand; a second adder-subtractor for computing a second intermediate sum of the element of a third source operand and the element of the fourth source operand; and a third adder-subtractor for computing a third intermediate sum of the element of the fifth source operand and sixth resultant data obtained based on the element of the sixth source operand. Further, the processor (core A3) has a selector for selecting the maximum value and / or the minimum value from among at least the first intermediate sum, the second intermediate sum, and the third intermediate sum.SELECTED DRAWING: Figure 21

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 method is provided that includes performing, by a processor in response to a vector matrix multiply instruction, multiplying an m×n matrix (A matrix) and a n×p matrix (B matrix) to generate elements of an m×p matrix (R matrix), and storing the elements of the R matrix in a storage location specified by the vector matrix multiply instruction.

This specification discloses a data processing and visualization method, apparatus, medium, and device, which relate to the computer technology field. The method includes selecting a data platform to acquire source data of a target company, organizing the source data and loading it into a data system as a source database table, performing primary data operations to process the source data into target data, performing secondary data operations to process the target data into value-added data, performing tertiary data operations to process related data into graphical data, and performing data visualization to render the graphical data into a visualization diagram. The method of the present invention at least partially solves the problems of inability to timely and completely acquire source data and inability to effectively utilize source data without processing it, thereby improving data processing efficiency, increasing the accuracy of data analysis of the target company through data modeling, and automatically combining background data with front-end displays such as panel interfaces through data visualization, thereby improving the effectiveness of data presentation and the availability of the data system.

The application discloses a circuit and method for long sequence data sorting. The circuit is used for sorting N data in K base, and the maximum element in the data contains m bits in K base. The circuit comprises a radix counting unit, a first address generating unit, a data distribution unit and two sorting buffers. The two sorting buffers are used as a source buffer and a target buffer respectively, and can read data in a given address or write a data into a specified address. The radix counting unit, the first address generating unit and the data distribution unit are connected in sequence. The two sorting buffers are connected with the data distribution unit and the radix counting unit respectively. The data sorting circuit of the application has simple structure, can be flexibly adjusted according to specific requirements, and the data sorting method has linear order time complexity and short sorting time.

A processor and a vector sort instruction for the processor to execute are provided, in which the vector sort instructions includes instructions for comparing a first element of a set of vector elements of a vector to a remainder of the set of vector elements; determining, based on the comparing, a control vector that specifies a respective sorted position for each element of the set of vector elements; and reordering the set of vector elements based on the control vector.