241 results about "Data patterns" patented technology

Apparatus and method for generating an output using an AI-PII model. The apparatus includes at least a processor and memory communicatively connected to the at least a processor. The memory instructs the processor receive personally identifiable information (PII) data, receive one or more model constraints, map, using an AI-PII model, the PII data to at least a data schema as a function of the one or more model constraints by identifying at least a PII datum of the PII data, categorizing the at least a PII datum to one or more categories of a plurality of categories, and mapping the PII data to the at least a data schema, modify the data schema based on a refinement datum, wherein the refinement datum is generated based on a temporal datum of the one or more model constraints, and generate an output as a function of the refinement datum and data schema.

Systems and methods for intelligent data sanitization employing PCM and AI / ML are provided. The idea uses AI / ML to detect specific facts that needs sanitization rather than full properties in incoming records. Data sanitization is optimized using this focused method, saving computational resources. To properly manage changing data volumes, PCM shifts between Logical 0 and Logical 1 states. Logical 0 processes smaller volumes with high resistance and low conductivity, while Logical 1 processes large volumes with low resistance and high conductivity. The AI / ML module organizes and directs data to maximize resource and processing efficiency. The PCM processes data in-memory and directly overwrites, eliminating erasure. AI / ML and PCM integrate to sanitize data quickly, efficiently, and securely, improving system performance and data integrity without a central repository. The system dynamically adjusts to changing data patterns, protecting and optimizing data.

The invention discloses a multi-agent collaborative application big data management system and method. The system comprises a data acquisition unit, a data preprocessing unit, a multi-agent collaborative processing unit, a data storage unit and a data scheduling unit. The data acquisition unit is used for acquiring multi-source heterogeneous original big data and transmitting the multi-source heterogeneous original big data to the data preprocessing unit; and the data preprocessing unit is used for executing preprocessing operation including de-duplication and format standardization on the original big data. The invention relates to the technical field of big data management. According to the multi-agent collaborative application big data management system and method, through a multi-agent collaborative algorithm, the system can realize a self-adaptive data classification and optimization strategy, a data processing mode is dynamically adjusted, the processing efficiency problem of a traditional static rule during data mode fluctuation is improved, and the data processing efficiency is improved. The system adopts a dynamic resource scheduling strategy based on a data access demand and a system load, efficiently distributes calculation and storage resources, and facilitates priority processing of high-priority tasks.

The invention provides an intelligent test data generation and verification method and system based on a hierarchical architecture, and relates to the technical field of data generation, and the method comprises the steps: constructing a cross-layer constraint dependency graph by analyzing each layer of data mode and constraint and conversion logic of a hierarchical architecture software system; generating candidate test data meeting the constraint based on the dependency graph in a back-stepping manner, and verifying the candidate test data through projection; and iteratively optimizing the dependency graph and the test data by monitoring the difference between the actual execution and the prediction form. According to the method, cross-layer data consistency verification is realized, the test coverage rate and the defect detection efficiency are improved, and the quality risk of a software system is reduced.

A method for verifying a low-density parity-check (LDPC) unit capable of being applied in a memory system can include receiving original data corresponding to a memory device, encoding the original data by the LDPC unit to be verified, injecting errors into the encoded original data by a data pattern for generating verifying data, and verifying a soft decode capability of the LDPC unit by utilizing the verifying data. The data pattern can include the errors generated by threshold voltage (Vth) distributions interlaced between two neighboring logic states of 2n logic states of the memory device. The method and system can provide an error injection to accurately and efficiently verify a LDPC soft decode capability of the LDPC unit, decrease errors, increase error correction accuracy and efficiency, more accurately model actual threshold voltage (Vth) distributions, increase flexibility, increase speed, increase performance, and reduce firmware overhead.

The invention discloses a data simulation method based on a database function and a timed task. The method comprises the following steps: S1, acquiring source data based on a multi-modal lightweight intelligent acquisition strategy; s2, performing data cleaning and standardization so as to ensure engineering processing of simulation data quality and consistency; s3, creating a user-defined function in the database to package intelligent logic of data simulation; according to a time condition and a data mode in the timed task, intelligently performing simulation trigger judgment and simulation data generation through the user-defined function; s4, performing efficient data injection based on a buffer and batch optimization mechanism of the intermediate table; and S5, performing timed task period regulation and control. The invention further discloses a corresponding system, electronic equipment and a computer readable storage medium.

The invention relates to the technical field of medical information, in particular to an interventional operation patient perioperative period nursing information intelligent interaction and management system which comprises the steps that through an integrated data collection module, a timeline generation module, a risk identification module, a nursing intervention module and an interaction management module, time axis alignment and structured coding are conducted on scattered physiological data and diagnosis and treatment events; and a dynamic comprehensive nursing timeline is formed. And performing real-time analysis on the timeline based on a preset risk knowledge graph, identifying an abnormal data mode and generating an early warning. The system automatically matches a nursing rule base, pushes individualized intervention suggestions including specific operations, priorities and bases, synchronizes information and collects feedback through a multi-role interaction channel, and realizes closed-loop management. According to the system, fragmented nursing information can be integrated into a continuous dynamic view, and risk early warning and structured nursing actions are directly linked, so that the collaboration, accuracy and timeliness of nursing management are improved.

Methods, systems, and computer-readable storage media for receiving a user query provided in natural language and requesting data stored in a resource according to a data schema, determining a set of relationship and context data represented in a knowledge graph, the set of data schema and context data determined to be relevant to the user query from a super-set of data schema and context data stored in a graph database, generating a prompt using the user query and the set of data schema and context data, receiving a data query from a LLM that is responsive to the prompt, the data query being in a structured format, processing the data query to provide a query result, and displaying the query result to a user.

According to embodiments, a memory system includes a data management circuit, a memory, a write control circuit, and a data determination circuit. The data management circuit manages data, which is received from a host on a first data size basis, on a second data size basis. The second data size is greater than the first data size. The data determination circuit determines whether or not data of the second data size received by the write control circuit matches a data pattern set in advance. In a case where first data received by the write control circuit matches the data pattern, the data management circuit sets a flag indicating that the first data matches the data pattern in an entry of a management table corresponding to the first data, and the write control circuit discards the first data without writing the first data to the memory.

An apparatus and method for generating at least a data connector. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor. The memory instructs the processor to receive input data from an external source, wherein the input data comprises a plurality of source parameters, identify a first data schema associated with the external source and a second data schema associated with an internal system, instantiate a runtime instance of a data connector associated with the external source as a function of the first data schema and the second data schema, configure the data connector by binding the data connector to a web service protocol, registering the data connector to the internal system, and validating the data connector using at least a compliance protocol, and deploy the data connector to a service-based orchestration layer of the internal system.

A framework for easy development of a machine learning solution is provided. The framework includes connecting data sources at all scale levels from a user; assessing data schema, data risk, and data quality status; performing scalable feature engineering and transformation; and automating machine learning model optimization. The framework also includes functionalities for generating production code and automatic model documentation.

A diagnostic system for communication behavior is provided. The diagnostic system includes a wireless device configured to communicate with an intranet, and a diagnostic device. The diagnostic device includes a data receiving module, a pattern identification module, an event detection module, and a processing module. The data receiving module is configured to receive at least one log data from the wireless device. The pattern identification module is configured to receive the log data and identify whether the log data has at least one key pattern of data. The event detection module is configured to obtain an event corresponding to the key pattern of data. The event represents a communication behavior of the wireless device. The processing module is configured to perform a specific operation corresponding to the event.

The invention relates to a spatio-temporal data rule extraction method based on spatio-temporal Fourier expert mixing, belongs to the technical field of urban spatio-temporal data processing, solves the problem that global stability characteristics and local disturbance characteristics driven by real physical rules are difficult to effectively describe in the prior art, and comprises the steps that S1, a spatio-temporal data acquisition module acquires historical data information; s2, establishing a space-time Fourier expert hybrid network, and carrying out expert mixing to obtain mixed features; s3, using adaptive group normalization as a conditional fusion module to obtain fusion features; s4, establishing a space-time Fourier attention mechanism module to obtain output features; s5, establishing a diffusion model for performing back diffusion on the noisy data in combination with the noise estimation network to obtain predicted denoised data, and performing training to obtain a trained diffusion model; and S6, performing sampling to obtain spatio-temporal data, inputting the spatio-temporal data into the trained diffusion model, and obtaining an extracted causal law for urban traffic flow prediction.

This invention provides a multi-model fusion method, device, electronic device, and system for debris flow early warning. The method includes: collecting geographical and geological data, real-time rainfall, runoff, and auxiliary monitoring data; fusing a shallow flow numerical model, a moving least squares machine learning model, and a generalized regression neural network model, and obtaining a comprehensive runoff prediction through weighted least squares calculation; constructing critical runoff indicators based on the safety requirements of disaster-bearing bodies such as roadbeds, retaining walls, and culverts, and inversely calculating critical rainfall thresholds, then making differentiated adjustments based on the average gradient of the gully; generating a debris flow early warning report by comparing real-time predictions with thresholds, achieving final graded and zoned early warning for the warning area. This invention improves the accuracy and engineering relevance of early warnings by integrating physical mechanisms and data patterns, achieving refined early warning from "whether it will occur" to "when, where, and what the impact will be," while reducing the error of a single model and adapting to different terrains and disaster-bearing bodies along highways.

The invention relates to the technical field of Internet data processing, in particular to an industrial Internet equipment data extraction method and system, and the method comprises the steps: S1, analyzing multi-source equipment data through a full-protocol compatible layer, and outputting a standardized data stream through protocol analysis middleware and a data fusion mechanism; when the system is used, by seamlessly integrating industrial internet equipment with more than three hundreds of protocols, plug-and-play is realized, the cost of docking industrial data islands and systems is reduced, and by standardizing a data output uniform format, format fragmentation, semantic conflicts and time sequence chaos of original data are avoided, and a data circulation barrier between equipment is broken; a weak data mode is introduced through the edge computing terminal, data preprocessing, caching and mode optimization are completed, real-time screening, rapid positioning and target data extraction of high-throughput data are achieved, the rear-end computing pressure and the data integration cost are synchronously reduced, and the flexibility and efficiency of industrial data full-link processing are improved.

The embodiment of the application discloses a data conversion method, system and device, the method comprises the following steps: when a trigger condition corresponding to a scheduling task is met, triggering the scheduling task, and transmitting task parameters to a push task main program; through the push task main program, target data in a database is queried based on the task parameters in a stream mode, the target data is parsed according to a template file, and the target data is written into a landing shard data file; after the target data in the database is queried and written into the landing shard data file, the total number of the target data under each data mode is determined, a text file corresponding to each data mode is generated according to the total number of the target data under each data mode and each landing shard data file under each data mode; and the text file corresponding to each data mode is compressed, and the compressed text file is uploaded to a target server. The method can ensure the timeliness and accuracy of data pushing, and realizes decoupling of the database and a data consumer.

Matrix multiplication operations can be implemented, at least in part, on one or more tensor cores of a parallel processing unit. An efficiency of the matrix multiplication operations can be improved in cases where one of the input operands or the output operand of the matrix multiplication operation is a square matrix having a triangular data pattern. In such cases, the number of computations performed by the tensor cores of the parallel processing unit can be reduced by dropping computations and / or masking out elements of the square matrix input operand on one side of the main diagonal of the square matrix. In other cases where the output operand exhibits the triangular data pattern, computations can be dropped or masked out for the invalid side of the main diagonal of the square matrix. In an embodiment, a library implementing the matrix multiplication operations is provided.

A memory device operates with multiple internal ranks that enable concurrent access. The memory device receives multiple commands from a memory controller, which are assigned to different individual ranks and executed concurrently. Data is transferred between the memory device and the memory controller in an interleaved data pattern that includes interleaved data associated with multiple commands for multiple different ranks.

Systems, methods, and computer-readable media are provided for triggering functionality on data to be generated in a user interface and / or data shown or visualized in a user interface based on a natural language request that references actions to be performed and data items to use in performing the actions. The user interface actions are triggered based on a structured object generated by a large language model (LLM), which may then be processed, validated, and used to carry out the actions. The structured object may cause generation, on the user interface, of a representation of slice(s) of data across different dimensions with different filter(s) applied to include one or more dimensions and exclude one or more other dimensions. The slice(s) of data may be determined based on available dimension(s) and / or dimension value(s) specified, in the prompt, for the data schema. If the representation is a grid, a shape of a grid may be recommended for showing the slice(s) of data determined.

In some implementations, a memory apparatus may receive a command to write data to volatile memory of the memory system. The memory apparatus may compare the data to one or more data patterns to identify whether the data matches a data pattern of the one or more data patterns. The memory apparatus may increment a reference counter associated with the data pattern based on identifying that the data matches the data pattern. The memory apparatus may write, based on identifying that a measure flag associated with the data pattern is set, the data to the volatile memory.

Techniques for adapting data collection in vehicles for enhanced security and privacy are provided. A computer-implemented method, performed by a data processing device of a vehicle, comprises selecting a data mode, regarding data capable of being collected via one or more data collection devices integrated on or within the vehicle, from amongst a plurality of different data modes configured for the vehicle that vary with respect to one or more settings selected from the group consisting of: a collection setting that controls collection of the data via the one or more data collection devices, a storage setting that controls storage of the data, and a transfer setting that controls transfer of the data. The method further comprises controlling the collection, the storage, and the transfer of the data in accordance with the data mode, as opposed to another data mode of the different data modes, based on the selecting.

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may generate and transmit at least one TCP ACK message according to a delay timer. A duration for the delay timer may be based on a downlink data pattern associated with downlink data packets for the UE and an uplink scheduling pattern associated with uplink data packets for the UE. The UE may generate and transmit multiple TCP ACK message segments corresponding to the TCP ACK message based on the downlink data pattern and the uplink scheduling pattern. A TCP ACK message segment may correspond to feedback for a respective portion of the quantity of data. The UE may generate and transmit a TCP ACK message associated with the quantity of data based on a transmission blanking pattern.

To accurately detect read errors in postcodes. [Solution] The magnetic disk device of the embodiment comprises a magnetic disk having a plurality of tracks, a magnetic head for reading and writing data to the magnetic disk, a plurality of servo regions provided on each of the plurality of tracks and including at least a portion of a first physical pattern which is the physical pattern after writing of a first data pattern including position information for each of the plurality of tracks, and a control unit for controlling the magnetic head. Each of the plurality of servo regions includes either a first servo region including the first physical pattern, or a second servo region including a second physical pattern which is the physical pattern after writing of a second data pattern different from the first data pattern, for each of the plurality of tracks, and the control unit determines whether the third servo region among the plurality of servo regions of the plurality of tracks has the first or the second physical pattern.

An error correction circuit includes an error processing circuit, a poison processing circuit, and a data pattern conversion circuit. The error processing circuit receives data, generates delayed data, a parity signal, and syndrome information, and outputs error correction data generated by correcting errors in the data according to the syndrome information. The poison processing circuit outputs one of a write poison flag and a read poison flag as a poison pattern control signal in accordance with a write command. The data pattern conversion circuit converts the delayed data and the parity signal into a poison pattern according to the poison pattern control signal and outputs the poison pattern.

The invention provides a fan fault sensing method and system based on variable optimization of a multi-dimensional time sequence measurement mutual information estimator. The method comprises the following steps: constructing a multi-dimensional time sequence measurement data set; constructing a non-parametric mutual information estimator to perform space division and mutual information calculation; constructing a candidate variable set, and performing feature optimization on the candidate variable set based on a non-parametric mutual information estimator; constructing a long and short term memory network deep learning model and performing model training; and carrying out online fan fault classification perception by using the trained long and short term memory network deep learning model. According to the method, a data driving method is adopted, feature selection is carried out on multiple variables in an SCADA system based on a multi-dimensional time sequence measurement mutual information estimator, and feature variables which are more beneficial to a fan fault early-stage detection data mode are recognized and selected. Based on the optimized feature subsets, a more accurate and more efficient deep learning model for online fan fault prediction can be obtained, and real-time probability prediction is facilitated.

A system and method are provided for artificial intelligence (AI) model selection and loading. The method includes receiving an input data stream for an AI application and analyzing the stream to determine temperature scores based on topic frequency, importance weightage, and access frequency. The method also includes mapping these temperature scores to relevant expert AI models and predicting the temperature and context of the next prompt using historical data patterns through a sliding window that smooths temporary anomalies. The system dynamically selects AI models based on temperature scores and predicted contexts, then pre-loads these models into a memory hierarchy where higher-temperature models are placed in faster memory. The method processes incoming prompts using these pre-loaded models and outputs responses, improving processing efficiency through predictive model management. The system addresses challenges of managing large-scale AI models by implementing intelligent pre-loading mechanisms that anticipate and prepare for upcoming processing needs.

The present application relates to a kind of time series prediction methods under multi-scale double clustering, the present application can simultaneously capture the data pattern and relevance under different time scales by multi-scale input mechanism, improve the modeling ability to complex real-world signal.Effective decoupling and modeling TDS and CCT: by parallel patch intra-domain clustering module and mask attention module, respectively, special modeling is carried out on two core challenges of time distribution drift and channel correlation conversion, avoids causal confusion, and improves the robustness and generalization ability of the model.The model carries out twice dynamic, hybrid expert-based dependency filtering at patch level, can accurately capture instantaneous and time-varying correlation, avoids the information loss or noise interference caused by traditional coarse-grained or static dependency modeling.

An object of a system according to an embodiment is to efficiently and accurately perform a checking operation when an alarm is generated.SOLUTION: A system includes an alarm data collection part, a pattern learning part, a confirmation work proposal part, a confirmation work execution part, a report part, and a history storage part. The alarm data collection unit collects past alarm data. The pattern learning unit learns a pattern based on the data collected by the alarm data collection unit. The confirmation work proposal unit proposes confirmation work when an alarm is generated. The checking work execution unit executes the checking work proposed by the checking work proposal unit. The reporting unit reports a result of the checking work executed by the checking work execution unit. The history storage unit stores a history of the confirmation work executed by the confirmation work execution unit.SELECTED DRAWING: Figure 1

A system and method for evicting a resource from a unified network topology (UNT) of a computing environment, is disclosed. The method includes: generating a representation of a UNT for a network topology of a computing environment, the representation based on a unified data schema, wherein the unified data schema is mapped to a plurality of data schemas, each corresponding to a unique data source; receiving an event record at a first time, the event record including a resource identifier; updating in the UNT a representation corresponding to the resource based on extracted data from the event record at the first time; determining that a time period has lapsed from the first time which is greater than a threshold; detecting that a number of event records which include the resource identifier is less than a threshold value; and removing the representation of the resource from the unified network topology.

The invention relates to a photovoltaic output prediction method based on a damping Caputo fractional order grey model, and belongs to the technical field of new energy prediction. Aiming at the problem of insufficient prediction precision caused by randomness, volatility and small sample characteristics of photovoltaic output data, the method provides a scheme: firstly, establishing an initial sequence based on historical data, and performing data preprocessing by calculating an r-order damping Caputo accumulative generation sequence, an accumulative reduction sequence and a mean value sequence; the method comprises the following steps: constructing a damping Caputo fraction Logistic grey prediction model, estimating model parameters by using a least square method, adaptively optimizing by using a particle swarm optimization algorithm to determine the optimal order of the model, and finally obtaining a photovoltaic output prediction value through a time response function and a cumulative reduction formula. According to the method, the adaptability and prediction robustness of the model to a complex data mode are improved, and the intelligence of the parameter optimization process is realized.