57 results about "Complex data structures" patented technology

The invention belongs to the technical field of aviation energy design, and particularly provides an aviation energy power device fault intelligent inquiry system and method based on a knowledge graph. The method comprises the following steps: firstly, showing a complex data structure and a knowledge relationship to a user in a visual and dynamic manner through knowledge extraction, knowledge graph database construction, knowledge graph visualization and other methods, and constructing a knowledge graph basic framework; according to the method, the depth information of the target object can be extracted from the images from different angles, the fault characteristics of the equipment can be intuitively displayed, and the fault state of the equipment can be visually represented by extracting the depth information of the target object from the images from different angles. The evolution of the equipment state can be tracked through the change of the three-dimensional space; a conversion algorithm of the knowledge graph and the Bayesian network is realized through the Bayesian network, and rapid reasoning of fault causes is realized in combination with a Bayesian network reasoning algorithm; and finally, based on an intelligent inquiry system of a large model, a deep learning technology and a large-scale training model are utilized, and previous fault data and professional knowledge of the aviation energy power device are combined to provide intelligent auxiliary diagnosis support.

A system with graph-based Natural Language Processing (NLP) for querying, analyzing, and visualizing complex data structures is described. Such a system executes a generalized AI language model; defines and migrates a training dataset into a graph database by exposing data sources to an executing AI language model that self-defines a structure and self-writes an executable script to query the original data sources and self-writes code to load the extracted data into a graph database in the form of new nodes and new relationships with directionality between the nodes. The system further includes means for loading the extracted data into the graph database; condensing the information stored within the graph database into a condensed data structure representing the full architecture of the data in a natural language format; and responding to human language inquiries with responsive text, speech, and visualizations using the data loaded into the graph database.

The invention relates to the technical field of data warehouses, and particularly provides a statistical aperture generation method and device, a storage medium and computer equipment. The large aperture generation model can automatically understand the complex data structure of the data warehouse through the physical table building statement of the data warehouse. When the initial statistical caliber is obtained, mutual exclusion verification, grammar compliance verification and data source matching verification can be performed on the initial statistical caliber so as to verify whether the statistical caliber generated by the large model has the problems of logic conflict, grammar non-compliance or data source inconsistency and the like which influence the reliability of the statistical result; and judging whether the statistical caliber generated by the large model can be directly put into use or not. And if yes, setting a final statistical caliber corresponding to the target data index according to the initial statistical caliber. Therefore, by introducing the large model technology, the technical threshold is greatly reduced, the generation efficiency of the statistical caliber is improved, and the reliability of data analysis can also be ensured.

The invention discloses a data model field mapping method and device, a storage medium and a processor. According to the scheme, metadata of a source model and metadata of a target model are obtained; constructing a candidate field mapping set based on the metadata of the source model and the metadata of the target model; traversing each pair of candidate field mappings in the candidate field mapping set, calculating a comprehensive similarity value of each pair of candidate field mappings in the candidate field mapping set, and determining a mapping rule corresponding to each pair of candidate field mappings based on the comprehensive similarity value of each pair of candidate field mappings, and converting each pair of candidate field mapping from the source model to the target model based on the determined mapping rule. Compared with the problems of low efficiency, high error rate and the like of a manual maintenance mode in the prior art for a large number of fields and complex data structures, the method has obvious advantages.

The invention relates to the technical field of data storage and synchronization of intelligent equipment, and provides an off-line data storage synchronization method and system, comprising a data and file downloading storage method in an off-line mode, a data synchronous uploading method in an on-line mode, and a data and file downloading storage method in the off-line mode. Comprising the following steps: a front end sends a downloading request for acquiring data, files and a file-data relation table to a rear end through a calling interface: the rear end locks the data requested to be downloaded according to the downloading request, and then downloads the data so as to realize exclusive downloading of the front end; and the back end determines the number of files downloaded at a single time according to the sizes of the files aiming at the corresponding files in the file-data relation table, and performs fragmented downloading on the files. Storage and synchronization of complex data structures are effectively achieved, and the method is particularly suitable for storage and offline query of large data volumes.

An artificial intelligence (AI) language model is trained using input data from one or more original data sources and a selected training algorithm that includes data extraction and condensing techniques. The trained AI model generates self-written code to create an executable query script for extracting data and an executable load script for loading the extracted data into a graph database. The extracted data is stored as newly created nodes within the graph database. The AI model transforms the loaded data into a condensed data structure that represents the full architecture of the data in a natural language format. The AI model is further trained to respond to user input in natural human language using the condensed data structure. Upon receiving a user query, the AI model processes the input, generates a structured data query, retrieves relevant information from the graph database, and outputs a response in natural language.

The invention relates to a large-scale education data migration method based on dynamic service routing, and the method comprises the steps: carrying out the optimization of a ClickHouse batch writing operation, employing a transaction control and distributed lock mechanism in the data conversion and writing process, effectively preventing the data from being repeated or lost, enabling the converted data to be efficiently written into a ClickHouse target table through a batch loading mode, and enabling the data to be more stable and reliable. Dynamic table name adaptation and complex data structure conversion are supported, efficient and reliable migration of ten-million-level education data is achieved through asynchronous parallel processing, fragmentation task scheduling, a multi-dimensional duplicate removal mechanism and an abnormal retry strategy, and the throughput and consistency of data writing are remarkably improved.

The invention relates to a data synchronization and comparison method and device, and aims to efficiently and accurately solve the problem of data consistency of a source end and a target end. The method comprises the following steps: (1) data partitioning: partitioning data of a source end and a target end according to a preset rule; (2) hash calculation: generating a hash value for each data block; (3) hash comparison: rapidly screening inconsistent data blocks by comparing hash values; (4) recording and comparing: comparing inconsistent data blocks item by item and field by field to determine specific differences; and (5) data restoration: restoring the target end data according to the comparison result. According to the method, the comparison data volume is reduced through data partitioning and Hash calculation, and the comparison efficiency is remarkably improved; hash comparison and record comparison are combined, so that the comparison accuracy is ensured; the scheme does not depend on external factors, is suitable for various complex data structures, and has wide applicability.

The invention discloses an agricultural Internet intelligent connection platform integrating multi-modal data. The platform comprises an Internet of Things sensing work center, a data intelligent work center, a business cooperation work center, an intelligent application and an agricultural cloud. The Internet of Things sensing work center is used for equipment optimization and protocol adaptation, mass connection management, data standardization, data storage and sharing and equipment management; the data intelligent working center is used for data aggregation and processing and intelligent algorithm and model development and application; the business cooperative work center is used for unifying user management, authority distribution and message pushing and management; the intelligent application is used for interacting with a user and realizing interaction and display of agricultural supply and demand information, agricultural production information, agricultural information and product traceability; the agricultural cloud is used for storing, managing and distributing data. According to the method, the analysis efficiency of a complex data structure can be improved, the compatibility of multi-modal data is improved, and the adaptation and analysis capability to different devices is enhanced.

The invention belongs to the related technical field of runoff forecasting, and discloses a runoff forecasting method and system based on an improved generative adversarial network, and the method comprises the steps: building an initial sample set according to runoff historical data; training a variational auto-encoder model by using the initial sample set, the trained variational auto-encoder model being used for generating a reconstructed sample according to an initial sample in the initial sample set, and combining the initial sample and the reconstructed sample to form a reconstructed sample set; constructing a forecasting model based on the generative adversarial network, and training the forecasting model by using the reconstructed sample set to obtain a trained forecasting model; and forecasting the runoff by using the trained forecasting model. According to the method, the data are reconstructed by constructing the variational auto-encoder, the original data are merged and input into the subsequent forecasting model, the understanding ability of the model for a complex data structure is enhanced, the forecasting model is trained based on the reconstructed sample set, the training difficulty is reduced, and the forecasting precision of the trained forecasting model is improved.

The invention discloses a multi-source complex table-oriented trusted question and answer agent construction method and system, a question and answer agent is constructed based on a ReAct framework, at least a code interpreter tool and a document reading tool are used in the ReAct framework, and the multi-source complex table-oriented trusted question and answer agent construction method comprises the following steps: constructing a multi-mode complex table analysis tool; the analysis module is used for analyzing a non-editable unstructured table file into an editable structured table text; constructing a single-mode complex table cleaning tool for performing abstract modeling, identification extraction and structure reconstruction on a complex data structure of the structured table text to generate a normalized table which is simpler than an original table in structure; and constructing an illusion detection and repair tool for performing real-time illusion verification on the output content of the question and answer agent, and triggering a fallback regeneration mechanism when illusion is detected. According to the invention, through tool set expansion and multi-tool cooperation, an efficient and credible table question and answer agent is constructed.

This invention belongs to the technical field of runoff forecasting, and discloses a runoff forecasting method and system based on an improved generative adversarial network (GAN). The method includes: establishing an initial sample set based on historical runoff data; training a variational autoencoder (VA) model using the initial sample set; using the trained VVA model to generate reconstructed samples from the initial samples in the initial sample set; combining the initial samples and reconstructed samples to form a reconstructed sample set; constructing a forecasting model based on a GAN; training the forecasting model using the reconstructed sample set to obtain a trained forecasting model; and using the trained forecasting model to predict runoff. This invention enhances the model's ability to understand complex data structures by constructing a variational autoencoder to reconstruct data and merging the original data into the subsequent forecasting model. Training the forecasting model based on the reconstructed sample set helps reduce training difficulty and improves the forecasting accuracy of the trained forecasting model.

This invention discloses an FPGA-based image classification method and system in the field of image classification technology. It aims to solve the problems of high computational load and complex data structures in existing deep neural network modules, which increase the difficulty of FPGA deployment and result in low operating efficiency. The method includes steps such as system initialization, loading model weights, acquiring image data, generating feature matrices, and displaying results. The image classification method provided by this invention can fully utilize hardware resources, linearly reduce the computational load of basic convolutional modules, and achieve the goal of improving operating efficiency.

The invention concerns an OPC-UA (Open Protocol Communications Unified Architecture) system and its use in creating light digital cognitive twins for use in industrial automation, especially in edge IoT devices. OPC UA clients can retrieve real-time data updates and historical data for analysis and interaction with complex data structures. Cognitive Digital Twins (CDTs) represent a significant evolution in digital twin technology, integrating cognitive abilities into the digital representations of physical assets. The lightweight cognitive and interlinked digital twin embedded system a) is connected to digital assets provided by PLC server (or relevant edge IoT-enabled architecture) and b) is powered by a lightweight Open Protocol Communications Unified Architecture (OPC-UA) to enable seamless and continuous generation and update of the digital twin.

This invention relates to a quantum-enhanced clustering method based on spheres, belonging to the field of quantum computing technology. This method addresses the problems of high computational complexity, difficulty in capturing intrinsic geometric features, and sensitivity to noise in traditional clustering algorithms when processing large-scale, high-dimensional, nonlinear data. The technical solution includes obtaining a compressed set of spheres from the original dataset and extracting the center vector; encoding the center vector into a quantum state through quantum feature mapping; constructing a quantum kernel matrix using quantum circuits to calculate fidelity; calculating cohesion based on the quantum kernel matrix to identify and remove noisy spheres; performing hierarchical clustering on the core sphere set; and mapping the results back to the original data points. This invention improves clustering accuracy, robustness, and the ability to handle complex data structures.

The invention discloses a metabolite structure prediction method and system based on a BERT twinning neural network model. According to the method, the BERT-based twinning neural network model is constructed, and the structure identification of the metabolite is carried out through feature extraction, similarity calculation and network training optimization. Compared with a traditional method, the feature extraction capability is improved by adopting deep learning, a complex data structure can be better dealt with, and then the accuracy of similarity calculation is improved. The application of the invention can more quickly and accurately identify metabolites, and especially has important theoretical significance and practical value when being applied to large-scale and high-dimensional data sets. A verification result shows that the method shows relatively high prediction precision in public and actual data sets, and a more reliable metabonomics data analysis tool is provided.

The present application provides a data verification method, device, system and storage medium, the method comprising: respectively initializing source data and data to be verified to obtain source data frame and data frame to be verified; respectively extracting labels from the source data frame and data frame to be verified according to the loaded template rule to obtain label data corresponding to the source data and label data of the data to be verified; comparing the label data corresponding to the source data with the label data of the data to be verified to obtain a data verification result. The present application is applicable to the verification processing of complex data structures, can effectively improve the verification efficiency of data, and has strong scalability.

This invention relates to the field of software development technology and discloses a method, apparatus, device, and storage medium for interface iteration risk alarm. The method includes: acquiring data returned by the interface after an iterative update as comparison data, and acquiring pre-stored data returned by the interface before the iterative update as reference data; converting the comparison data and reference data into a structured data model in a preset format using a nested structure to generate a first structured data representation and a second structured data representation; performing a multi-dimensional comparison based on the first and second structured data representations to obtain a multi-dimensional comparison result; and determining whether to trigger an alarm notification based on the business impact of the multi-dimensional comparison result. Applying the technical solution of this invention can solve the problems of inefficiently handling complex data structure comparisons during interface iteration and the imperfect alarm mechanism in the prior art.

In accordance with various embodiments, described herein is a system (Data Artificial Intelligence system, Data AI system), for use with a data integration or other computing environment, that leverages machine learning (ML, DataFlow Machine Learning, DFML), for use in managing a flow of data (dataflow, DF), and building complex dataflow software applications (dataflow applications, pipelines). In accordance with an embodiment, the system can provide support for auto-mapping of complex data structures, datasets or entities, between one or more sources or targets of data, referred to herein in some embodiments as HUBs. The auto-mapping can be driven by a metadata, schema, and statistical profiling of a dataset; and used to map a source dataset or entity associated with an input HUB, to a target dataset or entity or vice versa, to produce an output data prepared in a format or organization (projection) for use with one or more output HUBs.

The invention belongs to the technical field of additive manufacturing, and discloses an LDED dilution rate monitoring method based on a Markov transfer field and related equipment.The method comprises the steps that one-dimensional spectral signals of excited atoms in a molten pool are obtained in real time; preprocessing the one-dimensional spectral signal to obtain a preprocessed one-dimensional spectral signal; converting the preprocessed one-dimensional spectral signal into a two-dimensional feature map by adopting a Markov transfer field method; inputting the two-dimensional feature map into a trained ResNet18 model, and outputting a prediction range of the dilution rate; the training data of the ResNet18 model comprises a two-dimensional feature map with different dilution rate range labels. The objective of the invention is to solve the limitation of a one-dimensional spectral signal analysis method in processing a complex data structure and realize real-time monitoring and accurate classification of the dilution rate.

The invention discloses a geographic target query method, which comprises the following steps of: determining a specified area by utilizing longitude and latitude information of a reference point and a query condition, further determining a query result of a geographic target in the specified area according to the longitude and latitude information of each candidate target, and quickly screening each geographic target on a map based on a longitude and latitude range. Judgment can be carried out without calculating the distance from the reference target one by one, a complex data structure is not needed, and implementation is simpler; the calculation efficiency of longitude and latitude information is higher, and the number of points needing to accurately calculate the distance can be reduced; the query method provided by the invention is small in memory occupation, does not need to configure additional index storage space, is easy to integrate with an existing system, and has no special requirement on a data storage format.

The present invention discloses an AI-driven efficient data redundancy detection and cleaning method and system, which relates to the field of data processing technology. It includes constructing an AI-driven fragment recognition architecture based on the categories of different paragraphs in the target data chain, and the fragment recognition architecture is used to locate the location of duplicate data in the target data chain; and the duplicate data is used as the positioning reference. The present invention performs structured analysis on the target data chain through the fragment recognition architecture, combines structural features, semantic features and functional features, and realizes the positioning and recognition of duplicate data in different contexts and different data types. It is suitable for diversified data scenarios such as text sequences, data streams, multimedia images, etc. In addition, by constructing a feature mapping relationship table, it is possible to construct a fragment recognition framework based on the feature content of different paragraphs, and adopt AI recognition logic matching, so that it has higher adaptability and recognition accuracy when facing complex data structures and implicit redundancy problems.

An early fault positioning method for a complex data structure of a large hydraulic turbine set belongs to the technical field of hydraulic turbine set fault detection, and comprises the following steps: collecting 52 key operation parameters, covering multiple types of data such as rotating speed, power, vibration and the like, classifying the complex data structure according to the collection position, and carrying out abnormal value processing, vacancy value filling and normalization preprocessing to obtain a complex data structure; dimension differences are eliminated, and data quality is improved. The Pearson's correlation coefficient is calculated by using a sliding window, the parameter correlation change is monitored in real time, a dynamic threshold value is set based on the historical data standard deviation, and the abnormality of the unit is accurately judged. And after abnormality is detected, constructing fault feature vectors containing correlation variable quantity and abnormal moment parameter values, and dynamically adjusting fault position scores according to pointing weights of a plurality of feature vectors so as to realize accurate fault positioning. According to the method, the multi-source data processing problem is systematically solved, the fault early warning and positioning accuracy of the unit is remarkably improved, and safe and stable operation of the large hydraulic turbine unit is powerfully guaranteed.

The invention relates to the technical field of railway signal safety control, and provides an interlocking product data analysis method, system and equipment, and the method comprises the steps: analyzing program language data in interlocking software, and obtaining a structural body type definition and a structural body variable; a multi-way tree is constructed according to the structural body type definition, and a multi-way tree model is optimized through node replacement; the optimized multi-way tree is adopted to carry out assignment analysis on the structural body variables, and structural body variable values are obtained; and verifying the obtained structural body variable value and input data, and outputting a verification result. According to the interlocking product data analysis method, system and equipment, structured analysis of interlocking program language data is realized by constructing the multi-way tree model and carrying out node replacement optimization, a complex data structure is efficiently processed, unanalyzed nodes are ensured to be correctly supplemented through a queue traversal and type replacement mechanism, and the data analysis efficiency is improved. And the integrity and accuracy of data analysis are improved.

The invention relates to the technical field of electric digital data processing, in particular to a multi-source heterogeneous data fusion method and system, and the method comprises the steps: connecting a distributed data collection platform to obtain a to-be-fused data set, constructing a hierarchical processing path, carrying out data format conversion and authority verification, and optimizing a decision to generate a fusion strategy. By introducing an error model, a calling probability set and a risk mapping mechanism, the accuracy, efficiency and safety of data fusion are improved. According to the method, the problems of low processing efficiency, insufficient flexibility and limited data security guarantee of a complex data structure can be effectively solved, and the reliability and adaptability of data fusion are remarkably improved.

The invention discloses a coding module for an optical neural network, an input port and a three-dimensional input port. The coding module comprises a plane wave generator; an information encoder; the plane wave generator is arranged at the upstream of the information encoder along the light path propagation direction; the plane wave generator is used for converting the received light beams into plane waves, and the information encoder is used for carrying out information encoding on the plane waves; wherein the information encoder is a reconfigurable metasurface, and the reconfigurable metasurface comprises at least two reconfigurable micro-nano units. According to the method, the light beam is converted into the plane wave before information coding, so that the precision of information coding is improved, and the accuracy of processing a light calculation task is further improved; moreover, the adjustable metasurface serves as an information encoder, so that the number of information which can be borne in a unit area is larger, the input information density of the optical neural network is improved, a data structure with higher dimensionality or more complexity can be processed, and the optical neural network can process a large-scale optical calculation task.

The invention provides a cross-project defect prediction method based on semi-supervised clustering data screening, and aims to effectively improve the performance of a defect prediction model. According to the method, firstly, software module data is processed by utilizing a semi-supervised clustering algorithm, and the data is divided into a plurality of sub-clusters, so that a complex data structure is effectively organized and simplified; then, by analyzing the historical software module category of the current project, screening out cross-project historical software module data consistent with the category label of the historical software module, and forming a screened cross-project data set; and then, combining the screened data with historical software module data of the current project, modeling by using a naive Bayes classification algorithm, and finally constructing a cross-project defect prediction model. In the whole process, the interference of irrelevant data on model training is avoided by screening the relevant data, and meanwhile, the historical software module information of the cross-project and the project is fully fused, so that the actual application effect of defect prediction is enhanced while the accuracy and generalization ability of the model are improved.

The invention discloses a face image clustering method, which is carried out through a self-adaptive multi-view face image clustering model based on non-negative matrix factorization, and comprises the following steps of: 1, preprocessing data; the second step is density segmentation; 3, preliminary clustering is carried out; 4, calculating cluster centrality and deviation degree; 5, dimension conversion is carried out; 6, constructing a model; and 7, optimization solution is carried out. According to the face image clustering method, the number of clusters does not need to be predefined, unsupervised learning can be carried out, the loss function can contain two data adaptive parameters through the second step to the fifth step, a novel double-information graph structure and an auxiliary clustering algorithm are combined, and the face image clustering method which is good in robustness and optimal in performance is obtained. The robustness, the accuracy and the purity of a clustering algorithm during processing of a complex data structure related to a face image can be improved.

An embodiment of the present application provides a data structure for multi-party call records, and a call record aggregation method based on the above data structure. The multi-party call record adds a summary field on the basis of the existing two-party call records. This field can record summary information of the entire call record. Therefore, when aggregating multi-party call records, the electronic device can determine whether the compared call records meet the aggregation conditions by comparing the summaries in the call records. Implementing the method provided by the embodiment of the present application can achieve the aggregated display effect of multi-party call records. At the same time, the improved multi-party call record data structure can also make the aggregation process more efficient, thereby solving the problems of lengthy logic, complex data structure, low efficiency, etc. in the process of aggregating multi-party call records using the two-party call record aggregation method.

The application discloses a large-scale water turbine set complex data structure-oriented early fault positioning method, belongs to the technical field of water turbine set fault detection, and comprises the following steps: collecting 52 kinds of key operation parameters, covering multiple types of data such as rotating speed, power and vibration; classifying the complex data structure according to collection positions; and performing preprocessing such as abnormal value processing, missing value filling and normalization, so as to eliminate dimension differences and improve data quality. A sliding window is used to calculate a Pearson correlation coefficient, real-time monitoring of parameter correlation change is performed, a dynamic threshold is set based on historical data standard deviation, and unit abnormalities are accurately judged. After detecting abnormalities, a fault characteristic vector containing a correlation change amount and an abnormal time parameter value is constructed, a fault position score is dynamically adjusted according to the pointing weight of multiple characteristic vectors, and accurate fault positioning is realized. The method systematically solves the problem of multi-source data processing, significantly improves the accuracy of unit fault early warning and positioning, and effectively guarantees the safe and stable operation of large-scale water turbine sets.