52 results about "Heterogeneous source" patented technology

The present disclosure describes systems and methods for detecting and preventing data misconfigurations within a security-focused data fabric platform. The system integrates an advanced script migration engine designed to streamline the translation of security rules and scripts across different scripting languages while ensuring alignment with the fabric's unified schema. The method involves receiving inputs from data sources, mapping these inputs to entities of a target schema, monitoring real-time data changes, and simulating impacts on operational dependencies to detect misconfigurations proactively. Leveraging AI-driven mechanisms, including Large Language Models (LLMs), the system dynamically identifies breaking changes in third-party data streams, issues alerts, and provides suggested fixes. The script migration engine further enhances the platform's functionality by automating cross-platform script translations and enabling faster onboarding of security tools. Together, these innovations ensure scalable, accurate, and resilient integration and management of security data across heterogeneous sources, strengthening operational integrity and minimizing security risks.

The embodiment of the invention discloses a dynamic metadata sensing and self-adaptive mapping method and system, and the method comprises the following steps: deploying a lightweight probe to be agented to a plurality of heterogeneous source database systems, and building a trusted communication channel between the probe and a coordination center through equipment fingerprint generation and security authentication; a database log event is captured in real time, log analysis is carried out through an FPGA module, the change of a field structure or a semantic level is identified, and structured event data is generated; performing context embedding representation on the captured event based on a semantic understanding model, performing strategy reasoning in combination with a historical strategy and a knowledge graph, and generating a matched mapping strategy; loading and executing the mapping strategy in a trusted execution environment (TEE) to realize data conversion, cleaning and complementation, and performing desensitization processing on sensitive data through a differential privacy mechanism; and collecting performance indexes and abnormal logs of a strategy execution result, continuously training the model through a closed-loop optimization mechanism, and correcting an unreasonable strategy or updating a rule base.

The invention relates to the technical field of information retrieval, in particular to a vertical website deep crawling method based on intention recognition, which comprises the following steps: crawling data from a plurality of heterogeneous information sources, and constructing a dynamic knowledge graph; identifying a weak signal association path formed by connecting a plurality of relation edges in the dynamic knowledge graph, and converting the weak signal association path into a to-be-verified risk hypothesis; generating a data acquisition intention, and based on the to-be-verified risk hypothesis, generating a rejection query for searching reverse evidence as a first data acquisition intention; receiving an input query instruction, and performing semantic analysis on the query instruction to generate a second data acquisition intention; a multi-source crawling strategy is generated based on the type of the data collection intention, and the multi-source crawling strategy comprises a plurality of vertical website sources and crawling priorities of the vertical website sources; executing a deep crawling task according to the multi-source crawling strategy to obtain target data; and processing the target data, verifying the to-be-verified risk hypothesis, and judging whether the reverse evidence exists or not and judging the intensity of the reverse evidence.

The invention relates to the technical field of electric digital data processing, and discloses an intelligent data processing method for clinical research. The method comprises the following steps: constructing a basic clinical knowledge graph fused with an international medical ontology; the method comprises the following steps: performing deep learning driven entity recognition and knowledge graph linking on original data from heterogeneous sources such as an electronic medical record and an inspection system; executing cross-source entity alignment and knowledge fusion based on the graph attention network; performing automatic data quality verification and restoration according to clinical logic rules embedded in the atlas; and finally, extracting and generating a standardized analysis ready data set from the enhanced atlas according to research requirements. According to the technical scheme, high-quality, automatic and semantic integration and flexible delivery of clinical research data are realized, and the data processing efficiency and research reliability are improved.

The invention provides a heterogeneous data mapping optimization method, and relates to the technical field of software development. A JSON Schema model library containing business and voucher objects is constructed by obtaining heterogeneous source data, the construction process comprises data format analysis and feature extraction, and construction and classified storage are completed by means of DSL rules and visual editing; based on the model library, generating a mapping relation description document by utilizing a preset abstract mapping rule template and an expanded JSON Schema grammar; a mapping engine analyzes the document to construct a field-level dependency graph, data conversion is executed according to the field-level dependency graph, and an inerting execution and caching mechanism is adopted in the conversion process; and verifying the converted data and the mapping process by adopting a mode of combining static pre-verification and incremental verification, dynamically adjusting a mapping rule according to a verification result or model change, and performing version management. Through the steps, the data mapping efficiency, flexibility and accuracy are improved, and the system performance is optimized.

The model generation platform enables generation of machine learning models based on modular storage of user-specified training data. The platform can obtain a dataset from a different heterogenous source (e.g., a structured database, an unstructured database, a semi-structured file system, manual upload of a comma-separated value file, a spreadsheet, and / or big data) through an associated application programming interface and store this data in a first storage medium. The model generation platform can obtain an indication of a portion of the dataset from a user via a user interface and determine a second storage medium for this portion of the dataset based on an associated estimated performance metric. In response to a request for generation of a machine learning model, the model generation platform can generate a machine learning model using training data comprising a subset of the portion of the dataset.

A database comparison tool is provided that compares a source database to a target database. The database comparison tool performs the comparison in two steps: an initial comparison step and a confirmation step. In the initial comparison step, a server retrieves rows of a table from source and target database tables with a query using an agent. The server compares rows by using a hash value for all non-key columns. The resulting rows are flagged as potentially being out-of-sync. In the confirmation step, the server fetches the potentially out-of-sync rows from the source and target databases and performs a literal value-by-value comparison. Each agent receives a batch of rows, inserts the batch of rows into a temporary table, and performs a join operation between the temporary table and the actual source or target table. The agent returns the rows resulting from the join operation to the server.

The application provides a multi-level index calculation system and method based on configuration, and relates to the technical field of data processing, and comprises an acquisition module, a rule configuration module and an index calculation module; the acquisition module is configured with a metadata model; the acquisition module is used for acquiring a plurality of heterogeneous source data and extracting field metadata by using the metadata model; the rule configuration module is used for configuring a calculation rule for a target index based on the index type of the target index; the index calculation module is used for matching the field metadata of the plurality of heterogeneous source data according to each calculation rule, and determining the heterogeneous source data with matched field metadata as index calculation data required for calculating the target index; and in the order from high to low of the calculation priority, the statistical calculation method of the corresponding calculation rule is sequentially executed on the index calculation data to obtain index result data of the target index. The whole process dynamic generation from original data to multi-level indexes is realized.

The invention discloses a method for integrating and analyzing heterogeneous data (3) from a heterogeneous source application (2). The method is used for solving the inherent non-standardization problem of a data model used by a Web-based database platform. All platforms adopt different data configurations, so that a unified and standard data presentation mode cannot be followed, and structural aggregation of information is not facilitated. In order to achieve the purpose, the computer implementation method comprises the steps that a calculation data collection program (1) is executed to access and load heterogeneous source data (3); providing a heterogeneous data integration module (4) for converting the heterogeneous source data (3) into target data (5); performing a data analysis process (6) based at least on the target data (5) to generate output data (7); and writing the output data (7) to the normalized database (8).

This invention relates to a fault early warning system for hydro-generator units based on CEGD-FNN. The system includes: a feature extraction module for early fault signals of the hydro-generator unit, a fault diagnosis module for the hydro-generator unit, a status assessment module for the hydro-generator unit, a fault prediction module for the hydro-generator unit, and a human-machine interface module. These modules are sequentially connected. This invention employs a unified data platform integrating heterogeneous source information and a unified platform for open knowledge resource sharing. It can promptly obtain heterogeneous data from distributed systems and knowledge resources from different users, thereby achieving intelligent collaborative diagnosis of equipment faults and reducing losses caused by insufficient timely handling of emergency faults.

The application provides a connection code generation method and device, electronic equipment, computer readable storage medium and computer program product; the method comprises the following steps: determining a plurality of uniform interface models in the memory for a plurality of source heterogeneous source files corresponding to a plurality of hardware modules; based on each uniform interface model, a hierarchical signal tree is constructed for each hardware module; based on a plurality of signal trees, determine the signal pair to be matched, and determine the bit width topology characteristics of the interface signal in the signal pair to be matched; calculate the similarity between the names of the interface signals in the signal pair to be matched, and the compatibility score of the direction; based on the bit width topology characteristics, the similarity and the compatibility score, determine the consistency score of the signal pair to be matched, and when the consistency score is greater than the score threshold, it is determined that the signal pair to be matched has a connection matching relationship; based on the connection matching relationship, a first connection code is generated. Through the application, the generation accuracy of the connection code of the interface signal can be improved.

The application provides a data management method, comprising: acquiring heterogeneous source data; the heterogeneous source data comprises basic information and power consumption data of an enterprise; performing data adaptation on the heterogeneous source data to obtain first data; performing standardization on the first data to obtain second data; performing quality evaluation on the second data; and generating a quality report according to the result of the quality evaluation. By adapting and standardizing the data, the type of data that can be analyzed is increased, the adaptability of analysis is expanded, the heterogeneous source data is comprehensively analyzed, and the efficiency of data processing is improved.

This application provides a method, apparatus, device, and medium for heterogeneous code translation. The method includes: acquiring heterogeneous source code running on a source heterogeneous hardware system; compiling the heterogeneous source code to generate an intermediate representation of the source heterogeneous hardware system; mapping the intermediate representation of the source heterogeneous hardware system to obtain an LLVM intermediate representation; and mapping the LLVM intermediate representation to obtain RISC-V extended code, which is used to run on a target hardware system. This application improves the compatibility between code and target hardware systems.

The invention discloses a mammary gland endoscope preoperative anatomical variation identification method and system, and relates to the technical field of computer data processing. The method comprises the following steps: S1, acquiring heterogeneous source data, and performing spatial registration and data standardization fusion to generate a multi-dimensional fusion data set; s2, performing basic topological feature extraction based on the multi-dimensional fusion data set, and calculating edge weight information to construct a target topological graph; s3, matching and comparing the target topological graph with a preset standard topological database, and calculating topological compliance parameters based on comparison differences; s4, identifying a parameter abnormal region according to the spatial distribution characteristics of the topological compliance parameters, and generating a spatial security constraint model in combination with local spatial medium attributes; and S5, based on the spatial security constraint model, outputting structure abnormal information and corresponding optimized path planning parameters. And through multi-channel tensor quantization variation characteristics, safety constraints are constructed by using an anisotropic bounding box, and a path is optimized in combination with a potential field algorithm, so that high-precision automatic obstacle avoidance is realized.

Systems and methods for generating deployment databases based on preloaded data from heterogeneous sources are disclosed herein. The system may receive first structured data. The system may extract first data of a first format and second data of a second format. The system may determine that the first data has a first update rate. The system may determine that the second data has a second update rate. The system may preload the first data by retrieving a first identifier and by storing a first preloaded representation of a first parent dataset for the first data. The system may receive a first request for a deployment database. The system, based on the first request, may retrieve the first preloaded representation. The system may generate the deployment database.

This invention provides a method, device, and storage medium for generating and utilizing a multimodal discourse tree (MMDT). [Solution] The method generates an Extended Discourse Tree (EDT) from a text corpus (e.g., from a Discourse Tree (DT) or Communication DT (CDT)), links data records (e.g., records containing numerical data) to the Extended Discourse Tree, and generates a Multimedia Discourse Tree (MMDT). The MMDT links any suitable text / records from heterogeneous sources. For example, entities identified from the basic discourse units of the EDT are matched with entities in the data records. The method also identifies causal links between EDTs and / or between data records, identifies rhetorical relationships for each entity match / causal link match, and merges the data records with the EDT to generate the MMDT.

A system for training a model from a subset of data representing decentrally stored source databases. A key variable repository module operably couples the databases and includes an AI program with a scanner algorithm and a profiler algorithm. The scanner algorithm receives the training data from a source interface, compresses the training data, and synchronizes the training data with the meta-data using a meta-database interface. The profiler algorithm receives the meta-data from the meta-database interface, generates granular data types for the meta-data, determines training variables indicative of the meta-data, generates variable probability distributions, produces training variable associations, and modifies the meta-database to include the probability distributions and associations using the meta-data interface. The key interface allows for searching the meta-database for training variables, variable probability distributions, and / or variable associations. A model of the system may be trained in less time with a subset of data associated with the training variable.

The invention relates to the field of intelligent doorplate management, and particularly discloses an intelligent doorplate dynamic management system and method based on an LORA technology, which are characterized in that external heterogeneous source data such as OA, educational administration and the like are grabbed, cleaned and packaged in a standardized manner at a front end through protocol adaptation and data engine technologies, and information islands are broken to realize automatic content production; the transmission layer reconstructs network data into high-penetrability LORA radio-frequency signals for wide-area broadcasting by utilizing a cloud compiling and gateway protocol conversion mechanism; and the terminal side adopts a periodic wake-up strategy based on a CAD mechanism, and drives the electronic ink screen to refresh and immediately and automatically cut off the power supply only after the verification of the effective signaling is passed. Therefore, real-time synchronization of the physical identification and the digital service is realized without manual intervention through full-link data closed-loop circulation, and meanwhile, extremely-low-power-consumption long-term operation of the system is ensured by utilizing a bistable display characteristic.

This invention provides a three-dimensional registration method for neutron / X-ray heterogeneous source images of cylindrical multilayer structures. The method includes edge extraction operators and PCA principal component analysis for vertical adjustment of image pairs; an edge contour extraction module to determine the height mapping of heterogeneous source tomographic images; a circle extraction algorithm and calculation of the circle center positioning translation; an image super-resolution reconstruction algorithm to achieve image scaling; and image rotation matching based on feature tomography and phase cross-correlation methods. This method effectively corrects geometric and modal differences between heterogeneous source images, achieving accurate alignment of metallic and non-metallic regions. It has significant advantages in registration accuracy, edge preservation, and modal feature alignment. It exhibits high adaptability and stability for complex structures and multimodal characteristic characterization, and is suitable for three-dimensional registration of neutron / X-ray heterogeneous source images of cylindrical multilayer structures such as cylindrical lithium batteries.

This invention proposes a heterogeneous data mapping optimization method, relating to the field of software development technology. It constructs a JSON Schema model library containing business and credential objects by acquiring heterogeneous source data. The construction process includes parsing the data format, extracting features, and completing the construction and classification storage using DSL rules and visual editing. Based on this model library, a mapping relationship description document is generated using a preset abstract mapping rule template and extended JSON Schema syntax. The mapping engine parses the document to construct a field-level dependency graph, and performs data transformation accordingly. The transformation process employs lazy execution and caching mechanisms. The transformed data and mapping process are verified using a combination of static pre-verification and incremental verification. Furthermore, the mapping rules are dynamically adjusted and version managed based on the verification results or model changes. This invention improves the efficiency, flexibility, and accuracy of data mapping and optimizes system performance through the above steps.

The invention provides a logistics path control system based on machine learning. Relates to the technical field of logistics technology and intelligent control, and comprises a data acquisition module used for acquiring logistics related data in real time and outputting original multi-source heterogeneous data, a feature processing module used for performing cleaning, self-supervised space-time pre-training and feature construction on the source heterogeneous data, generating a high-quality comprehensive feature vector used for path optimization, and outputting the high-quality comprehensive feature vector. And the feature processing module comprises a data cleaning unit. According to the logistics path control system based on machine learning, multi-target logistics path optimization is realized through a deep reinforcement learning algorithm, and the driving distance, the distribution time efficiency and the operation cost of a distribution path are effectively balanced. By dynamically adjusting the target weight, the system can automatically adjust the optimization strategy according to the real-time traffic condition, weather change and other multi-dimensional factors, so that the distribution path can respond to the change in real time, and the distribution efficiency is improved while the total distribution cost is reduced.

The invention relates to the field of data communication, in particular to an oil and gas station heterogeneous data integration method and device, electronic equipment and a computer readable storage medium. The method comprises the following steps: acquiring heterogeneous source data generated by each data source through a source channel corresponding to each data source; environment interference intensity in the oil and gas station is collected; adjusting each source channel according to the environmental interference intensity so as to reduce the packet loss rate of the heterogeneous source data; and mapping the heterogeneous source data to a standard data form based on a preset mapping rule to obtain standard source data in one-to-one correspondence with the heterogeneous source data, and fusing all the standard source data. According to the oil and gas station heterogeneous data integration method and device, the electronic equipment and the computer readable storage medium, the technical effect of improving the data quality of the data obtained through multi-source heterogeneous data integration and fusion in the oil and gas station can be achieved.

The present invention relates to a hybrid optimization method and system for island-type heterogeneous source load-storage inertia assessment. This method first models multi-source inertia and constructs a spatiotemporal distribution discrete model based on the multi-source inertia dynamic model and related parameters. An ant colony algorithm is then used to generate an initial solution set for the spatiotemporal distribution discrete model. A determination is made as to whether the solution set satisfies a first constraint; if not, the solution set is returned and regenerated. Finally, a genetic algorithm is used to further process the solution set, output an optimized solution set, and determine whether the optimized solution set satisfies a second constraint. If so, the optimal inertia configuration is output; if not, the optimization is returned and continued. Compared with the prior art, the present invention has advantages such as improving the effectiveness of inertia optimization results.

The invention belongs to the technical field of data processing, and provides a scientific research domain knowledge graph construction method, system and device and a storage medium, and the method comprises the steps: building a database, employing a dynamic data strategy, employing a BERT-wwm model to solve entity ambiguity, constructing a domain dictionary tree, and forming a heterogeneous source database processing system; constructing a hierarchical progressive reasoning framework which comprises a basic relation layer, a business relation layer and a dynamic evolution layer; an entity relation network is stored through Neo4j, time sequence change data is recorded by using TTDengine, a dynamic relation topology is established, and a knowledge graph is generated. According to the method, a hierarchical progressive reasoning framework combining a basic relation layer, a business relation layer and a dynamic evolution layer is constructed, so that the limitation of a traditional knowledge graph in the aspects of dynamics and business adaptability is effectively solved. Not only is the capture capability of static association enhanced, but also domain characteristics and implicit logic are deeply analyzed, and the reasoning accuracy in scenes such as scientific research supply chain optimization and the like is remarkably improved.

The invention provides a multi-region interconnection system carbon emission factor calculation method considering multivariate heterogeneous source load access. The method comprises the steps that a topological structure of a multi-region interconnection system, operation parameters of all source load devices and carbon emission basic data are acquired; calculating a carbon emission factor of the energy storage element by adopting an electric quantity pack sequence model, dynamically adjusting an electric quantity pack sequence according to a magnitude relationship between a discharge demand and the extracted electric quantity of the electric quantity pack, and determining a discharge carbon emission factor based on the carbon emission factor corresponding to the extracted electric quantity pack; based on the real-time carbon emission intensity of the system, a flexible allocation coefficient is dynamically generated through a nonlinear correlation mechanism, and the carbon emission responsibility corresponding to the power loss power in the region is bidirectionally allocated to a load side and a power supply side; and according to the distribution result of the network loss carbon emission responsibility, constructing a virtual lossless network, and based on the virtual lossless network, correcting the carbon emission factors of each node between the regions and in the regions and outputting the carbon emission factors.

The invention relates to the technical field of large model application, and discloses a data visualization narrating method and system based on a large model. The method comprises the steps of obtaining structured original data in an ERP system, a CRM system and IoT equipment; generating a unified event set; performing synchronous identification of cross-source events; constructing a heterogeneous event graph; outputting a causal chain set based on a graph attention path decoding model; and outputting the visual narrative template and the display instruction set. In the prior art, causal modeling is carried out mainly depending on a single-mode data flow or a static business rule, and especially under the condition that an enterprise has multi-business system collaboration and semantic description is inconsistent, high-credibility causal chain reconstruction and structured narrative output cannot be realized. Due to the fact that a graph attention mechanism driven path decoding and cross-modal event recognition mechanism is introduced, dynamic causal path extraction and visual structured reconstruction of complex business events between heterogeneous sources are achieved.

A system for determining a degree of dependence on screens of a user, includingat least one software module embedded in a set of information processing apparatuses, adapted for collecting first measurements about the behavior of the applications installed on said information processing apparatuses and second measurements about the behavior of said information processing apparatuses;a database for storing said measurements;a server for periodically determining an aggregation of said measurements and a degree of dependence on screens according to said aggregation.

The invention provides a natural reserve intelligent checking method and system based on multi-source data driving, and relates to the technical field of environment monitoring and information fusion, and the system comprises a data collection unit which is used for asynchronously obtaining and caching original monitoring data, and a self-adaptive preprocessing unit which is connected with the data collection unit and is used for processing the original monitoring data. The self-adaptive preprocessing unit is used for executing dynamic quality evaluation, weighted filtering based on quality scores, unified space-time calibration and multi-modal tensor formatting on the original monitoring data, and the fusion engine unit is connected with the self-adaptive preprocessing unit. According to the natural reserve intelligent checking method and system based on multi-source data driving, quality self-evaluation and dynamic weighting of heterogeneous source data are realized for the first time through cross-modal embedding fusion of self-adaptive quality evaluation and contrast learning driving, and fusion bottlenecks caused by different data formats, precision and timeliness differences are effectively eliminated.

This invention discloses a heterogeneous extension component concurrent loading and dependency coordination architecture. Through concurrent acquisition of heterogeneous sources (remote repositories, temporary injection, pre-embedded components, dynamic inlining), declarative pattern verification of extension component descriptors, multiple heterogeneous source type adapters, offline resource discovery, dependency coordination and degradation, runtime configuration optimization, hierarchical exception management, adaptive dependency resolution, intelligent hot updates, sandbox isolation, dynamic capability negotiation, and predictive preloading, it achieves an efficient, reliable, and secure extension system. The system supports trusted source authentication, access control decisions, and functional unit extraction, and is suitable for command-line tools, integrated development environment extension systems, and extensible application frameworks. This invention effectively solves the shortcomings of existing extension component loading mechanisms in terms of heterogeneous source support, dependency coordination, security isolation, and loading efficiency, and can be widely applied in plug-in systems, modular applications, and microservice architectures.

The embodiment of the specification provides a hyperspectral image cross-domain small sample classification method and system based on heterogeneous data, wherein the method comprises the following steps: first stage training of a spatial feature extraction backbone network based on a metric-based meta-learning method by using labeled natural images in a heterogeneous source domain, extracting spatial features of the natural images to obtain a spatial feature extractor; second stage training of a spatial-spectral feature extraction network composed of the spatial feature extractor and a spectral feature extractor by using labeled hyperspectral images in the heterogeneous source domain and a target domain by adopting a small sample classification method FSL, obtaining spatial-spectral features of the hyperspectral image HSI which is domain adaptive, and obtaining a spatial-spectral feature extractor; and classifying unlabelled hyperspectral images in the target domain by using the spatial-spectral feature extractor.