86 results about "Information integration" patented technology

This invention discloses a method, device, terminal, and storage medium for sensing operating system fault information, belonging to the field of operating systems. The method includes: deploying a resident lightweight resource detection tool to detect external manifestations of anomalies; when anomalies appear, activating an information collection daemon to periodically collect system resource information using a small amount of resources, closing the resident lightweight resource detection tool, and generating system resource trend information; using the system resource trend information to determine if there is an abnormal trend deterioration; when there is an abnormal trend deterioration, selecting and loading a corresponding medium-weight resource information collection plugin based on the system resource information corresponding to the abnormal trend; using the medium-weight resource information collection plugin to collect system information and recording the system information. This method can simultaneously ensure resource effectiveness and information integrity.

The application provides an API information processing method and an electronic device, wherein the method comprises: fusing abstract syntax tree analysis, large language model and retrieval enhancement generation technology, obtaining code fragments and a vector library by analyzing target application source code, extracting a route group and constructing a route mounting tree representing hierarchical relationships by using a large language model, determining complete target API information in combination with the route mounting tree and the extracted initial API information, and finally generating a function description through semantic retrieval of the vector library to associate related code fragments. The API information extraction and description generation full process are automated, the API information integrity and the accuracy and richness of the function description are improved, the manual maintenance burden is effectively reduced, and the application scenarios and business change scenarios lacking API document maintenance are adapted.

The application discloses a kind of distributed polarization navigation sensors and polarization navigation systems based on optical fiber light guide, it is related to bionic polarization navigation field.The sensor includes the front stage distributed polarization detection unit, optical fiber transmission unit, rear stage information processing unit, information integration unit and centralized information processing module connected in sequence.The front stage distributed polarization detection unit can be installed on the external surface of carrier, for collecting polarized light signal;Optical fiber transmission unit realizes the optical signal transmission between front stage and rear stage unit;Subsequently, data processing and solution are completed by information processing, information integration unit and module.The application separates the front stage and rear stage of polarization navigation sensor, miniaturizes the front stage unit, can be lightweighted on the external surface of carrier, effectively saves space.The method proposed in the application can make the whole sensor realize distributed polarization sensing and assist carrier orientation, navigation, with the characteristics of anti-electromagnetic interference, flexible structure installation, strong environmental adaptability.

The method for analyzing incomplete channel EEG signals based on adaptive orthogonal regression belongs to the field of EEG signal analysis and pattern recognition technology. This invention proposes an embedded feature selection framework for incomplete channel EEG data, enabling robust modeling directly without channel completion and effectively reducing the impact of missing channels on the analysis results. By introducing a channel missingness perception mechanism and an adaptive channel weighting strategy, this invention achieves dynamic modeling of the contribution of information from different channels, significantly improving the model's stability in multi-channel and high-missing-rate scenarios. The orthogonal regression-based feature selection mechanism can maintain the integrity of discriminative information while suppressing redundant features, making it suitable for high-dimensional EEG data analysis tasks. This method has good versatility and can be adapted to different channel scales (such as few channels and high-density channels) and single-modal or multi-modal fusion scenarios, possessing high engineering application value.

The application provides a kind of industrial vertical field corpus construction data cleaning method, belong to industrial big data technical field, the application is through collecting multi-modal industrial data and recording original time stamp, unified time reference is generated using dynamic time warping algorithm for time alignment, cross-modal fusion feature vector is generated by multi-modal attention mechanism learning feature interaction weight, key data section is identified based on event-driven trigger rule and is associated as event data package, quality evaluation and abnormal calibration are carried out through step response consistency detection, multi-modal knowledge graph is constructed and cross-modal alignment model based on dual optimization constraint satisfaction framework is used for feature completion, finally, data quality is ensured through adaptive verification process, solve the technical problems that multi-modal industrial data is difficult to ensure alignment accuracy and original information integrity in time dimension alignment and semantic level fusion process.

This invention discloses an intelligent dynamic integration method and system for mobile office applications, addressing issues such as fragmented mobile office applications, inefficient information integration, and low recommendation matching accuracy. The method includes collecting user attributes and application usage data; establishing a user attribute mapping layer with a hash table structure and assigning initial weights to applications; constructing a neural network recommendation model with a multilayer perceptron structure; obtaining real-time application weights after supervised learning training and inputting updated data; sorting and grouping applications according to their weights; and then accurately pushing applications to users through the mapping layer. The system includes corresponding acquisition, mapping, modeling, classification, and push units, and also provides electronic devices and storage media for implementing the method. This invention achieves personalized and accurate matching between applications and user needs, improves the real-time performance and stability of recommendations, adapts to dynamic changes in enterprise organizational structure and user needs, and significantly improves mobile office efficiency.

The application provides an improved optical remote sensing image automobile target detection method based on YOLOv10, which improves detection precision and robustness through multi-stage network optimization. First, a remote sensing automobile dataset containing accurate labeling and data enhancement is constructed, and the DOTAv2.0 dataset is used for preprocessing and enhancement processing. In the model architecture, an improved PSA-INEMA module is introduced into the backbone network PSA structure, the layer normalization (LN) is used to replace the group normalization (GN) and the residual connection is fused, the ability of micro target feature reservation and cross-channel information integration is enhanced, the dynamic interactive dual attention mechanism (DIDA) module is embedded in the key feature fusion node at the Neck end, the spatial and channel attention map dynamic interaction fusion is used to improve the automobile target positioning precision, and the GMIn-IoU hybrid loss function is introduced at the Head end to replace the CIoU, the dynamic non-monotonic focusing mechanism is combined to optimize the gradient distribution, and the small target detection performance is improved.

The application discloses a kind of based on agent long-term memory and optimization search's streaming video compression system, including: memory graph processing module, no edge minimum maximum sampling module and edge perception weighted pruning module and time decay memory search module, the structural characteristics of the present application based on memory graph, text node in long-term memory is distinguished according to whether it is connected with entity node, and differential compression strategy is used, effectively removes redundant memory node, while maintaining the integrity of key information is significantly reduced in storage, based on time decay memory search mechanism, different time segments of memory are given dynamic weight, pay more attention to recent memory more relevant to current query in search stage, so as to alleviate the performance loss caused by memory compression, while improve retrieval efficiency.Through the above technical scheme, the application realizes efficient compression and fast retrieval of agent long-term memory under the premise of guaranteeing streaming video understanding accuracy, significantly improves the real-time performance and scalability of the system.

The application relates to the technical field of transformer monitoring, and discloses a main transformer multi-dimensional state online monitoring method based on big data analysis, which comprises the following steps: mapping a transformer body into a knowledge graph node, dynamically calculating an edge weight, obtaining a dynamic knowledge graph, combining the dynamic knowledge graph with a three-dimensional data matrix, generating an evolution rule set and a parameter threshold, reasoning on the basis of the evolution rule set, and outputting a rule priority. Through multi-dimensional data fusion and dynamic rule evolution, the application realizes high self-adaptation and accurate abnormality detection of main transformer state monitoring, and the core advantage lies in a flexible architecture which is not model-driven, so that the detection logic can be dynamically adjusted through a rule self-evolution system without relying on a fixed model, equipment aging and working condition evolution are adapted to, a full-parameter sensor network covers four core areas, working condition information is integrated, a three-dimensional data matrix is formed, and comprehensive data support is provided for abnormality detection.

The application provides a work injury audit management method and system, relates to the technical field of work injury audit, and the method comprises the following steps: information extraction: obtaining to-be-settled cost information, converting a cost list in the to-be-settled cost information into editable text by using an OCR technology, and extracting project information in the editable text; integrity judgment: judging whether the project information is complete, if yes, executing the step of auditing; if not, correcting the incomplete project information, obtaining a correction result, updating the correction result as the project information, and executing the step of auditing; auditing: auditing the project information according to a preset operation path by using an RPA robot, obtaining an auditing result, and sending an alarm signal when the auditing result does not meet the expectation. The application can improve the efficiency of the work injury audit process.

This application provides a method and system for optimizing user input information based on a large language model, belonging to the fields of artificial intelligence and regional management technology. The method receives initial user information and extracts key semantic features such as entities and actions through deep learning algorithms. An expanded semantic set is generated based on a dynamically updated semantic association rule base, and a precisely targeted subset is obtained through a specially trained large language model. The initial information and the subset are fused according to relevance weights, and optimized information is output after verification and correction through syntactic analysis and logical reasoning. Parameters can be dynamically adjusted based on user feedback. This invention solves the problems of low accuracy and insufficient standardization in existing technologies, improves information completeness and adaptability, and enhances the usability of user input information.

This invention relates to the field of network communication and online conferencing technology, and discloses a method and system for compensating for core information lost during conferencing. After a conferencing terminal experiences a network outage and reconnects, the server determines the compensation range based on the outage information reported by the terminal, processes the cached conferencing data from the outage period, and extracts the key information most impactful on conferencing continuity through a multicast and unicast collaborative transmission mechanism and a core information filtering model based on semantic similarity and time decay. This information is then encapsulated into a data compensation packet and sent to the terminal, achieving rapid compensation and synchronization of core content. Compared with existing technologies, this invention can effectively reduce redundant data transmission under bandwidth-constrained conditions, improve conferencing continuity and information integrity after network recovery, and enhance system stability and user experience.

Provide a system. 【Solution means】 An information collection device that receives information regarding budget and interests from a user, An information analysis device that analyzes the collected information regarding budget and interests, A candidate selection device that selects a plurality of candidate destinations based on the analysis result, A plan generation device that generates a travel plan adjusted based on the selected candidate destinations, An evaluation device that aggregates evaluation information of other users related to the generated travel plan and evaluates the plan, An information integration device that integrates the latest information related to the destination and travel period, A display device that integrates this information and displays a travel plan suitable for the user, A smart device application that provides a real-time and customized travel plan based on the user's interests and stay period, A system including the above.

Systems and methods are directed to incorporating large language model (LLM) prompts into graph queries using graph query language and executing these graph queries. The system receives a graph query that includes a LLM prompt in the graph query language. The LLM prompt comprises an input context and a query to be answered. Based on the graph query, the system traverses a knowledge graph to obtain context data associated with the input context. The context data may be constrained by access control policies of an enterprise providing the knowledge graph. A context table may be generated using the context data based on the input context indicated in the LLM prompt. The system then invokes an LLM to determine a result by applying the query of the LLM prompt to the context data obtained from the traversal of the knowledge graph. The result is then presented on a user interface of a client device.

This invention discloses a method and system for constructing image generation anchor points with multi-dimensional constraint information, belonging to the field of multimodal ultra-long text generation technology. First, it receives a reference file and generation instructions, parses the text semantics and chapter structure, locates the image insertion position, and extracts the corresponding text fragments. Then, based on the text fragments, it refines the image requirements and constructs standardized anchor point tags containing unique identifiers, types, styles, contexts, and detailed requirements. Subsequently, it performs semantic matching and information integrity verification on the anchor points, iteratively optimizing substandard tags. Qualified anchor points are inserted into designated positions, the anchor point information is parsed, and a multimodal model is scheduled to generate images. Based on the anchor point identifiers, precise replacements are made to form complete text with images. Finally, it evaluates the fit between the image and the anchor point requirements, optimizing and redrawing substandard images until they are qualified. This invention is applicable to ultra-long texts such as proposals, papers, and reports, and can improve the accuracy and efficiency of multimodal text generation.

The application discloses a compliance file generation method based on OpenXml analysis and template deep configuration, and relates to the technical field of file generation.The compliance file generation method based on OpenXml analysis and template deep configuration comprises the following steps: S1, collecting and preprocessing structure mapping data, node positioning data and binding rule data in a file audit process, and constructing a standardized audit state data set; S2, analyzing the information integrity of the structured audit, and adjusting an audit description generation process; S3, evaluating the mapping redundancy degree of the audit chain, and adjusting a redundancy writing strategy; S4, evaluating the consistency of the audit data, and identifying the matching accuracy of the annotation insertion and the structure binding relationship; and S5, completing interface annotation reservation and audit content embedding, and generating a compliance file.The method solves the problem that the audit information recording mechanism lacks unified mapping capability of the structure layer and the interface layer, and the audit chain is prone to be broken, so that the whole-process audit management cannot be supported.

This application discloses a network connection decision method, apparatus, device, and medium based on NAT state, applied to a network server and relating to the field of vehicle diagnostic technology. The method includes: obtaining a first message sent by a target device based on the monitoring results of a first socket and a second socket on the network server; if both the first socket and the second socket receive the same device serial number and socket number, determining whether the public network ports of the devices obtained by each socket are consistent, determining the target NAT type based on the determination result, and returning a response message to the target device within a preset time period; if both the first device and the second device meet the target information integrity condition, sending a second message to each device based on the target NAT type, so that the device can perform network connection operations based on the second message. This method improves the efficiency of UDP hole punching at low cost and quickly enables communication between the two devices.

The present application relates to the technical field of information entry, and discloses a multi-terminal adaptive labeling system for preoperative information entry of children, which comprises a multi-terminal interaction module for receiving preoperative basic information of a single child from various terminals; an integrity checking module for checking whether the preoperative basic information entered by any terminal is missing, and if so, generating a corresponding preoperative core information dimension-specific completion prompt until the preoperative complete information of the single child is obtained; an adaptive labeling module for determining an adaptive labeling granularity based on the type of the terminal for entering the preoperative complete information, and dynamically matching and labeling non-professional entry information in the preoperative complete information of the single child based on a child specialty terminology library and the adaptive labeling granularity; and a secondary verification module for triggering a secondary verification prompt based on structured evaluation items and cross-terminal entry conflict information until the preoperative standard information is obtained; thus, the quality of preoperative information of children is ensured, and reliable basis is provided for preoperative evaluation of children.

This invention proposes a housing information integration system. Considering the data processing load exceeding that of small shops, and to facilitate the migration of large and small shops, this invention introduces a jump-type housing information management module, which is superior to existing technologies. By taking over the housing information of specific small shops within a specific time period, this module avoids data processing confusion during the migration process and prevents business processing interruptions that may occur due to the migration. It also avoids the technical problem of increased business data load on the original regional control center caused by the management of large and small shops, resulting in the loss of new housing information and business data processing failures. At the same time, this invention introduces a token holding judgment for the jump storage area to realize the specific fixed storage requirements of specific small shops, introducing adaptive functions to system management and avoiding system downtime caused by the uniform management of specific high-throughput small shops.

The application discloses a kind of multi-agent (Agent) dynamic cooperation and information aggregation method for complex query task, the method includes: by extracting the key semantic feature vector of query text, similarity matching is carried out with pre-set field feature library, accurate field recognition and shunt are realized;For real estate field query, further identify subtask type, dynamically load the most suitable professional agent from special sub-agent pool;Agent detects parameter gap automatically during execution, and obtains complete parameters through multiple rounds of directional interrogation;Finally, call professional tools to obtain original information, and carry out semantic deduplication, conflict detection and multi-source fusion, to generate comprehensive answer.The application realizes the specialization of complex query processing, intelligent interaction, efficient dynamic scheduling and powerful information integration, and improves the processing efficiency and answer quality of complex query, especially professional field query.

This invention discloses a multi-channel reversible information hiding method and system, relating to the field of information security technology. The method includes a carrier partitioning and data embedding stage, and a data extraction and carrier recovery stage. In the carrier partitioning and data embedding stage, the sender partitions the original binary image carrier into multiple sub-binary images and the secret data into multiple sub-secret data. The sub-secret data is then embedded into the corresponding sub-binary images to generate a coded sub-binary image, which is transmitted to the receiver via multiple channels. In the data extraction and carrier recovery stage, the receiver can losslessly recover the original binary image and secret data using any number of coded sub-binary images. This invention improves the fault tolerance of information transmission by transmitting coded sub-carriers through multiple channels, solving the problem that existing single-channel reversible information hiding methods are susceptible to interference leading to data extraction and carrier recovery failures. It is suitable for multimedia communication scenarios with high requirements for information integrity and security.

The application discloses a low-illumination multi-modal scene understanding method based on dynamic feature selection complementarity, and belongs to the technical field of artificial intelligence, wherein according to the complementary characteristics of multi-modal feature information, a parallel feature extraction network based on a Transformer is constructed; a short-distance selective attention module is constructed to reduce information interference between the same modalities and ensure the reservation of complementary information under different modalities; a channel selective self-attention module is constructed to enhance the information integrity of the feature channel dimension; and a lightweight fully-connected multi-layer perception decoding module is constructed to receive the feature information of the encoding network and perform feature reconstruction on feature maps of different scales; the application can fully utilize the complementary advantages of multi-modal feature information, avoid the insufficient low-illumination scene understanding ability caused by single-modal information, and be applied to scene analysis of unmanned systems and unmanned navigation systems under complex low-illumination scenes.

This application provides a multi-source stability collaborative compensation method and apparatus for a panel scanning system. Through innovative data processing system design, it achieves effective information integration via parameter analysis and feature extraction. A compensation model mechanism is constructed, combining mapping analysis and rule formulation to establish a reliable correction strategy. Dynamic optimization is introduced, ensuring imaging stability through multi-dimensional correction and collaborative compensation. This method effectively addresses the shortcomings of traditional technologies in data processing, model construction, and dynamic compensation, providing technical assurance for panel scanning.

We provide the system. [Solution] Information gathering means for collecting information on machinery within a factory, Information integration means for integrating and centrally managing the information collected by the aforementioned information collection means, An analysis means that analyzes the information managed by the aforementioned information integration means and proposes the arrangement of machines and a work plan, A presentation means that presents the optimal arrangement and work plan for the machine based on the analyzed results, A system that includes this.

This invention discloses a customer intelligent recommendation method, apparatus, and computer equipment based on a large-scale language model. The method includes: acquiring users' natural language business opportunity needs and converting them into text information in a unified format; using a large-scale language model to perform deep analysis of the text information, extracting user need types and related core business entities; judging the information completeness of the preliminary structured information based on the large-scale language model's generation capabilities, and supplementing missing or ambiguous information through multi-round interactive dialogue to obtain structured data; converting the structured data into a domain-specific DSL query, verifying it, and sending it to a downstream system to filter and match customers, and generating matching results; displaying the results in a user-friendly manner through a customizable wbit module, and providing visual annotations. Implementing this invention reduces manual intervention and improves processing efficiency and accuracy.

The application discloses a shared bicycle demand prediction method based on a geographical distance enhanced space-time graph neural network, belongs to the technical field of bicycle demand prediction, and inputs bicycle data cleaned and aggregated into a double-branch framework integrating local and global space-time information, which is composed of a local information branch and a global information branch. In the local branch, the distance information between stations is introduced, a multi-scale Gaussian kernel is designed for processing, and is added to the calculation of attention scores to solve the deficiencies of existing models in global and local information integration and real geographical distance information extraction. In the global information branch, global spatial patterns and long-term time trends are extracted through the combination of GAT and TCN, and information is further fused through shared GNN and shared MLP to comprehensively integrate global space-time information, better grasp the overall change rule of bicycle demand in time and space, and improve the prediction accuracy and stability.

The application relates to the technical field of fault positioning, and discloses an intelligent cable fault positioning method and device for a photovoltaic system. The method comprises the following steps: obtaining a feature data matrix containing voltage difference between components, string current change rate, component power value and adjacent component voltage ratio by performing characteristic parameter calculation and processing on component-level synchronous sampling data; obtaining component abnormal state marking data by performing multi-dimensional threshold analysis and sliding variance analysis processing on the feature data matrix; obtaining cable fault type judgment data by performing decision tree classification processing on the component abnormal state marking data; obtaining cable fault point position data by performing voltage gradient analysis processing on the region where the cable fault type judgment data is located; and performing fault information integration processing on the cable fault point position data and the cable fault type judgment data to obtain cable fault diagnosis data. The application improves the efficiency and accuracy of intelligent cable fault positioning of the photovoltaic system.

The application provides a DXF file generation method and device, electronic equipment and a storage medium. The method comprises the following steps: inputting an accident scene picture and a corresponding text prompt into a preset multi-modal large model to generate a target vector mask containing target contour coordinates and semantic labels; converting the target vector mask into SVG code with semantic labels and physical attributes according to the target contour coordinates and the semantic labels; creating a corresponding DXF layer through the semantic labels, converting the graphic elements into corresponding DXF primitives, and inheriting the physical attributes as metadata to the DXF file to convert the SVG code into a DXF file. The application realizes end-to-end automatic generation from an accident scene image to a standard DXF engineering drawing, and effectively solves the deficiencies of traditional methods in efficiency, accuracy and information integrity.

This invention discloses a method and apparatus for dynamic configuration of a fusion tree based on a virtual interface, belonging to the field of multi-sensor information integration. The method includes: Step 1, performing an uncertain mapping of sensor measurement parameters using a virtual sensor interface, where the sensor measurement parameters include target motion feature parameters and target recognition feature parameters; Step 2, setting fusion nodes after the virtual sensor interface mapping, and dynamically configuring the fusion tree using processor-run program code; Step 3, storing the dynamically constructed fusion tree in memory using a neighbor table, where each fusion tree is treated as a directed graph, and any fusion tree structure is described by a unique neighbor table; the neighbor table specifies the input sensor combination used by each fusion node and the order of these combinations, thus completing the fusion tree configuration based on the virtual sensor interface for different task and multi-situational requirements. This invention's configuration method is unaffected by changes in the sensor interface and has high fusion efficiency.