202 results about "Graph analytics" patented technology

A system and method for real-time team intent modeling using persistent cognitive machines with federated human profiles which processes individual team member behavioral signals through geometric intent analyzers that generate high-dimensional vector representations of individual objectives and preferences. A team intent orchestrator aggregates individual vectors into collective representations within a dynamic geometric manifold that evolves based on team coordination patterns. Federated human profiles enable privacy-preserving knowledge sharing across teams through geometric abstraction techniques that preserve coordination utility while protecting individual privacy. The system implements proactive conflict detection through trajectory analysis that identifies potential coordination issues before performance impact, and provides real-time synchronization mechanisms that maintain team coordination coherence despite individual behavioral changes. Cross-team learning capabilities enable organizational intelligence development through pattern abstraction and context-aware adaptation of successful coordination strategies. The persistent cognitive architecture maintains coordination patterns across sessions and team composition changes, enabling continuous improvement through accumulated team experience.

The invention provides a social robot intention recognition system and method based on heterogeneous graph analysis, and the method comprises the steps: carrying out the unified modeling of multi-type nodes and multiple relation edges based on a heterogeneous graph structure, and systematically capturing the complex multi-dimensional behavior association in a social platform; according to the method, technologies such as a large language model and a knowledge graph are fused, fine recognition and classification of multiple types of behavior modes are supported, and meanwhile, hierarchical evaluation for different organized risks is realized by combining account behavior aggregation and emotional polarity analysis, so that a decision basis is provided for supervision and governance; according to the method, automatic collection and preprocessing of multi-source heterogeneous data and heterogeneous graph construction processes are integrated, manual intervention and repeated work are reduced to the maximum extent, the overall data processing efficiency and response speed are improved, and the real-time analysis requirement in a complex dynamic social environment is met; transparent analysis of robot control sites and behavior strategies is supported, and the interpretability of the system, the user credibility and the executive force of policy supervision are greatly improved.

A system and method for privacy-preserving communication validation and multi-attribute trust scoring is disclosed. The system analyzes communication metadata to determine pattern legitimacy by comparing current communication patterns against relationship fingerprints without accessing communication content. The system validates relationship context between communicating parties using interaction graph analysis and historical communication data. Cross-network intelligence correlation compares current patterns against aggregated patterns across voice, email, and messaging services, creating a self-strengthening security framework that recognizes emerging threat patterns while validating legitimate communication behaviors. The system generates comprehensive multi-attribute trust assessments comprising individual trust attribute scores including engagement rate, reliability index, channel preference, temporal pattern, and behavioral pattern, combined into overall trust levels. Trust context is displayed through a user interface presenting simplified, intuitive, and actionable information with progressive disclosure capabilities, enabling informed user decisions while preserving privacy. Communication processing actions provide users with appropriate engagement options tailored to specific trust assessment results.

The invention relates to an open source component multi-modal dependency risk tracing method and device, and the method comprises the steps: employing a mode of combining static analysis and dynamic analysis to analyze a component dependency relationship of software, and combining AST analysis and construction script analysis; function-level features are extracted, a binary fingerprint database is constructed, the similarity between different versions is analyzed through LSH, behavior patterns in binary codes are analyzed, and hidden dependencies or malicious code injection is detected; the version updating history of the dependent component is analyzed and monitored by using a time sequence, and the vulnerability security of the component is evaluated in combination with attack graph analysis; high-risk components on the path are calculated, potential supply chain attack points are identified, and risk points are subjected to cross validation; a time sequence diagram database is used for recording the component dependency relationship, and time backtracking query is supported. According to the method, a multi-mode dependency analysis method is adopted, potential dependency risks are rapidly identified, and potential supply chain attack risks are timely warned.

The invention relates to an electrical variable adjustment monitoring method and system for a distribution box, and the method comprises the steps: collecting the current, voltage, power factor and other core electrical variables of each distribution box in a region in real time, triggering an abnormal mark through dynamic gradient analysis, and recording time-space information; performing space-time correlation analysis based on the power distribution network topological graph, calculating an abnormal time concentration ratio and a spatial distribution density, and generating a correlation circle; determining a fault type (a local problem or an overall power grid problem) according to the characteristics of the associated circles, the number of branch lines and the change of electrical variables, and matching a corresponding processing scheme; and continuously monitoring a fault processing result, and generating a repair report until the system is stable. The system comprises a core electric variable real-time acquisition module, a power distribution network topological graph analysis module, a fault type judgment module and a fault processing monitoring module. According to the method, the anomaly detection accuracy and the fault processing efficiency are improved, and reliable operation of the power distribution network is guaranteed.

The invention relates to a cloud-edge collaborative medical image intelligent diagnosis method and device, equipment and a medium. The method comprises the following steps: firstly, obtaining original image data, and carrying out calibration processing on the original image data through a preset multi-modal image standardization model to obtain a standard image data set; performing feature extraction and grading processing on the standard image data set at an edge end to obtain a grading feature set, transmitting the grading feature set to a cloud end through a bandwidth allocation strategy, and generating a cloud end receiving feature subset; constructing a feature expression matrix based on the cloud receiving feature subset, using a network topology structure to carry out graph analysis, and extracting a path to generate a diagnosis path set; and finally, feedback diagnosis information is formed based on the diagnosis path set, and a task allocation optimization result is obtained by dynamically adjusting a cloud side task allocation proportion. According to the method, efficient processing and diagnosis optimization of medical image data are realized, efficient collaboration of cloud edge resources is ensured, timeliness and accuracy of medical image diagnosis are effectively improved, and resource scheduling requirements in different scenes are met at the same time.

The invention relates to the technical field of attribute graph collection and release, in particular to an attribute graph collection and release method, system and device based on local differential privacy and a medium. Receiving disturbed data from each client; based on all disturbed data, estimating community division of the original graph structure and triangular counting of each node; obtaining a target graph structure based on community division and triangular counting; and on the basis of the target graph structure and the disturbed data, node attributes are optimized, so that node attribute distribution and community division are kept consistent, and a final publishable target attribute graph is obtained. The method has the advantages that strict requirements of local differential privacy are met on the privacy protection level, and multi-dimensional features such as a community structure, triangular counting and node attribute homogeneity are reserved in the published attribute graph at the same time; and the practical value of the synthetic data in applications such as downstream graph analysis, community discovery and personalized service is greatly improved.

The invention discloses a building plane element recognition method based on a graph neural network and a convolutional neural network, and belongs to the technical field of building plane element recognition. The identification method comprises the following steps: inputting a building plane pixel image, and carrying out preprocessing through image layer screening and information extraction to obtain a binary pixel image; vectorizing the binarized pixel image to obtain a vectorized image; performing image segmentation on the vectorized image; using the segmented vectorized image to construct a regional adjacency graph; performing regional adjacency graph optimization on the regional adjacency graph through graph analysis and regional merging by using a graph neural network, and identifying a room edge; calculating the area of the room; using the convolutional neural network image recognition model to recognize articles in the room; and performing room function prediction according to an identification result. According to the method, the problem of shape blurring caused by convolution in a traditional method is avoided. The graph neural network model training effect is improved, and the classification accuracy is improved. The method is suitable for various planar graph styles and is high in universality.

The invention discloses an LCD display screen production whole-process collaborative management method and system based on the Internet of Things, particularly relates to the technical field of production process collaborative management, and is used for solving the problems of circular waiting and resource stiffness caused by lack of global resource allocation dependency relationship analysis in a complex production scene in an existing collaborative management method. Equipment state data and material position data of each process are acquired in real time, a resource allocation relation graph with production equipment as resource nodes and production tasks as task nodes is constructed, a resource access sequence of each production task is analyzed to identify a circular waiting path, and a process association network and a production flow network are combined to evaluate a blockage propagation risk. And finally, on the basis of the relief urgency degree, the production tasks are selected to carry out resource reallocation so as to realize the relief of the circular waiting path, and the overall efficiency of the production line and the resource utilization rate are improved.

The invention discloses an iterative collaborative learning method for node characterization and a topological structure, which comprises the following steps: acquiring initial graph data containing a node set and an initial edge set, and initializing node characterization; constructing a dynamic edge predictor, calculating the structure confidence degree between the node pairs based on the current node representation, generating a candidate edge weight matrix, and obtaining an optimized adjacent structure through a sparsification strategy; inputting the optimized adjacent structure into a graph neural network encoder, performing information propagation and aggregation in combination with the node feature matrix, and outputting enhanced node representation; the enhanced node representation is fed back to the dynamic edge predictor, and the edge weight and the node representation are iteratively updated until the convergence condition is met; and outputting a final optimized graph structure and node representation for downstream graph analysis. According to the method, through iterative collaborative optimization of the graph structure and the node representation, the graph topology can be dynamically corrected, the discrimination capability of the node representation is enhanced, the method is suitable for various complex scenes such as a noise graph and a sparse graph, and the performance of a downstream graph analysis task is improved.

The invention provides a multilevel modular personal memory large model and a construction method thereof. The model comprises a receiving module for receiving information input by a user, converting the information into a unified text and language vector, storing a personal historical language vector and context information, extracting key features, analyzing intention, retrieving by using Elasticsearch and returning related context information, and generating a reply according to the related context information. According to the method, the multi-modal data processing capability is integrated, the traditional limitation is broken through, and image, voice and text associated storage is supported; constructing a dynamic memory enhancement mechanism to prevent important information from being covered; then an intention analysis module is innovated to accurately capture deep demands; the correlation and the adaptability of memory fragments are improved by adopting a multi-modal retrieval and graph structure correlation technology; and finally, through a language habit adaptation and logical reasoning module, the output conforms to the style of the user, potential requirements are deduced, a whole-process closed loop is formed, and breakthrough is achieved in the aspects of memory management, intention understanding and interaction naturalness.

Methods, systems, and storage media for generating contextually relevant text from image descriptions and user intent are disclosed. Exemplary implementations may: receive an image and a user-defined intent for text output; analyze the received image to generate a contextual description of the image; generate a query based on the contextual description of the image and the user-defined intent; and generate the text output based on the query.

The invention discloses an Internet of Things real-time task distribution system based on AI scheduling, and the system comprises a data collection module which is used for generating a task input vector, and constructing a node resource vector set; the semantic modeling module is used for inputting the task input vector into an improved StructBERT model to construct a task semantic vector; the intention analysis module is used for constructing a historical task intention vector library and a semantic conflict graph; the comparison judgment module is used for executing task conflict judgment and scheduling priority adjustment operation according to the task aggregation degree and the semantic difference vector; and the scheduling execution module is used for executing a scheduling calculation operation, generating a task allocation result and updating the historical task intention vector library and the semantic conflict graph. The method has the advantages of high task semantic understanding precision, high conflict judgment efficiency and high scheduling decision intelligent degree, and is suitable for a real-time task allocation scene in a multi-node heterogeneous Internet of Things environment.

The invention discloses an education content recommendation method and system based on group intelligence, and the method comprises the steps: constructing a user dynamic cognition knowledge graph, and carrying out the dynamic attenuation of the memory intensity attribute of a knowledge point node based on a cognition forgetting model in combination with the memory stability; defining utilization pheromones on edges of the knowledge graph, and strategically injecting exploration pheromones according to forgetting states of users or knowledge breadth indexes; guiding a swarm intelligence algorithm to construct a learning path according to a rule jointly influenced by bimodal pheromones; and finally determining an optimal path and generating recommendation. According to the method, accurate intervention on a learning process and intelligent guidance on knowledge exploration are realized through deep fusion of a recognition scientific model and a swarm intelligence algorithm of bimodal pheromones, the learning stability is remarkably improved, information cocoon houses are effectively broken, and the interpretability of recommendation is enhanced through path intention analysis.

The invention provides an electric vehicle battery health state assessment monitoring method, which comprises the following steps: carrying out image mapping processing on a protection plate depression degree index and a bracket stress concentration point displacement estimation value by adopting a convolutional neural network, and converting a protection plate deformation parameter into a two-dimensional thermodynamic diagram representation form, obtaining a correlation characteristic graph of stress transmission between the protection plate and the support; analyzing a time sequence change mode in the correlation characteristic graph of stress transmission between the protection plate and the support, predicting the accumulated influence of the deepened depression of the protection plate on the displacement of the support, if the predicted displacement exceeds a preset threshold value, triggering an adjustment signal, and judging the updating demand of a dynamic detection strategy; and obtaining reference parameters in a similar road surface bulge scene from a historical impact data set according to the adjustment signal, fusing the current vehicle speed and impact strength data, and determining an optimized sensor sampling frequency and a stress monitoring focus area.

The invention discloses an integrated energy low-carbon operation model construction method and device and a management and control system, and the method comprises the steps: analyzing the operation state and carbon emission condition of an integrated energy system based on a constructed system directed graph, and giving an operation constraint and a carbon emission constraint; determining a system action space and a system reward function based on the current operation state and the scheduling operation of the integrated energy system; fusing the initial deep learning model, the system action space and the system reward function, analyzing state distribution of expected cumulative reward values corresponding to different system actions and different system states, constructing a deep learning model based on a Bayesian neural network, training the deep learning model based on the Bayesian neural network until convergence, and obtaining a final deep learning model; and obtaining a low-carbon operation model. The deep learning model based on the Bayesian neural network is introduced, topology change detection is analyzed, addition and deletion of nodes and lines can be rapidly recognized, real-time updating is achieved, and the operation efficiency is improved while the stability of the comprehensive energy system is kept.

The invention discloses a high-risk call identification method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a to-be-matched call text, carrying out the preprocessing of the to-be-matched call text, removing called information, and reserving calling information; constructing an event evolution chain based on the calling information, extracting cue words, and calling a large language model to generate a to-be-matched event evolution chain; converting the to-be-matched event evolution chain into a to-be-matched event evolution diagram; based on a preset event evolution graph, performing multi-stage graph analysis processing on the to-be-matched event evolution graph to obtain a key node number, a key node matching rate and key node connectivity; and determining whether the to-be-matched call text is a high-risk call based on the key node number, the key node matching rate and the key node connectivity. According to the method, a large number of positive and negative labeling samples are not needed, the manual labeling cost is reduced, the fraud sequential logic is accurately captured, the limitation of a static map is overcome, and the high-risk call recognition accuracy and practicability are improved.

The invention provides an intelligent processing method and system from structured data to a text based on a natural language, and relates to the technical field of data processing.The method comprises the steps that the text chunk analysis process is optimized and adjusted according to analysis optimization parameters, sentences are segmented into non-overlapping phrases with syntactic function labels, and the text after chunk analysis is obtained; performing syntactic and semantic structure analysis on the text subjected to block analysis, and establishing a semantic association relationship between internal structures of sentences through component analysis, dependency analysis and semantic dependency graph analysis to obtain structured semantic information; performing multi-sentence logic association analysis on the structured semantic information on a chapter level to obtain semantic information of the whole chapter; and based on the semantic information of the overall chapter, obtaining a target natural language text by utilizing a pre-trained large language model. According to the method, the accuracy and fluency of conversion from the structured data to the text are improved.

The invention belongs to the technical field of nuclear power, and particularly relates to an asset risk assessment method and device based on an attack surface and graph centrality. According to the method provided by the invention, on the basis of traditional vulnerability severity assessment, an asset attack surface measurement and graph centrality analysis method is introduced, and vulnerability availability, attack path transmissibility and structure importance of assets in a network are organically fused, so that an asset risk assessment mechanism closer to a real threat environment is formed. Through the method, high-risk vulnerability assets can be identified, nodes having key effects on attack diffusion in the network can be highlighted, and accurate quantification and sorting of risks are realized. The attack surface is closely combined with the network structure; the attack graph analysis fully embodies the key node effect; and the vulnerability severity and the utilization probability are fully fused. And the asset risk level in the network environment can be reflected more accurately.

The invention discloses a computing power resource determination method, apparatus and device, a medium and a product. The method comprises the steps of obtaining real-time computing power resource data of a natural language demand and a resource library; performing intention analysis on the natural language demand based on the historical performance database, and determining a computing power task instruction; and on the basis of the real-time computing power resource data, performing dual-algorithm collaborative matching on the computing power task instruction, and determining a target resource combination meeting requirements from a resource library. According to the method, intention analysis is carried out on natural language requirements in combination with a historical performance database, fuzziness of intention analysis is eliminated from the source, cooperative matching is carried out through a bi-pass algorithm so as to reduce matching deviation of computing power resources, accurate analysis of the requirements is achieved, and then accurate matching of the computing power resources is achieved.

A method for detecting escalation paths in a cloud environment is provided. The method includes accessing a security graph representing cloud objects and their connections in the cloud environment; analyzing each cloud object to detect an escalation hop from a current cloud object to a next cloud object, wherein the analysis is based, in part, on a plurality of risk factors and reachability parameters determined for each cloud object; and marking the security graph with each identified escalation path in the security graph, wherein an escalation path is a collection of escalation hops from a source cloud object to a destination cloud object.

The invention relates to the field of network security, and particularly provides a security cooperation system and method in a hybrid cloud environment. The system comprises a unified security management module, a security solution cooperation module and a security function module. The unified security management module analyzes a self-healing algorithm through a policy graph, carries out graph modeling on a multi-cloud security policy, automatically identifies policy conflicts and drifts through graph difference analysis, and realizes continuous consistency and self-healing of a cross-cloud policy. And the security solution cooperation module is embedded into a reinforcement learning agent, dynamically decides an optimal deployment scheme of security resources according to a reward function by sensing a threat situation, a resource utilization rate and response delay, and realizes intelligent self-adaptive arrangement of protection resources. The security function module is used for micro-servitizing and containerizing security capability based on a cloud native technology, transparent injection is realized through a service grid, an automatic compliance repair and verification mechanism is built in, and a compliance closed loop for detection, repair and verification is formed. And the management efficiency and the protection accuracy of the security policy are improved.

The invention discloses a machine tool operation state prediction method based on edge calculation. The method comprises the steps of 1, collecting multi-source state parameter data of machine tool operation; 2, preprocessing the multi-source state parameter data in an edge node; 3, modeling and predicting the operation state of the machine tool through an improved TSMIXer network, and outputting a machine tool operation state predicted value and a time sequence response scoring graph; step 4, comparing the machine tool operation state prediction value with an operation state threshold value, generating a state evaluation label and an early warning level label, and analyzing a disturbance condition based on a time sequence response scoring graph; 5, executing a response control strategy according to the early warning level label, and collecting operation state feedback data; and 6, visually displaying the prediction result and the early warning information. According to the invention, the machine tool abnormity identification accuracy and the running state prediction real-time performance are improved.

The invention provides a honey matrix situation map analysis method and device, a medium and electronic equipment. The honey matrix situation map analysis method comprises the steps that collected data are processed to form honey matrix network data; constructing a honey matrix situation map based on the honey matrix network data; performing time sequence modeling on the feature sequence of the edge, performing enhancement by applying a multi-head attention mechanism to obtain an enhanced edge feature representation, calculating an importance score of the edge, and performing weighted summation on the enhanced edge feature representation of the direct connection edge of the node to obtain a dynamic traffic feature of the node; selecting a new neighbor node set according to the security state similarity of the node and the initial neighbor node thereof, and obtaining node feature representation based on the new neighbor node set; and according to the node feature representation and the dynamic flow feature of the node, generating a final representation of the node feature, and performing node security state prediction to obtain a security state category. By applying the method, multi-level and multi-dimensional expression of the node features can be realized, the richness of the node features is enhanced, and the sensing precision of the honey array network security situation is improved.

Mechanisms are provided for generating a cryptographic asset bill of materials of a source code. The mechanisms generate flow graph of the source code and execute a parsing and analyzing the source code based on a cryptographic asset knowledge base to identify an initial set of cryptographic artifacts referenced in the source code. The mechanisms execute, for each cryptographic asset in the initial set of cryptographic artifacts, a flow graph analysis to identify one or more dependent cryptographic artifacts to form sets of related cryptographic artifacts. In addition, the mechanisms generate, for each set of related cryptographic artifacts, a cryptographic asset, compile the generated cryptographic assets into a cryptographic bill of materials, and generate and output a report of the cryptographic bill of materials.

A computer-implemented method comprising: obtaining an asset graph representing a network, the asset graph comprising a plurality of nodes representing assets of the network and edges representing relationships between the assets; calculating, using the graph, a local connectivity metric for a node of the graph; based on the local connectivity metric, identifying a local pattern associated with a potential security risk; and identifying a candidate node or a candidate edge based on the identified local pattern associated with the potential security risk.

The invention relates to the technical field of intelligent analysis of drawings, in particular to an AI intelligent drawing auditing system for architectural drawings based on deep learning. The system comprises a drawing structured analysis module, a building component feature fusion module, a drawing violation drawing examination analysis module and a violation drawing examination optimization visual module, a drawing primitive data set, an attribute feature set and a standard rule base corresponding to a building engineering drawing can be obtained, and meanwhile a building component multi-dimensional feature matrix is generated; inputting the multi-dimensional feature matrix of the building component into a deep learning model for intelligent identification, compliance verification and drawing violation examination analysis, and generating a building drawing violation feature vector; based on the building drawing violation feature vector, violation point location aggregation analysis and reverse drawing checking optimization application are carried out, so that a corresponding building drawing intelligent drawing checking result is output, and a building drawing compliance thermal bitmap is displayed. According to the invention, intelligent and standardized review of the architectural engineering drawings can be realized, so that the drawing review efficiency of the architectural drawings can be improved.

The invention discloses a code abstract generation method and system based on hierarchical context awareness, and belongs to the field of natural language processing and software engineering. The method comprises the following steps: generating hierarchical annotations for a code warehouse based on a function call graph, firstly extracting a call relationship in the graph and eliminating a ring structure, and then classifying leaf nodes and generating annotations; a topological sequence from bottom to top is adopted in the core step, the importance score of a child node is calculated for each father node, child node annotations, class function descriptions and code bodies of the child nodes are fused, and a large language model is input to generate annotations till the whole graph is covered. The method is suitable for an automatic document generation scene of a complex software project, and through a technical path combining function call graph analysis, child node importance measurement and class function description, the accuracy and semantic consistency of code annotation are improved; the method can be widely applied to the fields of software maintenance, code understanding, intelligent development tools, automatic document generation of enterprise-level code warehouses and the like.

The invention discloses a Java decompiling method, device and equipment and a computer readable storage medium. The method comprises the following steps: extracting a class file, a resource file, a configuration file and a static resource by decompressing a Java deployment package; based on a predefined third-party library feature library, utilizing Hash comparison and package name / method signature feature matching to accurately classify the class file into a third-party library or a self-researched code; performing byte code analysis on the self-developed class file to generate structured data, and inputting the structured data into a customized large language model (based on byte code-source code parallel corpus training) to generate an initial source code; the optimization step comprises the steps of correcting errors by a static grammar checker and automatically supplementing annotations by a byte code constant pool; reconstructing a project directory structure based on the package name; a class dependency graph is generated through static call graph analysis; and finally, integrating the initial source code, the dependency graph and the configuration file, automatically generating an interface document, a framework document and a configuration document, and realizing complete reverse engineering of the deployment package.

The present application relates to the technical field of geographic information system, and more particularly to a kind of city land planning analysis method and system based on GIS, by integrating remote sensing, historical planning, mobile terminal and social media data, construct multi-scale spatiotemporal cube model, generate standardized city temporary land data, adopt adaptive feature extraction to identify temporary land, establish hierarchical classification system, and monitor abnormal change through time series model, construct the association diagram of temporary land and city function network, analyze interactive relationship, finally propose differentiated city planning optimization suggestion, help city land resource efficient allocation and management, by fusing multi-source data, realize the multidimensional cognition of space, time, function and relationship, fundamentally improve the understanding depth of temporary land.