54 results about "Graph sampling" patented technology

The embodiment of the invention provides a power distribution network dynamic reconstruction method based on knowledge-guided large model enhanced fine tuning and related equipment, and belongs to the technical field of smart power grids and artificial intelligence. The method comprises the following steps: constructing a power distribution network dynamic knowledge graph, and providing structured knowledge guidance for model training; subgraph sampling is carried out based on the timestamp and converted into a fine tuning sample, and a training data set is generated; utilizing the data set to supervise, finely adjust and preheat the large language model; designing a multi-dimensional reward function of fusion format accuracy, economy and security based on mechanism knowledge in the knowledge graph and expert experience; a group relative strategy optimization mechanism is adopted to carry out reinforced fine tuning on the large language model, and the large language model is guided to output a safe, reliable and economical dynamic reconstruction strategy in interaction with the environment. According to the method, the problems of lack of training data, lack of physical knowledge guidance and insufficient decision reliability of a large language model in the power grid field are solved, and the intelligent level and decision quality of dynamic reconstruction of the power distribution network are remarkably improved.

The invention relates to the technical field of government affair entrepreneurship guidance service intelligent dialogue systems, in particular to a dialogue intention recognition and correction method based on a knowledge graph, and the method comprises the steps: obtaining a voice signal and text information inputted by a user, and carrying out the cross-modal fusion processing to generate a semantic vector; carrying out evolution modeling on the dialogue state by adopting a graph neural network, and outputting a dynamic dialogue state graph containing a time sequence relationship between a user entity and a business object; taking the semantic vector and the focus information as query conditions, executing dynamic subgraph sampling from the government affair knowledge graph, and generating an intention correction strategy vector; and finally, utilizing reinforcement learning to generate personalized guide statements and outputting the guide statements to the user. According to the method, through deep coupling of semantic understanding, state tracking, knowledge reasoning and decision optimization, real-time perception and active completion of dynamic updating of hidden nodes and policies of the government affair process are achieved, and the intention recognition accuracy and systematic adaptive capacity are effectively improved.

The invention provides a workpiece surface target merging and partitioning method, system and device, and relates to the technical field of image processing, and the workpiece surface target merging and partitioning method comprises the steps: obtaining a surface image of a target region of a target maintenance workpiece; performing graph sampling and visual processing on the surface image to generate a binary image of the target area, and analyzing the binary image to obtain a plurality of connected domains of the target area; performing convex hull generation according to the connected domains to obtain a convex hull corresponding to each connected domain; generating a rotation enclosing rectangle set according to the leading edge vector of each convex hull; and through a variant ant colony optimization algorithm, carrying out global optimization on the rotating enclosing rectangle set, and generating a combined partition of the target area. According to the invention, through accurate capturing of the original image and subsequent layer-by-layer optimization, the redundancy of the laser maintenance path is reduced, and the operation cost is reduced; through construction and global optimization of a high-adaptation form carrier, precise coverage of a complex form target is realized, and the maintenance precision is improved.

The invention particularly relates to a CAD part feature recognition method based on a graph neural network, and the method comprises the steps: extracting geometric entity information and topological relation information based on C # and NX Open API, and generating standardized AAGJSON format data; converting the geometric data into a graph structure which comprises a topological adjacent matrix and a multi-dimensional node feature vector comprising geometric, topological and shape features; a graph neural network model comprising a graph convolution layer, a graph attention layer and a graph sampling aggregation layer is adopted for training, and feature categories and confidence degrees of holes, grooves, bosses and the like are output; and the prediction result is corrected based on geometric consistency, topological connectivity and context rationality constraint. The method has the characteristics of high recognition precision, strong adaptability and good robustness of complex parts, can be integrated in UG NX to realize real-time recognition and visualization, provides important technical support for design automation and manufacturing industry digitization, and solves the problems of low recognition precision, poor adaptability and insufficient topological information utilization in the existing CAD part feature recognition method.

The application relates to a kind of smart grid dynamic graph data generation methods based on deep adversarial training, comprising the following steps: obtaining smart grid time series graph data, inputting the smart grid dynamic graph data generation model based on deep adversarial training, and generating smart grid dynamic data graph;The smart grid dynamic graph data generation model based on deep adversarial training includes sampling module, generative adversarial network and reconstruction module, wherein the sampling module is used to generate the center graph using the center graph sampling method;The generative adversarial network includes generator and discriminator, the generator includes decoder and encoder, the encoder is used to encode the center graph through the time series graph self-attention network constructed based on the multi-head self-attention mechanism, and obtain the hidden variable of the center graph;The reconstruction module is used to generate the classification distribution of each time series edge according to the center graph score matrix, and then generate the smart grid dynamic data graph.Compared with the prior art, the application can efficiently and reliably generate smart grid dynamic data graph.

The invention provides a large-scale supply chain solving method and system based on graph sampling and graph coarsening, and the method comprises the steps: building a target supply chain model with the maximization of total welfare as a target according to the basic element parameters of an electric truck in a supply chain network; constructing a first supply chain model based on an active transportation path in the target supply chain model and a graph sampling technology, and solving the first supply chain model to obtain an optimal lower bound and an optimal feasible candidate solution; constructing a second supply chain model based on active supply chain nodes in the target supply chain model and a graph coarsening technology, and solving the second supply chain model to obtain an optimal upper bound; and according to the optimal lower bound and the optimal upper bound, solving to obtain an optimal gap, and based on the optimal gap and the optimal feasible candidate solution, determining a target large-scale supply chain solving result. According to the invention, the solving efficiency and accuracy are improved.

A heterogeneous optimization method and device based on subgraph link prediction, the method comprising the steps of: based on the behavior or relationship between users, commodities and shop entities, abstractly constructing a no-weight and no-direction graph required by a model with node labels; implementing closed subgraph sampling extraction of target node pairs on the no-weight and no-direction graph; simply controlling the upper limit of the size of the subgraph and reasonably filtering the samples; calculating the node heterogeneity index of each closed subgraph; generating a new heterogeneous feature vector according to the calculated index data; replacing the obtained heterogeneous feature vector with the original node feature to become a new node feature; extracting subgraph features and performing model training, and finally realizing the prediction of whether an edge exists between the target node pairs; obtaining a recommended candidate list of each user according to the obtained prediction result of the existing edge; and according to the prediction score, completing the sorting of the recommended candidate list obtained by each user to obtain the final recommended result displayed to the user.

The invention discloses a power system load flow analysis training data enhancement method and related equipment, and the method comprises the following steps: taking all buses in a power grid as nodes in a topological graph, taking all connection equipment as middle edges of the topological graph, and obtaining a structure graph of the power grid; the sample enhancement method is recorded as a sample enhancement method based on subgraph sampling and / or a sample enhancement method based on subgraph grafting based on the structure graph, and samples are generated to enhance the training data. According to the power system load flow analysis training data enhancement method based on subgraph sampling and subgraph grafting, node scale change is carried out on the actual operation scene of a power grid, the number of training samples is increased, the number of nodes of the samples is increased, and the richness of a topological wiring mode is improved. The invention provides a sample expansion method based on sub-graph sampling and sub-graph grafting, and the adaptability of the model to a node number reduction scene caused by starting and stopping of a generator and a node number increase scene caused by power grid expansion can be improved respectively.

The invention discloses a paper classification method based on graph matching and self-supervised graph learning, and relates to the technical field of document classification based on deep learning. According to the method, literature data is represented by adopting a literature relation graph, a graph learning model ConGM based on the literature relation graph is constructed, and reference and theme association between literatures are mined through sub-graph sampling and data enhancement, linear node matching, secondary edge alignment and double-layer negative sample selection, so that precise classification of fields to which papers belong is realized.

The invention provides a method and a system for constructing a knowledge representation learning model based on a periodic perception contrast graph attention network. The method comprises the steps of constructing a sub-graph sampler, constructing a graph attention network and performing time sequence perception contrast learning. According to the method, firstly, adjacent neighbor entities are screened and queried through a period time sensing sub-graph sampler in a period dynamic weighting mode, consumption of computing resources is reduced, and key time sequence context and period information are reserved; then, the graph attention network with the time perception function generates dynamic representation of time perception by using graph attention in the constructed query related sub-graph, and entity embedding representation fused with neighbor information is obtained; then, time sequence perception contrast learning improves robustness in time knowledge graph reasoning through dynamic and static representation contrast learning of an entity, and entity prediction is carried out through embedded representation obtained through learning. Finally, the entity representation is decoded using a decoder.

The invention relates to a random subgraph sampling method and system for machine or deep learning training and reasoning. The method comprises the following steps: step 1, estimating a budget; 2, scale solving is carried out; 3, at least one sampling rule is selected from a preset sampling rule base to generate sub-image batches meeting constraints; step 4, obtaining sub-graph labels; 5, implementing a boundary retention mechanism and / or virtual neighbor convergence on the cut edges of the cross-sub-graph to compensate cross-sub-graph information; 6, in the training stage, importance weighting and / or reweighting loss are / is carried out on estimation deviation introduced by non-uniform sampling; step 7, training and / or reasoning the machine / deep learning model based on the subgraph batch, and performing budget adaptive adjustment on the subgraph scale and the sampling rule according to actual occupation and time delay during operation; and executing a corresponding task through the machine / deep learning model. According to the method, data which cannot be trained / inferred originally runs on limited equipment in a sub-graph batch form, and the problem of memory overflow is avoided.

The invention belongs to the technical field of artificial intelligence and high-performance computing, and discloses a dynamic sampling and parallelization acceleration method for pulse graph neural network computing. An innovative dynamic graph sampling mechanism is provided, the instantaneous importance of nodes is evaluated according to historical pulse activities of the nodes, limited computing resources are dynamically allocated to the nodes and areas with the most abundant information, and a sampling strategy is adaptively adjusted. Secondly, the invention further provides a state decoupling parallelization method, pre-calculation and caching are carried out on a stateless layer of the SNN, part of serial dependence between time steps is broken, and remarkable acceleration of the calculation process is achieved. According to the method, a better effect is achieved under different data sets, and compared with other methods, the method has competitiveness in performance. On a MINIST data set and a DvsGesture data set, on the premise that the accuracy rate is basically kept unchanged, a remarkable acceleration effect is achieved, and meanwhile memory consumption of the method is obviously reduced.

The application discloses a digital quantum computing-based computing power network DDoS detection method and system, and relates to the technical field of network space safety management and control.The application comprises the following steps: collecting real-time network flow data in a jurisdictional area through a dispersed detection point, constructing a graph neural network DDoS detection algorithm based on quantum random walk sampling, and performing quantum random walk to obtain quantum probability distribution of each node; performing quantum probability weighted non-replacement sampling on neighbor nodes of each target node to obtain a sampled neighbor set; performing message passing on the sampled neighbor set through a graph sampling aggregation network; uploading local model parameters to a safety management center through a federated learning architecture, and outputting a DDoS attack detection result.The application realizes DDoS flow feature extraction based on a graph structure, and utilizes a neighbor sampling mechanism guided by quantum random walk to complete real-time identification and detection of DDoS attacks.

The invention provides a data analysis method and device, electronic equipment, a storage medium and a program product. The method comprises the steps of obtaining input data; performing data matching in a knowledge graph by utilizing a graph sampling and aggregation model according to the input data, performing vector representation conversion and text representation conversion according to a matching result, and splicing the vector representation and the text representation to obtain intermediate data; determining whether the intermediate data accords with a set rule or not; and in response to the intermediate data conforming to a set rule, determining a calculation rule of an analysis result in the knowledge graph according to the intermediate data, determining the analysis result according to the calculation rule and the intermediate data, and outputting the analysis result.

The application relates to a communication-efficient decentralized federated learning method based on subgraph division, comprising the following steps: constructing a physical network topology graph of a decentralized federated learning environment; dividing the physical network topology graph into non-conflict subgraphs, and setting a mixing matrix for each non-conflict subgraph; constructing a problem constraint model with the minimum total iteration time as the target; optimizing a bandwidth allocation matrix with the delay value of the link with the maximum communication delay in the current non-conflict subgraph as the optimization target; optimizing the mixing matrix with the minimum expected communication delay as the optimization target; optimizing the non-conflict subgraph sampling probability based on the optimized mixing matrix with the minimum expected communication delay as the optimization target; and performing global updating of the decentralized federated learning model based on the optimized non-conflict subgraph sampling probability. The application can significantly reduce the global model convergence delay.

The embodiment of the invention provides a graph data processing method and system and a model training method and system. Wherein the graph data processing system obtains a sampling sub-graph of a target node from original graph data, and the sampling sub-graph comprises a plurality of target neighbor nodes corresponding to the target node; a node text corresponding to each sub-graph node in the sampling sub-graph is obtained, the node text is input into the large language model to obtain a discriminative text corresponding to each sub-graph node, and the discriminative text comprises discriminative information used for classifying the sub-graph nodes; and inputting the sampling sub-graph and the discriminative text corresponding to each sub-graph node into a graph neural network to obtain a category discrimination result corresponding to the target node.

The application discloses a zero sample graph learning method based on a large language model, comprising: neighborhood subgraph sampling on to-be-processed graph structure data to generate graph query text; based on the graph query text, a structured prompt template containing multi-stage reasoning instructions is constructed, and the prompt template instructs the large language model to sequentially perform topological structure analysis, semantic attribute interpretation, multi-candidate answer enumeration and deep reevaluation before generating a final answer; a graph reasoning training data set is constructed; the basic large language model is fine-tuned by full parameters by using the graph reasoning training data set; the large language model is optimized by using a reinforcement learning algorithm based on group relative advantage; for a new graph task to be reasoned, the new graph task is input into the optimized large language model, and a final prediction answer containing a reasoning process is directly output. The application discards the dependence of a graph neural network, and only uses the text reasoning capability of the large language model to complete a graph task, so that the deployment complexity and the calculation cost are greatly reduced.

The invention relates to the technical field of big data analysis, and discloses a spatio-temporal big data intelligent tracing system and method for new infectious diseases, and the method comprises the steps: converting original multi-source data into a standard spatio-temporal format, and carrying out the entity recognition matching and quality evaluation; constructing a multi-layer dynamic graph structure comprising an individual layer, a region layer and a global layer, and extracting spatio-temporal features to obtain node embedding vectors; calculating the propagation probability between nodes through a graph sampling mechanism and a time sensing mechanism; establishing a multi-dimensional risk system, and designing an adaptive risk scoring function; carrying out cross-mechanism analysis by adopting federated learning and differential privacy protection; constructing a Markov decision model, and generating an intelligent decision scheme through deep reinforcement learning; according to the method, efficient fusion of multi-source heterogeneous data is realized by establishing a unified space-time reference coordinate system and a self-adaptive fusion mechanism.

This invention relates to the field of lithium-ion battery state monitoring and health management technology, and proposes a battery health state estimation method based on an improved graph neural network and a hybrid PINN. The method includes constructing a hybrid graph structure containing time edges and similarity edges based on weighted multidimensional health features and time series information, and inputting it into an improved graph sampling aggregation network. High-dimensional spatiotemporal features of each battery charge-discharge cycle node are extracted and embedded through multi-layer graph convolution operations. A hybrid physical prior model is constructed and combined with a residual correction module to form a physical constraint dynamic equation. The high-dimensional spatiotemporal feature embeddings are input into a prediction network to obtain a preliminary prediction of the battery health state. A loss weight annealing strategy is used to dynamically weight the data-driven loss and the physical constraint loss, and the improved graph sampling aggregation network, prediction network, hybrid physical prior model, and residual correction module are collaboratively trained to finally output the estimated battery health state.

The application discloses a kind of based on low rank contrast learning and reconstruction graph anomaly detection method, comprising: 1) obtain graph data, and generate low rank graph data by SVD singular value decomposition dimension reduction, using restart random walk algorithm to carry out subgraph sampling, obtain original view and low rank view;2) on original view and low rank view, construct contrast pair and carry out contrast learning, obtain contrast learning loss;3) low rank attribute reconstruction is carried out on original view and low rank view, and the final reconstruction loss of original view and low rank view is obtained;4) the graph neural network model is trained in combination with contrast learning loss and reconstruction loss;5) according to the trained graph neural network model, the abnormality of node in the graph to be measured is judged, and the potential abnormal node is determined.The application generates low rank view by low rank approximation to node attribute and topological structure, effectively filters the interference of abnormal node and noise on the basis of retaining the original structure of graph.

The invention discloses a city parking lot prediction method driven by an artificial intelligence algorithm, and belongs to the technical field of intelligent traffic. In order to solve the problem that sensor-free parking lot vacancies are difficult to obtain in real time, the method comprises the following steps of: a) calculating the distance between parking lots by using a threshold-free Gaussian kernel function to obtain a forward / reverse normalized adjacency matrix; b) sampling by adopting an inductive sub-graph, randomly extracting nodes, generating a mask matrix, and forming a sub-graph sample; c) sending the sample into a three-layer diffusion diagram convolutional network (second-layer residual connection), extracting spatial features by using a bidirectional adjacent matrix layer by layer, and outputting vacancy prediction; d) updating the network by using observable and unobservable nodes by taking a mean square error as loss; and e) the newly added parking lot can be predicted by directly updating the adjacent information without retraining.

The invention provides an APT attack traceability method and system based on time sequence characteristics and traceability graph sampling, and solves the technical problems of low detection precision and poor real-time performance of an existing APT attack traceability method. The method comprises the steps of constructing an initial traceability graph; obtaining threat knowledge, analyzing conditional probabilities of different node types in the benign behavior, and supplementing the initial traceability graph according to the node types and the conditional probabilities to obtain an enhanced traceability graph; using an encoder-decoder structure to learn structural features and time sequence features in the enhanced traceability graph, predicting an edge type, comparing the edge type with an actual edge type, and minimizing a prediction error to obtain an attack traceability model; and constructing a K-order sub-graph by taking the edge to be detected as a center, and judging whether the edge to be detected is abnormal or not by the attack traceability model according to the interactive behavior type in the range of the K-order sub-graph. The method can be widely applied to the technical field of network security.

The invention provides an industrial process key index prediction method and system, and a computer readable storage medium. The industrial process key index prediction method comprises the steps of obtaining input sequence data, and extracting an embedded feature of each node in the input sequence data through convolution operation; the embedded feature of each node in the input sequence data is input to a GRKAT-GSL model, a predicted value of the industrial process key index is determined, the GRKAT-GSL model comprises a graph structure learning module and a knowledge guidance graph attention neural network module, in the graph structure learning module, a prior graph is constructed based on the embedded feature of each node, and in the knowledge guidance graph attention neural network module, the prior graph is constructed based on the embedded feature of each node; carrying out the sampling of the priori graph through a Gumbel-Softmax sampling method, so as to determine a learning graph; and the knowledge guidance map attention neural network module includes a gating loop unit that fuses aggregation results of an attention score matrix, the attention score matrix being determined based on the learning map.

The application discloses a sensor arrangement method for a water supply network based on a graph sampling theory, and comprises the following steps: establishing a user node undirected graph, determining a graph Fourier operator of the user node undirected graph; obtaining user node pressure of the water supply pipe network under different leakage conditions, obtaining a pressure sensitivity matrix of the water supply pipe network, and determining useful information in the pressure sensitivity matrix; performing Fourier transform on the useful information by using the graph Fourier operator to obtain a graph Fourier spectrum of each user node under different leakage conditions; screening out a graph Fourier spectrum group which is greater than or equal to a spectrum threshold to form a new spectrum matrix, and screening out a frequency component corresponding to the new spectrum matrix to form a new frequency component matrix; taking the number of spectrums in the new spectrum matrix as the number of sensor nodes to obtain an update signal; constructing a target function according to the update signal; and screening out user nodes which satisfy the target function as sensor nodes one by one until the number of sensor nodes is reached.

The invention discloses a graph anomaly detection-based encrypted currency transaction anomaly recognition method and device, which are implemented by adopting a community detection-based graph anomaly recognition algorithm (CDAD) and a time sequence analysis-based pattern recognition algorithm (TAPR) based on a double-layer adaptive anomaly detection algorithm framework. The core of the algorithm is that a transaction sub-graph sampling method generated by random walk is used for estimating an abnormal score and a money laundering mode probability, and good detection accuracy guarantee is achieved. The problem of how to detect and identify the money laundering behavior mode in real time in a large-scale encrypted currency transaction network is solved, and the defects that a dynamic transaction graph structure cannot be effectively processed and a time sequence correlation analysis capability is lacked in a traditional method are overcome; and various money laundering modes can be effectively identified under an independent chain model and a time decay propagation model.

The present application relates to the technical field of network security analysis, in particular to a network node full-granularity anomaly detection method and system based on attribute-enhanced sampling, which obtains an attribute-enhanced network of an original attribute network based on an attribute-enhanced manner, and generates positive and negative sample pairs of both the original attribute network and the attribute-enhanced network through subgraph sampling of interval random walk; a full-granularity contrast learning network containing nodes and subgraphs, nodes and nodes, subgraphs and subgraphs, and nodes and the whole is constructed by using the positive and negative sample pairs, so as to capture abnormal information of nodes at the subgraph level, the node level and the global level by using the full-granularity contrast learning network; the abnormal value score of each node is calculated based on the full-granularity contrast learning network, and the abnormal nodes in the attribute network are determined according to the abnormal value score, wherein the abnormal value score includes node and subgraph abnormal score, node and node abnormal score, and node and global abnormal score. The present application can improve the accuracy of attribute network anomaly node detection and facilitate deployment and application in actual scenarios.

This invention provides a short text clustering and fuzzy recognition algorithm for large-scale online subgraph sampling, comprising the following steps: Step S1, extraction and preprocessing of training samples; Step S2, construction of the neural network; Step S3, overall clustering prediction; Step S4, fuzzy sample recognition; Step S5, retraining of the neural network. This invention combines short text clustering with a large language model, which not only improves clustering accuracy but also enables the handling of clustering tasks with different themes and classification requirements, significantly reducing the manual cost of data annotation. Furthermore, this invention can annotate fuzzy samples for classification, using K-nearest neighbors combined with minimum spanning trees to assist subgraph sampling in the selection scheme. This utilizes sparse structure to reduce computational costs and exposes the fluctuations of boundary samples through spectral clustering, providing a more comprehensive perspective for fuzzy sample selection and improving the accuracy and interpretability of the clustering results.

The invention discloses a parallel sub-graph matching result cardinal number estimation method and device and a medium. The method comprises the steps that a data graph is preprocessed; a triangular index is constructed in a serial mode, and a parallel four-ring index with nested granularity division is constructed; constructing an initial candidate space of the query graph; refining the candidate space in parallel; sampling the spanning tree of the query graph, and evaluating the precision requirement of a sampling estimation value; when the estimation precision meets a preset requirement, returning an unbiased estimation value based on a sampling result, and directly outputting a cardinality estimation value; otherwise, carrying out parallel graph sampling. Through a parallelization method, the problem of long time consumption of index construction and result cardinal number estimation when an existing serial sub-graph matching cardinal number estimation algorithm processes a large-scale graph is solved; on the premise of keeping the estimation precision unchanged, compared with an existing serial method, the speed-up ratio of 1.53 times to 6.93 times can be obtained in the index construction stage, and the speed-up ratio of 1.13 times to 4.61 times can be obtained in the query processing stage.

The application provides a subgraph-based information cascade prediction method and system, belonging to the technical field of social network analysis and neural network, comprising the following steps: S1, constructing a deep learning information cascade prediction model CasSubTS, inputting collected user published information through an input layer, and constructing an information cascade graph G; S2, inputting G into a subgraph sampling layer, dividing G into a plurality of information cascade subgraphs according to different time steps, and converting the information cascade subgraphs into adjacency matrices; aggregating node features of the adjacency matrices to obtain a feature representation matrix B; S3, inputting B into a feature learning layer to obtain a characteristic vector with structural features and time sequence features; S4, inputting an input feature weighting layer, and utilizing a channel attention mechanism to perform weighted fusion to obtain a weighted characteristic vector; and S5, inputting an input prediction layer to predict a final macro cascade increment. The method can effectively predict information cascade in a social network.