21 results about "Hypergraph" patented technology

The present disclosure provides a Boolean formula unsatisfiability core prediction method based on hypergraph modeling. A CNF formula generated by a software and hardware formal verification problem coding in the field of electronic design automation is received from the outside, input into a hypergraph representation learning model, and modeling for solving a SAT problem corresponding to the software and hardware formal verification problem is obtained. The model is constructed in the following manner: first, SAT problem representation based on hypergraph is performed. Specifically, the CNF formula is modeled as a clause-literal hypergraph by using a SAT problem modeling method based on hypergraph and a message passing method based on hypergraph, and a clause correlation graph is further constructed to obtain problem representation. Then, a polarity-aware variable decomposition method is applied to the clause correlation graph, and the problem representation is structured and modeled. The hypergraph representation learning model is trained by using a training method based on polarity-aware consistency constraints, and is integrated with a SAT problem solver, so that the model can be used to solve the problems that the existing graph neural network-based SAT learning method has in the aspects of polarity modeling and high-order structure expression.

Embodiments of the present application provide a point cloud matching method and device, and a storage medium. The method can include determining a consistency matrix between a first point cloud and a second point cloud, wherein an element in the consistency matrix represents a consistency size between a corresponding relationship of two points included in a first group of points and a corresponding relationship of two points included in a second group of points; constructing a hypergraph according to the consistency matrix between the first point cloud and the second point cloud; performing graph convolution processing on the hypergraph to obtain at least two groups of initial transformation matrices; and obtaining a target transformation matrix between the first point cloud and the second point cloud according to the at least two groups of initial transformation matrices. The high-order consistency represented by the hypergraph structure is more accurate, and is more robust to abnormal points and noise. The method can provide more accurate information for identifying and registering inlier points, and is helpful for effectively registering point clouds in a low overlap rate situation.

This invention discloses an action recognition method based on pyramid-style segmented dynamic graph construction and ellipsoidal geometric constraint metric learning, comprising the following steps: acquiring a human skeleton action sequence and constructing initial spatiotemporal features; constructing a feature space with ellipsoidal geometric constraints; employing a pyramid-style temporal segmentation strategy, dividing the temporal features into segments of different granularities at different depth levels of the network, constructing local dynamic graphs for each segment, and constructing a global dynamic graph for the entire sequence; adaptively weighting and fusing the generated dynamic hypergraph association matrix, the generated local dynamic graph, and the global dynamic graph with a preset static human topology prior graph to obtain the fused dynamic graph structure; sequentially performing spatial graph convolution and multi-scale temporal convolution on the initial spatiotemporal features to extract deeply fused spatiotemporal features; mapping to the target action category through a classifier to complete action recognition; this invention improves recognition accuracy and generalization ability.

This invention belongs to the field of information technology and relates to a method and system for temporal knowledge hypergraph reasoning based on a two-layer linear attention recurrent network. The method includes: obtaining node representations in subgraphs with different timestamps under the original time series based on a relation-aware graph convolutional network (RGCN), and integrating historical temporal information through a first-layer linear attention recurrent network (RWKV) to obtain a spatiotemporal fusion representation; constructing a hypergraph based on density peaks based on the spatiotemporal fusion representation and obtaining the non-qualitative association matrix of nodes relative to hyperedges; encoding the spatial features of the hypergraph based on a hyper-relation-aware graph neural network (HRGNN), and encoding temporal features through a second-layer linear attention recurrent network (RWKV) to achieve spatiotemporal information encoding; and using a decoder for connection prediction to achieve knowledge reasoning. This invention fully considers and designs various aspects of hypergraph construction, learning, and processing, enabling the extraction of temporal evolution patterns from historical information and achieving better reasoning about future events.

A computer-implemented method is provided. The computer-implemented method includes constructing a feature hypergraph, performing feature selection and combination, and performing predictive analysis. Performing feature selection and combination includes updating a representation of a hyperedge in the feature hypergraph by aggregating information from vertices connected to the hyperedge and selected by a gate function, and updating a representation of a vertex in the feature hypergraph by aggregating information from hyperedges to which the vertex belongs and selected by a gate function.

The application discloses a kind of structural information enhanced multimodal heterogeneous data fusion representation method, belong to data processing technical field.Method includes: by text, image, audio and video obtain multimodal heterogeneous data, after executing pre-processing operation to multimodal heterogeneous data, feature extraction and conversion operation are executed, obtain the multimodal feature matrix of uniform feature space and using graph structure enhancement technology constructs graph structure with stability and explainability;Structural entropy regular discriminant representation learning framework is constructed, structure information optimization framework and soft allocation mechanism are designed, hypergraph structure is constructed, and the structural entropy of hypergraph structure is calculated;Based on hypergraph structure entropy, guide multimodal unsupervised clustering.The present application is based on structural entropy, constructs hypergraph structure and introduces soft allocation mechanism, has the characteristic of automatically learning data structure information, improves the processing efficiency of unstructured data, effectively breaks through the traditional restriction condition, provides new solution.

This invention discloses a stock price trend prediction method based on multi-level dynamic graph fusion, relating to the fields of financial technology and artificial intelligence data analysis. The method mainly comprises three parts: multi-channel temporal feature extraction, collaborative modeling of macro and micro spatial structures, and multi-level feature fusion prediction. The steps include: First, constructing a multi-channel technical indicator sequence based on historical stock trading data, and using a multi-channel attention pooling GRU network to extract differentiated temporal features in parallel; Second, constructing a stock association hypergraph based on industry and concept labels, introducing graph information loss (GIL) as a feedback signal to drive an adaptive hyperedge reconstruction mechanism, dynamically adjusting hyperedge weights to filter noise and capture macro market hotspots; Simultaneously, constructing a simple graph within the hyperedge and combining contrastive learning constraints to semantically align stocks with similar technical patterns to enhance the discriminative power of micro local features; Finally, integrating macro hypergraph features and micro simple graph features through a cross-graph fusion module, concatenating them with temporal features, and inputting them into the prediction layer to complete the prediction of the stock price trend at the next moment. This method overcomes the limitations of existing static graph models in dynamically capturing high-order stock correlations and fine-grained technical pattern resonances, providing a new approach for financial spatiotemporal data mining.

The application discloses a miRNA, lncRNA and disease multi-element association efficient prediction method using a contrast hypergraph generation technology, a heterogeneous network construction module is used for integrating miRNA-lncRNA interaction, disease association network and gene expression multi-source data, and a three-layer heterogeneous graph containing molecular nodes and disease nodes is constructed, a contrast hypergraph generation module is used for enhancing representation ability through hypergraph structure construction and interaction and contrast learning, a multi-element association prediction module is used for uniformly processing multi-source data, dynamically adjusting loss weight, an explainability module is used for clustering analysis on attention weight of molecular nodes and disease nodes in the hypergraph, and a data enhancement module is used for generating synthetic negative samples from topological characteristics of high-risk nodes to adjust training set distribution; through integration of heterogeneous network construction, contrast hypergraph generation and multi-task learning framework, and through the three-layer heterogeneous graph structure and the contrast hypergraph generation module, the application breaks through the limitation that a traditional graph model can only model a binary relationship.

The application discloses an intelligent intervention strategy generation method for whole-cycle health management, and belongs to the technical field of health management, and specifically comprises the following steps: acquiring heterogeneous health original data, inputting the heterogeneous health original data into a modal alignment network, extracting context correlation information through a cross attention mechanism, and generating a fixed-dimension cross-modal feature representation; inputting the cross-modal feature representation into a hypergraph topology construction module according to a time sequence, taking time sequence mapping hyperedges and feature nodes as vertices, and constructing a time sequence health hypergraph structure containing a time dimension; outputting a health time evolution feature matrix expressing the evolution relationship of time nodes; inputting the health time evolution feature matrix into a hidden Markov model to generate a dynamic digital health portrait matching the above-mentioned hidden state; introducing the dynamic digital health portrait into a reinforcement learning decision framework to output an action sequence as a candidate intervention path; and decoding and mapping the action sequence in combination with the constraint rules of the dynamic digital health portrait to generate an intelligent intervention strategy.

This invention provides a mobile communication domain RAG question answering method driven by an ontology-constrained hypergraph in the fields of artificial intelligence and natural language processing. The method includes: Step S1, extracting domain knowledge to construct a domain ontology; Step S2, mapping knowledge fragments to vertices of a hypergraph based on the domain ontology, constructing ontology-constrained hyperedges to form an ontology-constrained hypergraph; Step S3, identifying the query semantics of the user query, obtaining an initial set of candidate knowledge fragments from the ontology-constrained hypergraph based on vector similarity; performing iterative semantic diffusion by activating and filtering associated hyperedges; selecting retrieval results from the retrieved knowledge fragments; Step S4, inputting the retrieval results into a large language model, verifying and correcting the initial answer, and outputting the final answer. The advantages of this invention are: improved semantic representation and associative reasoning capabilities for complex high-order business events, enhanced domain knowledge orientation and accuracy in the retrieval process, and ensured business compliance and logical integrity of the generated answer.

The application discloses a hierarchical function alignment hypergraph learning method and system for auxiliary diagnosis of brain diseases. In view of the influence of high-order interaction level heterogeneity of changes caused by brain diseases and presented among different anatomical levels, a hierarchical high-order knowledge capturing mechanism is adopted to jointly construct pair graphs and hypergraph relationships on multiple anatomical levels, so that the model can cooperatively learn and effectively capture high-order interaction knowledge of a specific level, and the inappropriate modeling caused by multi-level rough learning is avoided. In view of the influence of functional role level heterogeneity of the same brain element presented under different observation levels, the functional role of a specific level is evaluated through hierarchical functional role alignment combined with signal variability and structural characteristics, and the functional role of the brain element is aligned among multiple anatomical levels, so as to solve the knowledge conflict problem caused by inconsistent functional roles, thereby improving the generalization ability of the graph neural network across levels and realizing more accurate diagnosis.

This application relates to the field of robot human-computer interaction and discloses a multi-user communication system based on an AI robot knowledge graph, including a multimodal perception component, a dynamic hypergraph construction module, a semantic potential field arbitration module, an isomorphic pruning verification module, and a semantic motion mapping control module. The method constructs a dynamic hypergraph containing hyperedges by parsing multi-source data, temporarily stores and dynamically completes low-confidence semantics using a time-decaying virtual node mechanism, calculates semantic resistance based on user negative emotions and environmental constraints, and searches for the optimal candidate path using the minimum potential energy principle. It blocks dangerous interactions by extracting local subgraphs and performing isomorphic matching with a taboo graph library. Finally, it establishes a mapping from semantic potential energy to physical space and adjusts joint stiffness and damping parameters in real time based on the semantic potential energy value. This invention effectively solves the problems of intent conflict and context discontinuity in multi-user scenarios, achieving a deep integration of cognitive decision-making and physical compliance control.

The application discloses an attack detection and tracing method and system based on a hypergraph, and the method comprises the following steps: collecting a kernel log, and performing compression processing to obtain a first origin graph; performing path matching according to an ATT&CK attack and defense matrix, matching a hyperedge in the first origin graph, and constructing a hypergraph based on the hyperedge; matching each hyperedge in the hypergraph based on attack behaviors obtained based on expert experience, marking the hyperedge that is successfully matched as malicious behavior, and performing a tracing operation according to the hyperedge. The application converts the origin graph by using the structure of the hypergraph, matches the attack behaviors through the hyperedge, improves the detection efficiency, and can trace out the initial attack entry node.

The application discloses a supergraph neural network-based SDN saturation attack detection method and relates to the technical field of network communication. In order to solve the technical problems that the existing SDN security protection work has the problems of insufficient description ability of complex network structure, limited dynamic adaptation ability and weak accurate detection ability, the technical scheme provided by the application is as follows: a CHS graph is established according to state information of a target network; in the preset model learning of a throttling point of the CSH graph, an attention mechanism is used to capture the influence of a node on a hyperedge to which the node belongs and the influence of the hyperedge on the nodes in the hyperedge; the features of the hyperedge are relearned; the preset model is trained according to a preset loss function; the throttling points in the CHS graph of the target network are classified through the trained preset model, and the saturation attack types are distinguished according to the classification results, including the saturation attack on the SDN switch and the saturation attack on the SDN controller. The application is suitable for application in the work of SDN saturation attack detection.

The application provides a recommendation method, comprising: in response to a recommendation request, obtaining historical interaction data of a request user corresponding to the recommendation request, and determining a to-be-recommended object according to the historical interaction data; for each to-be-recommended object, constructing an interaction pair; for each interaction pair, taking a feature corresponding to the interaction pair as a node, and constructing a default hypergraph corresponding to the interaction pair; inputting the default hypergraph corresponding to the interaction pair into a hypergraph learning model, to predict a hyperedge set of the default hypergraph by a hyperedge generation module, and to model the default hypergraph based on the hyperedge set by a hypergraph learning module, to obtain a complete hypergraph corresponding to the interaction pair, and to predict a probability that a recommended object corresponding to the interaction pair is liked by the request user based on the complete hypergraph corresponding to the interaction pair by a prediction module; and selecting candidate recommended objects from the recommended objects corresponding to each interaction pair according to the probability that the recommended objects corresponding to each interaction pair are liked by the request user, to perform recommendation.

The application relates to the technical field of production line modeling, and proposes a production line parameter change propagation evolution modeling method based on a hypergraph, which comprises the following steps: constructing production line parameters and a correlation relationship into an initial hypergraph model containing nodes, hyperedges, membership relationships, attributes and constraint functions. Then, the system receives a change operation instruction for the nodes or the hyperedges, and executes corresponding atomic or composite operations according to the instruction type, and dynamically updates the topological structure and the membership relationship of the hypergraph. After the change is executed, the system automatically locates the affected area, triggers an attribute recalculation mechanism driven by the constraint function, realizes the automatic propagation and consistency maintenance of the parameter change in the hypergraph model, and finally outputs the production line parameter model in the current state. The application solves the technical problems of high parameter coupling degree and complex change propagation link in a complex production line, and significantly improves the modeling accuracy and evolution efficiency in the parameter change process.

This invention discloses a hypergraph visualization method for knowledge graphs. It constructs knowledge graph triples and a knowledge graph hypergraph model, using targets in network attacks as entities and subordinate information as relations. A general knowledge graph based on triple data is constructed, mapping all nodes of the knowledge graph triple relations to a set of entity points in the hypergraph. All relations of the knowledge graph triple relations are mapped to hyperedges of the hypergraph according to their categories. Simultaneously, entities containing a relation are placed within the hyperedge, with the relation as the core. A two-dimensional planar bounded region entity point uniform distribution model is constructed, converging entity points with similar orders of relation at one point, resulting in a knowledge hypergraph model with circular hyperedges for visualization. This invention solves the problem of large-scale point and line aggregation in the visualization model of ordinary graphs.

This invention discloses a method for recommending points of interest (POIs) in different locations based on user preferences and personalized preference transfer, belonging to the field of terminal location-based recommendation. The method includes: constructing a heterogeneous hypergraph for five different types of nodes, and obtaining user preference representations through training the hypergraph; constructing a POI-category graph, and learning POI representations through a continuous skipping word model; constructing an attention network with POI representations as input to obtain user-transferable features; constructing a parameter learning network using a multilayer perceptron and user-transferable features as input, and constructing a transfer network with user preference representations as input and the output of the parameter learning network as parameters to achieve personalized user preference transfer; constructing a geographic map between POIs based on latitude and longitude, and learning the embedding representations of different POIs through a convolutional network; calculating the score for each POI by combining the user's transferred preferences with the embedding representations of different POIs, thus completing the final recommendation.

The application discloses a traffic accident risk prediction method and related devices. The method obtains and pre-processes multi-source heterogeneous traffic data; adopts a dual heterogeneous graph neural network to perform double feature coding on heterogeneous entity nodes and associated hyper-edges, and obtains heterogeneous entity node features and heterogeneous hyper-edge associated features; constructs a time sequence knowledge hypergraph based on the heterogeneous entity node features and the associated hyper-edge features, optimizes node embedding in the time sequence knowledge hypergraph based on dynamic hypergraph convolution, and obtains hypergraph heterogeneous features; takes the hypergraph heterogeneous features optimized by the dynamic hypergraph convolution as input, adjusts feature weights from a time sequence dimension and a heterogeneous entity dimension through a space-time-dual relationship cross attention mechanism, and obtains space-time cross fusion features; cross-fuses the hypergraph heterogeneous features and the space-time cross fusion features, and obtains comprehensive features; and predicts traffic accident risks based on the comprehensive features. The application can realize accurate traffic risk analysis and interpretable decision-making.

This invention provides a graph representation learning method based on a variational hypergraph mask autoencoder. The method includes: first, adaptively masking the input graph data according to node degree to generate a masked node feature matrix; then, obtaining initial node representations using an initial graph encoder incorporating an edge feature attention mechanism; subsequently, constructing a hypergraph structure based on the initial representations, and performing bidirectional information aggregation between nodes and hyperedges through hypergraph convolutional layers to output a hypergraph encoded representation; next, inputting the hypergraph encoding into a variational encoder to generate a probabilistic latent representation; finally, reconstructing node features using a decoder, calculating the total loss including reconstruction loss, KL divergence loss, and structure preservation loss, and optimizing the model using a progressive pre-training strategy. This invention effectively captures high-order association information of graph data, solves the problems of coarse masking and loss of structural information in existing methods, and improves the robustness and accuracy of graph representation learning.