123 results about "Graph classification" patented technology

The invention provides a knowledge data classification multi-modal reasoning evolution system based on a deep model, and the system comprises a dynamic semantic memory network module which is used for storing multi-source heterogeneous data through a memory network, and according to the relevance between the newly input multi-source heterogeneous data and the existing knowledge content in the memory network, obtaining the knowledge content of the multi-source heterogeneous data; dynamically updating the knowledge content in the memory network; the cross-modal collaborative classification enhancement module is used for extracting feature vectors of different modals from the multi-source heterogeneous data, performing feature alignment, and performing feature fusion on the feature vectors after feature alignment to generate comprehensive feature vectors fused with multi-modal information; and the multi-level layered attention classification module is used for processing the comprehensive feature vector by adopting a multi-level attention mechanism to understand the overall content and context information of the multi-source heterogeneous data so as to make a classification decision. According to the method, intelligent processing and classification decision making of complex information are realized, and a solid technical support is provided for knowledge evolution of a Chinese graph classification system.

The invention belongs to the technical field of graph classification, discloses a graph classification method based on subgraph integration and position sensing, and relates to a graph classification method SIPA combining substructure embedding and node position sensing. The SIPA firstly extracts sub-graphs through two different strategies so as to capture multi-sample sub-structures in the graphs, the structural features of the sub-structures are coded by adopting a graph convolutional network, and information of different sub-graphs is effectively fused through an attention mechanism. Then, anchor nodes are introduced to calculate relative position information of nodes, and the information is embedded into node representation to better capture global position features. And finally, a graph information bottleneck mechanism is used for optimizing node representation and removing redundant information irrelevant to a classification task. The method not only effectively learns the local structure information, but also enhances the perception capability of the relative position information of the nodes. Experimental results show that SIPA is superior to an existing baseline model in five data sets, and the superiority of SIPA in a graph classification task is verified.

Embodiments of the present systems and methods may provide techniques to predict the success or failure of a drug used for disease treatment. For example, a method of determining drug efficacy may include, for a plurality of patients, generating a directed acyclic graph from health related information of each patient comprising nodes representing a medical event of the patient, at least one first edge connecting the first node to an additional node, each additional edge connecting nodes representing two consecutive medical events, the edge having a weight based on a time difference between the two consecutive medical events, capturing a plurality of features from each directed acyclic graph, generating a binary graph classification model on captured features of each directed acyclic graph, determining a probability that a drug or treatment will be effective using the binary graph classification model, and determining a drug to be prescribed to a patient based on the determined probability.

The invention discloses a reference intention classification method based on a dynamic heterogeneous graph fusion network. The method comprises the following steps: dynamically extracting reference sentences and related contexts thereof, and combining chapter information to form a text T; utilizing a SciBERT pre-training model to extract a feature vector of the text T; constructing a heterogeneous graph to express a complex relationship among nodes such as papers, chapters and authors; node features are extracted through a dynamic heterogeneous graph network, and node feature vectors of a heterogeneous graph are generated; fusing the node feature vector with a text feature vector of the text T to form a comprehensive feature; and finally, predicting the reference intention by using a classifier, and optimizing the model through a loss function. According to the method, the semantic features with distinction are extracted from the internal structure of the paper and the cross-literature relation, high-precision citation intention recognition is achieved in a complex citation network, and the method is particularly suitable for citation intention classification tasks of academic literatures.

The invention relates to a graph model fine tuning method based on graph prompt learning, and aims to solve the problem that a graph model pre-training task is inconsistent with a downstream task target and improve the performance of tasks such as node classification and graph classification. The method comprises the following steps of: firstly, sampling graph data by restarting random walk, and learning a pre-training model through graph-level contrast to fully mine graph representation capability; secondly, aiming at a downstream node classification task, extracting a two-hop neighborhood construction induction graph for each target node, and selecting the most representative dominating node based on the minimum dominating set; further, in order to enhance hierarchy and discrimination of the graph structure, a corresponding sub-graph-level node is introduced for each dominating node, a global graph-level node is added, and a unified prompt graph is constructed by connecting the dominating nodes with the corresponding sub-graph-level nodes and connecting all the sub-graph-level nodes to the graph-level nodes. And then, in a representation generation stage, fusing information of different granularities through a multi-level graph embedding aggregation strategy, and weighting to obtain final graph-level embedding representation. And finally, performing similarity calculation on the graph representation and a predefined class prototype, and selecting a class with the highest similarity as a target node prediction result, thereby converting a node classification task into a graph classification task. According to the method, the performance is excellent when upstream and downstream task targets are unified, the pre-training encoder is frozen during fine adjustment, and only relevant parameters of the prompt weight are updated, so that the model performance, the adaptability and the practical value are remarkably improved.

The invention discloses an abnormal transaction dynamic detection method and system fusing multi-scale analysis and information entropy, and the method comprises the steps: S01, constructing a dynamic heterogeneous graph which comprises account nodes, equipment nodes and IP nodes; s02, extracting node features of each node, extracting a spectrum feature matrix of the dynamic heterogeneous graph, and forming a node feature vector of each node; step S03, clustering the dynamic heterogeneous graph to screen out candidate detection sub-graphs, inputting node feature vectors of nodes in the candidate detection sub-graphs into a pre-trained graph classification model, and identifying an abnormal account node set; and S04, calculating a corresponding dynamic risk score in real time according to the memory state vector of each abnormal account node in the time sequence diagram network and the diagram information entropy change rate so as to evaluate a real-time risk and identify a core abnormal account. According to the method, the group type abnormal transaction account can be quickly and accurately identified, and the core abnormal account can be positioned.

The invention discloses an autism classification algorithm based on combination of dynamic and static brain networks. A static brain network encoder effectively extracts global brain network features and generates functional connections with richer semantic information; a dynamic brain network encoder independently encodes each subsequence in the dynamic connection and fuses all codes according to an attention mechanism; and the graph classifier recombines the results of the first two modules into a high-order function connection brain graph and performs classification based on a GCN algorithm. According to the method, deep features in a static function connection matrix are effectively learned by introducing the static brain network encoder, and the dynamic brain network encoder module is introduced, so that the method has the capabilities of learning time dynamics of a brain network and integrating dynamic function connection, and the brain fitting capability of the model is greatly improved; and finally, through representation learning of a graph classifier, the model can deeply learn classification features of the brain to diseases from two aspects of a dynamic brain network and a static brain network, so that the classification performance of the model can be improved.

The invention discloses a smart contract state manipulation attack detection method, system and device based on a graph neural network, and a medium, and belongs to the technical field of block chain security analysis, and the method comprises the steps: defining a smart contract state manipulation attack type, collecting an attacked smart contract and a normal smart contract on a block chain, and constructing a training data set; positioning a creator of the target smart contract, and constructing a historical external transaction graph; decompiling the target smart contract, and constructing an external calling behavior graph; combining the historical external transaction graph with the external calling behavior graph to construct a creation behavior graph of the smart contract; constructing a classifier of a state manipulation attack; and monitoring the smart contract created in real time, constructing and creating a behavior graph, and inputting the behavior graph into the classifier to realize real-time detection of the state manipulation attack. According to the method, the blockchain nodes can be connected, the newly deployed smart contract is acquired in real time, and the smart contract creation behavior graph is constructed and input into the graph classification model to realize detection of the state manipulation attack smart contract.

The invention provides a graph classification method based on a quantum graph attention network. The method comprises the following steps: step 1, converting graph data into a quantum state for quantum calculation through angle coding and amplitude coding; step 2, establishing a quantum linear mapper, and for a quantum state obtained by angle coding, using a group of alternative trainable parameter-containing quantum gates and CNOT gates to form a group of unitary gates # imgabs0 # to evolve the quantum state; 3, establishing a quantum graph attention layer, splicing the nodes in pairs when the attention coefficient between the nodes is calculated, and iteratively inputting the spliced nodes into the quantum graph attention layer; step 4, taking nodes with small influence on the target node as noise items and removing the noise items; and 5, image classification is carried out. According to the method, strict dependence on a graph structure is avoided, the processing requirement of large-scale graph data can be better met, the method is a universal circuit design method, and the processing efficiency and flexibility of the quantum graph neural network are remarkably improved.

The invention discloses a graph classification method based on a spiking neural network, and the method comprises the steps: carrying out the aggregation of the node features of a graph G (V, E) through a graph convolution network, and obtaining the embedded representation of nodes; performing dimension conversion on the embedded representation through linear projection, and inputting the converted node representation X into a standard Transform layer; performing weight screening and sparsification through a sparse weight selection module to generate queries, keys and values; and carrying out dimension expansion and information enhancement on the Q and the K by utilizing a Kronecker product, and calculating the element similarity between the Q and the K through a Hadamard product. And sending the result to a pulse sequence generated by the leakage integral-issuing neurons, averaging the pulse sequences of a plurality of time steps to obtain an attention score, processing the attention score to obtain attention output, performing feature transformation by using a feedforward network to obtain final node representation, and finally converting Z into a classification result by using a classification layer. According to the method, the attention mechanism is combined with the pulse neural network, so that the expression ability of the attention mechanism can be kept, the utilization of computing resources is further optimized through the sparsity and discreteness of pulse signals, the parameter quantity in the model is remarkably reduced, and the accuracy of graph classification is improved.

The invention discloses an Ethereum Pincare contract detection method and system based on a runtime behavior graph. The method comprises the following steps: firstly, collecting runtime behavior information of a smart contract when the smart contract is executed; constructing a contract runtime behavior graph CRBG according to the runtime behavior information, wherein the CRBG comprises control flow and data flow information of the contract; and finally, carrying out graph classification analysis on the CRBG to identify a potential Pincare contract. By using the runtime behavior graph and the graph neural network model, the Pincare contract can be accurately identified, and the detection precision is obviously higher than that of a traditional method based on rules and static analysis. The method does not depend on a predefined rule in a specific field, and can meet the detection requirements of various novel Pincare contracts. Through a graph optimization technology, large-scale block chain data can be efficiently processed, and the method is suitable for Ethereum main network and test network environments.

The invention discloses a hyperspectral image classification method and device based on a subgraph adaptive neural network. The method comprises the following steps: S1, constructing a paired probability graph structure based on an SSAPGCN method; s2, reading a feature index and generating a feature matrix, performing PCA dimension reduction on the input data, and performing data preprocessing; s3, a global graph structure S is obtained based on dimension reduction preprocessing, a Metis graph segmentation algorithm is called for graph segmentation, and the global probability graph structure is divided into a plurality of sub-graphs; s4, calculating an intra-class distance intra (C) as a self-adaptive feedback threshold value, determining an optimal image convolution layer number k according to sub-image spectral difference, and avoiding over-smoothing or under-smoothing; and S5, training and testing the data by using the optimal image convolution layer number k to complete hyperspectral image classification. The method is mainly used for processing large image classification, the calculation cost can be reduced, spectrum and spatial features are fully mined, and the classification accuracy and efficiency are improved.

The invention relates to a test time training method based on invariant graph learning, and the method comprises the following steps: 1, a joint training stage, a joint training encoder fg, a main task classification head pi m and self-supervision task classification heads pi s1 and pi s2 are trained through a graph classification task and an auxiliary self-supervision task by using marked training set data, and the joint training stage is divided into two parts including invariant graph recognition and multi-level graph comparison learning; step 2, in a test time training stage, realizing self-adaption of the model on test data by minimizing the difference of feature distribution of a training domain and a test domain; and step 3, a test stage: in a model test stage, carrying out specific processes and strategies of model evaluation and fine adjustment by using a test sample. According to the scheme, through a test time training method, the model is finely adjusted by using label-free test data distribution, and the generalization performance of the model in a distribution offset scene is improved.

A domain-generalized person re-identification method and system based on meta-graph awareness includes: a ResNet50 network as the backbone network, a meta-global correlation awareness module, a meta-graph relationship sampling module, and a person matching module. This invention applies the Meta-Graph Aware (M-GRA) algorithm to domain-generalized person re-identification. In the meta-training domain, all block features are stacked using an affinity model to construct pairwise relationships. Then, a shallow convolutional model is used to learn this feature stacking relationship model, and a global knowledge graph is constructed. This classifies and weights structural information, suppressing noise while maintaining learning efficiency, and preventing model overfitting.

The application discloses a kind of graph classification training methods based on supervised contrast learning and structure inference, first, the model is fully mined by structure inference Topological information of graph data itself as originally graph data Enhanced graph.Then through label random sampling constructs sample positive-negative example pair, and using hierarchical graph neural network respectively to positive-negative example sample learning to extract whole graph embedding.Finally, through ordinary classification loss and supervised contrast learning loss jointly guide the learning process of hierarchical graph neural network, improve the classification performance of embedding.The application fuses structure inference and label information, the data enhancement mode based on structure inference does not need prior knowledge, expands the scope of model use, accelerates model learning speed.Self-supervised contrast learning on graph data is extended to contrast learning under label supervision, enhances its contrast learning ability.The application improves graph classification performance, and has good generalizability on generalized graph classification data.

The embodiment of the invention relates to a layout classification method, electronic equipment and a computer readable storage medium. The method comprises the following steps of: determining the similarity of each sub-layout on the basis of geometrical characteristics of graphs in a plurality of sub-layouts formed by segmenting the layout, wherein the sub-layouts have the same size; and classifying the sub-layouts based on the similarity to determine a classification number. According to the technical scheme, the layout can be efficiently classified.

Methods, systems, and non-transitory computer readable storage media are disclosed for dynamically generating and modifying interactive graph-based taxonomies associated with data processes in various domains. The disclosed system generates node data objects representing a domain-category hierarchy in connection with one or more computing data processes via a library of tools. The disclosed systems generates an attribute data object corresponding to an attribute assigned to a node data object in the graph-based taxonomy and links the attribute data object to node data objects according to parent / child relationships of the hierarchy. The disclosed system utilizes the parent / child relationships of the node data objects and attribute data object to aggregate attribute values of the attributes according to one or more aggregation operations. The disclosed systems provide indications of the aggregated attribute values for display via a graphical user interface for use in performing data processes.

This application relates to the field of integrated circuit design automation technology, and provides a method and system for verifying analog circuit structures based on graph neural networks. The method includes: converting a SPICE netlist into a graph structure representation, where nodes correspond to devices and edges correspond to electrical connections, and extracting feature vectors containing device type, parameters, functional roles, connection types, and signal directions; encoding the circuit diagram using a graph attention network to obtain node embeddings and graph embeddings; performing multi-task prediction based on the embedding results, including node classification, graph classification, structure scoring, and error localization; and finally generating a verification report containing an error list, location information, and repair suggestions. This application achieves automatic detection and precise localization of errors in analog circuit structures, improving detection accuracy by a significant margin and possessing strong generalization capabilities.

The invention relates to a design method for graph classification of a particle-ball-based quantum graph neural network, and belongs to the field of quantum machine learning. The method comprises the following steps: S1, a classic pellet generation method: generating a pellet sample from a data sample; s2, acquiring a node data set and an edge data set, and coding and preparing the node data set and the edge data set; s3, designing variable component sub-wire enhanced node features; s4, designing a quantum graph convolution line to realize information transmission and parameter sharing between nodes; s5, designing a quantum graph pooling line, extracting multiple pieces of quantum bit information to one quantum bit, and realizing feature dimension reduction; s6, inputting a training set to train parameters of the whole model; and S7, inputting a to-be-measured data set into the model, and executing Follis Z measurement on the specified quantum bits to obtain an expected value to finish final graph classification. According to the invention, the number of data sample points is reduced, and the defect that a huge amount of data cannot be prepared by a quantum experiment is overcome; and a quantum graph convolution line and a quantum graph pooling line are optimized.

The present application discloses a model training method and device, a text classification method, an electronic device and a storage medium. The method comprises: according to annotated text with a label, obtaining a relationship map with the annotated text as the central node, wherein the relationship map comprises neighbor nodes directly or indirectly associated with the central node, and the neighbor nodes comprise text nodes and entity nodes; inputting the text nodes in the relationship map into a text classification model to obtain a predicted pseudo-label and a predicted feature vector for each of the text nodes; according to the predicted pseudo-label, converging each text node with the central node to obtain a converged feature vector; using the converged feature vector as training data, training a graph classification model. The embodiment of the present application realizes an effective expansion of the annotated text, thereby improving the recognition accuracy of the trained model by adding a large amount of training data related to the annotated text.

The invention provides a node identification method and device based on graph classification and a related product, and relates to the technical field of data processing. The method comprises the following steps: calling a preset connected sub-graph recognition algorithm to recognize one or more connected sub-graphs of which the node number is greater than a set threshold value in a social network to obtain a connected sub-graph set; extracting a user node in the connected sub-graph, and extracting a local structure of the user node in the global network, namely a user case sub-graph taking the user node as a center; then classifying the user case sub-graphs by using a pre-trained graph convolutional network dichotomy model; in this way, user node recognition is converted into recognition of the local network where the user node is located, the user case sub-graphs with the user node as the center are classified through the pre-trained graph convolutional network dichotomy model, and therefore the classification result of the user node is obtained, the calculation amount is greatly reduced, and the recognition accuracy and efficiency are improved.

The invention relates to the field of molecular property prediction, in particular to a molecular property prediction robust heterogeneous graph classification method and device based on information bottleneck, and the method comprises the steps: obtaining a to-be-classified heterogeneous graph set which comprises at least one to-be-classified heterogeneous graph; the at least one to-be-classified heterogeneous graph is input into a pre-constructed robust heterogeneous graph classification model, the prediction probability of the category to which each to-be-classified heterogeneous graph belongs is obtained, and the pre-constructed robust heterogeneous graph classification model comprises a node semantic purifier and a global graph decoupling module; a node semantic purifier is used to generate robust heterostructures and node representations. Therefore, the problems that the node representation quality is reduced and the classification accuracy is relatively low under the confrontation attack in the related technology are solved, and the classification accuracy under the confrontation attack is improved.

The application discloses a graph classification method based on a saliency regularization graph neural network and related devices. The method comprises obtaining graph structure data corresponding to a target to be classified, and inputting the graph structure data into a saliency regularization graph neural network; determining a classification category of the target to be classified through the saliency regularization graph neural network; wherein the saliency regularization graph neural network learns a node feature matrix through a skeleton network, extracts the node feature matrix into a compact graph feature representation through a graph neural memory network, determines a saliency distribution vector based on the compact graph feature representation and the node feature matrix, and finally normalizes the aggregation weight of the skeleton network through the saliency distribution vector. In this way, the saliency regularization graph neural network focuses on the nodes more related to graph classification by measuring the compatibility between the entire compact graph feature representation and the node feature matrix, and can learn a more effective representation for the entire graph, thereby improving the classification effect of the graph classification task.

The present invention discloses a graph classification method based on fragmented data. The method includes: determining whether communication can be carried out between nodes according to the trajectory information and communication link information of the node fragmented data graph, and extracting the meta-structure model in the fragmented data graph; extracting the sub-graph structure from the extracted meta-structure model, and determining whether the meta-structures can be merged; finally, only retaining the nodes and meta-structures in the original graph that have mergeable nodes and meta-structures; the unmerged nodes and meta-structures are isolated nodes and isolated meta-structures, and the isolated nodes and isolated meta-structures are deleted from the original graph data set, and the remaining graph data is the constructed new graph data; the constructed new graph data is classified by using the self-adaptation based on spatio-temporal information. The fragmented data information attributes and spatio-temporal attributes are effectively utilized, and the interference of redundant information and noise is reduced.

The invention discloses a radar target recognition method and device based on a high-resolution one-dimensional range profile viewer. The method comprises the following steps: extracting a target time sequence in a preprocessed radar high-resolution one-dimensional range profile by using a noise threshold; obtaining a node set according to the amplitude value in the target time sequence; traversing slopes of a node i and a node j in the node set to obtain an edge set corresponding to the nodes in the target time sequence; wherein the range of the node i is 0 to L-1, the range of the node j is i + 1 to L-1, and L is the number of nodes in the node set; updating the node set for multiple times according to the edge set to obtain a plurality of node sets with different updating times; and performing graph classification according to the plurality of node sets with different updating times to obtain a classification result so as to realize target identification of the radar high-resolution one-dimensional range profile. The method has higher recognition accuracy, and improves the target recognition performance of a high-resolution radar system.

The invention discloses a graph classification method based on multi-view adaptive attention and feature integration enhancement, and aims to improve the representation capability and classification performance of graph structure data. An existing image classification method generally has the problems of single feature extraction view angle and inflexible fusion mechanism, and multi-level structure information in an image is difficult to fully mine. Therefore, the method has the following characteristics and contributions: firstly, a feature integration module fusing multi-graph convolution and one-dimensional convolution is constructed, graph node features are extracted from different view angles through multiple graph convolution, and self-adaptive integration of cross-view-angle features is realized by means of the one-dimensional convolution; secondly, designing a multi-angle adaptive channel attention mechanism, dynamically calculating weight distribution of each channel, and emphatically strengthening information expression of key semantic channels; finally, experimental verification on eight public graph classification data sets shows that the average classification accuracy of the method is improved by about 10% compared with that of a current advanced method, and the superiority and practicability of the method in a graph classification task are effectively proved.

The application discloses a personalized federated graph learning method and system suitable for cross-domain graph classification, and general graph spectrum knowledge is shared in view of the influence of structural heterogeneity in global cooperation, the general spectrum knowledge is shared in a global spectrum encoder including a global feature value encoder and a global filtering encoder, and customers benefit from the cooperation; wherein, other components representing non-general knowledge are reserved locally, each customer customizes personalized graph convolution for its own graph characteristics, and negative influence of spectrum bias is avoided; in view of the influence of structural heterogeneity in local application, a learnable preference is configured for each customer, a personalized preference module is used to cooperate with the work of the global spectrum encoder, personalized graph preference adjustment is performed, and the personalized graph preference adjustment is adapted to the unique graph structure of each customer; wherein, a regularization term is used to limit the personalized preference module to focus on local preference, and an over-reliance problem caused by the regularization term is solved.

To analyze images of more diverse graphs at higher speed and accuracy, disclosed herein is an information processing apparatus, comprising: a graph image acquisition unit configured to acquire a graph image; a graph classification unit configured to classify the graph image acquired by graph type; a probability map generation unit configured to generate, from the graph image, a probability map that is of a different type by the graph type; a component extraction unit configured to extract a plurality of components in the graph image based on the probability map generated by the probability map generation unit, and trace a graph with the plurality of components extracted; a value extraction unit configured to extract values of the components of the graph image extracted by the component extraction unit; and an output unit configured to output the values of the components extracted.

The present application relates to a kind of decoupling knowledge distillation hardware target detection method and system, method includes: according to data set mark frame in first model and second model respectively, decoupling is carried out to mark data set, and the foreground target area and background area are obtained, to determine the target kind probability of hardware, and according to target kind probability determine common feature score mask;According to common feature score mask and feature map classification score graph determine first distillation loss function;Based on global context module, according to the determination feature map, construct second distillation loss function;According to each distillation loss function, determine total loss function, and according to total loss function, the second model is trained, to carry out target detection to the hardware image to be measured.The present application uses the common shape feature of different kinds of hardware, adopts decoupling knowledge distillation method, for foreground target area, the common feature belonging to the unique hardware between is used for the information integration of target classification to migrate to small model, improves hardware detection precision.

The invention discloses a virus sample classification method and system based on a graph classification algorithm. The method comprises the following steps: firstly, collecting image samples of various virus categories, and carrying out preprocessing operation; secondly, performing data enhancement on the preprocessed image data, and constructing a classification model based on a VGG16 model architecture; training the constructed classification model, evaluating the performance of the classification model on a verification set, adjusting a training strategy according to an evaluation result, and optimizing the performance of the model; then acquiring a to-be-detected virus sample image, preprocessing the to-be-detected virus sample image, and classifying the to-be-detected virus sample image by using the trained virus sample classification model; and finally, outputting a classification result, and carrying out encrypted storage and safe deployment on the trained virus sample classification model. According to the method, the accuracy and efficiency of virus sample classification are effectively improved, the generalization ability of the model is enhanced, a large number of virus samples can be rapidly processed, and the real-time requirement in practical application is met.