139 results about "Bipartite graph" patented technology

The invention discloses an optimization solution system for a large-scale mixed integer linear programming problem, and the system is characterized in that the system comprises a problem decomposition module which is used for modeling the large-scale mixed integer linear programming problem into a bipartite graph for representation, and dividing the bipartite graph into a plurality of sub-graphs through a graph decomposition algorithm; the problem reduction module is used for obtaining a feature code of each sub-graph node, inputting the feature code into a pre-training neural network to obtain a variable prediction value, reducing the scale of the large-scale mixed integer linear programming problem through a redundancy constraint removal and coefficient priority variable fixing strategy based on the variable prediction value, and then outputting the problem; and the efficient solving module is used for solving the large-scale mixed integer linear programming problem based on the result output by the problem reduction module and weighted subgraph division. According to the method, the high-quality solution of the large-scale mixed integer linear programming problem can be quickly given, so that the high-quality feasible solution can be obtained within the acceptable time.

The invention discloses a medical image clustering method based on correlation entropy and an adaptive bipartite graph, and the method comprises the steps: collecting image data of medical imaging equipment, and forming a sample data matrix; selecting anchor point samples from the sample data matrix, and establishing an original sample low-dimensional representation-anchor point low-dimensional representation graph for capturing a local geometric structure of an embedded space; introducing an orthogonal basis matrix; carrying out joint modeling on the sample reconstruction error, the structure retentivity, the adaptive graph constraint and the discriminative features by adopting a joint objective function; iteratively updating the joint objective function; and inputting the finally obtained low-dimensional representation of the original sample as an embedded representation into a clustering device of K-means, and completing final clustering label distribution. According to the method, more stable, efficient and interpretable unsupervised clustering is realized, the clustering precision and generalization ability are improved, and the method is particularly suitable for medical image data analysis tasks of high-dimensional and complex structures.

The invention provides a personalized semantic understanding learning method, medium and system under an AI platform, and belongs to the technical field of semantic understanding. According to the technical scheme, the personalized semantic understanding learning method comprises the steps that a personalized semantic file recording user question and answer habits and personal terms is constructed, and dual time decay values are set; a semantic understanding optimization model is established based on an attention mechanism, personalized feature vectors and input semantic vectors are fused, a concept drift detection algorithm adopting bipartite graph maximum matching and a Hungary algorithm is designed to monitor semantic habit changes in real time, and an incremental learning mechanism based on a variable sliding window is established to calculate a semantic difference matrix. A forgetting function based on a Gaussian kernel is applied to adjust the historical semantic feature weight according to the time distance and the use frequency, and a reinforcement learning feedback module is constructed to collect user satisfaction evaluation and generate reward signal optimization model parameters; and executing a self-iterative optimization process to periodically update semantic archives and models so as to continuously adapt to personalized semantic requirements of users.

The invention provides a method and a system for accurately converting a large-model natural language into an SQL (Structured Query Language) based on multi-modal fusion. The method comprises the following steps: S1, performing named entity recognition and dependency syntax analysis on a natural language text, performing vision-text dual-path processing on image data, and constructing a bipartite graph structure for table data so as to obtain a vision text composite feature and a table feature; s2, designing a modal attention controller and calculating the weight of each modal through a gating network; s3, constructing a knowledge graph, clearly recording all legal table names and column names in a database, and retrieving the knowledge graph in real time when a decoder generates the table names or the column names; s4, generating an SQL draft; s5, performing bidirectional verification correction on the SQL statement; and S6, outputting the legal SQL statement. According to the method, multi-modal information can be integrated, information loss and semantic deviation are reduced, illegal statements are avoided through grammar and semantic dual verification, and the overall reliability of SQL statement generation is improved.

Methods, systems, and computer program products are provided for bipartite graph pre-trained dual transformers. Graph data for a bipartite graph is received, the bipartite graph including a first node / entity, a second node / entity, and an edge. The first entity context of each first entity includes a second entity connected to the first entity through the edge. The second entity context of each second entity / node includes the first entity connected to the second entity through the edge. A first embedding of each first entity is generated. A first encoded representation of each first entity is generated based on the first entity context and a first transformer encoder. A second embedding is generated for each second entity. A second encoded representation of each second entity is generated based on the second entity context and a second transformer encoder. The first embedding and the first transformer encoder are adjusted based on a first contrast loss. The second embedding and the second transformer encoder are adjusted based on a second contrast loss.

The invention discloses a large language model alignment method based on multi-user preference dynamic balance, and the method comprises the steps: collecting user interaction data, preference comparison data and public policy question and answer data in a livelihood consultation scene, and dividing the data into structured preference data and unstructured data; preprocessing the divided data to generate a user feature vector and a response feature vector; constructing a user-response bipartite graph, performing multi-hop preference propagation through a graph neural network, capturing potential association among users, and outputting a user preference embedding vector; embedding a vector selection expert path according to user preference through a dynamic gating function, and generating a personalized response; and through an optimization-free embedding aggregation strategy, similar users are retrieved based on a graph structure, and the embedding of the similar users is weighted and aggregated, so that rapid cold start of new users is realized. The precision and efficiency of livelihood consultation can be remarkably improved, and the method is particularly suitable for smart city scenes with high-frequency policy updating and frequent user flow.

The application relates to a recognition method for matching scene behaviors by using multi-modal features, and belongs to the technical field of scene behavior matching. The method acquires multi-modal scene behavior data, and carries out noise self-adaptive purification processing on the multi-modal scene behavior data, so as to obtain preprocessed scene image data, scene audio data and scene label text. Then, feature collaborative extraction is carried out on the preprocessed data, so as to obtain visual features, audio features and text features. The collaboratively extracted features are input into a scene behavior matching network, scene feature fusion and behavior feature fusion are respectively carried out, scene feature vectors and behavior feature vectors are obtained, a bipartite graph is constructed according to the scene feature vectors and the behavior feature vectors, the semantic similarity between nodes of the bipartite graph is calculated, the edge weight of the bipartite graph is dynamically updated according to the semantic similarity between the nodes, the updated bipartite graph is normalized, and a scene behavior matching result is obtained. The application can accurately quantify the correlation degree of scenes and behaviors, and greatly improves the accuracy of the matching result.

The application relates to a drug-lncRNA relationship prediction method and system based on embedding constraints, and the method comprises the following steps: collecting lncRNA-drug association data and preprocessing, and constructing a data set; based on the data set, an lncRNA-drug association matrix is constructed, and an lncRNA-drug bipartite graph is extracted; based on the lncRNA-drug association matrix, an lncRNA similarity matrix and a drug similarity matrix are obtained, and dimension reduction processing is performed; the lncRNA-drug bipartite graph and the dimension reduction processed lncRNA similarity matrix and drug similarity matrix are input into an LDA-GNN model and an LDA-DHG model with an embedding constraint strategy, and the probability score of lncRNA and drug association is obtained; and based on the probability score, the relationship prediction of the drug-lncRNA is realized. Through the embedding constraint strategy, the prediction performance and the distinguishing ability of the model are significantly improved.

A method for graph neural network-based recommendation is disclosed. The method comprises the following steps: receiving a user-project bipartite graph, wherein the user-project bipartite graph comprises a user node set representing a user set, a project node set representing a project set, and an edge set representing interaction between users and projects; obtaining user embedding of each user in the user set and project embedding of each project in the project set; performing an intra-layer aggregation process to generate an aggregated user embedding of each user in the user set and an aggregated item embedding of each item in the item set based on different weight coefficients of each user and each item; executing an interlayer propagation process to generate a propagation user embedding of each user in the user set and a propagation item embedding of each item in the item set based on different weight coefficients of each propagation layer; and making recommendations for the users in the user set based on the propagation user embedding and the propagation item embedding.

The invention discloses a classification and grading semantic enhanced dynamic graph community offset data access anomaly detection method and device. The method comprises the following steps: slicing an internal access log according to a time window, constructing a bipartite graph snapshot consisting of user nodes and data asset nodes, and classifying and grading sensitivity semantics for asset association; edge risk weights are calculated according to access statistical characteristics and sensitivity, and risk weights are introduced in graph neural network message passing to learn node embedding; community division is carried out based on embedding in each time slice, a stable community evolution trajectory is obtained through time sequence smoothing, a community offset degree and a node-community consistency deviation are calculated to form a node anomaly score, and an alarm is output. According to the scheme, structural anomalies such as collusion access, abnormal cross-community access and community migration can be identified, and interpretable clues on the community level are provided.

The present invention relates to a method for selecting training data for training a deep learning model, the method comprising: with reference to a bipartite graph in which each of multiple training images stored in a data pool and individual types corresponding thereto are matched, performing a cycle of selecting some training images matching all of the individual types by repeating a process of (i) identifying, among the individual types, the number of corresponding individual types matching each of the training images, and selecting a specific training image having the largest number of corresponding individual types, and (ii) identifying, among individual types remaining after excluding the corresponding individual types, the number of the remaining corresponding individual types matching each of the training images, and selecting another specific training image having the largest number of the remaining corresponding individual types; and repeating the process of performing the cycle until n training images for training the deep learning model are selected from among training images remaining after excluding some training images selected from among the multiple training images.

The invention provides an attack technology prediction method and device based on a large language model. The method comprises the steps of obtaining historical attack data of an attacker and performing data modeling to construct a bipartite graph; generating a structured context according to the bigraph, and inputting the structured context into a large language model; and constructing a prediction cue word, guiding the large language model to carry out attack technology prediction according to the structured context by the prediction cue word to obtain a preliminary prediction result, and carrying out attack stage filtering and sorting screening on the preliminary prediction result according to a set standard to obtain a final attack technology prediction result. By applying the method, the recommendation ability of the large language model can be stimulated, the adaptability and learning ability of the large language model to a specific security task can be improved, the scene learning ability of the large language model can be fully utilized, a new attack mode can be effectively dealt with, the dependence on a large amount of annotation data is reduced, and the user experience is improved. Particularly, under the condition that data is scarce or the cold start problem is serious, the accuracy of the generated result can be improved.

The application provides an AI intelligent calculation platform reasoning resource scheduling method and system, relates to the technical field of computer resource management and task scheduling, and obtains task load information of a to-be-reasoned task, memory allocation records of an AI intelligent calculation platform in a preset time window, and a retransmission event timing sequence of a bus data link layer; obtains the discrete entropy of each computing node through a Shannon entropy formula according to the memory allocation records; counts the retransmission times of the retransmission event timing sequence in a sliding time window to generate a communication congestion index of each computing node; determines a weight coefficient according to the task load information, constructs a weighted bipartite graph, and determines a matching edge set that satisfies the most matching edges and the smallest sum of edge weights in the weighted bipartite graph by using a Hungarian algorithm, so as to be used for resource scheduling of the to-be-reasoned task. The problem of throughput decline caused by serious memory fragmentation and communication interference in the large model concurrent reasoning scene is solved.

The invention provides a packet consistency research method for a cooperation-competition network discrete heterogeneous multi-agent system with packet loss, and the method is characterized in that the method specifically comprises the following steps: 1, determining a dynamic model of the heterogeneous multi-agent system; 2, designing a packet consistency control protocol about packet loss; 3, obtaining a closed-loop form of the kinetic model according to the control protocol in the step 2; 4, assuming that the topological structure of the heterogeneous multi-agent system is a bipartite graph containing a spanning tree, and setting establishment conditions; and 5, proving that the establishment condition is established, and indicating that the closed-loop form of the dynamic model can realize progressive grouping consistency.

The present application relates to a kind of coal mine underground number of people statistics method based on two-part graph, belong to artificial intelligence field.The image of coal mine entry or underground key place is used as input data, yolov5 is used as the main network of target detection, interval time T is obtained, and two-part graph of personnel in the coal mine entry area is formed, Hopcroft-Karp algorithm is used to match personnel at T1 and T2 time, and the number of people entering and leaving is counted;The present application is based on yolov5 lightweight personnel detection model, can realize rapid iteration and implementation deployment in field application under the condition of meeting the prediction accuracy.

This application belongs to the field of Ethereum transaction technology, and particularly relates to a high-precision Ethereum phishing account detection method based on high-order topology. The detection method includes: acquiring transaction information containing historical phishing nodes, and then cleaning it to obtain several qualified transaction records; constructing a training node graph based on the qualified transaction records; constructing a corresponding multi-order graph based on the training node graph; the multi-order graph includes a bipartite graph and a classic pairwise graph, the bipartite graph containing the edge relationships between the simplex nodes corresponding to each maximal clique in the training node graph and the original nodes; the classic pairwise graph containing the edge relationships between the original nodes; obtaining the optimal influence score of the current classification model based on the multi-order graph; calculating the spliced ​​feature matrix of the current actual Ethereum node graph based on the optimal influence score; and the current classification model predicting phishing nodes based on the current actual Ethereum node graph and the corresponding spliced ​​feature matrix. This application can accurately detect phishing nodes.

The invention discloses a bipartite graph-oriented maximum balance k-biplex search method. The method comprises the following steps of: 1, setting an MBKBP (Maximum Balanced Kalman Binary Pattern) model corresponding to the maximum balance k-biplex; step 2, based on an MBKBP model, proposing a scale-limited interlaced search framework; step 3, optimizing by using a plurality of pruning methods; the pruning method comprises enumeration pruning, vertex pruning and graph reduction pruning. According to the method, the strict connectivity requirement of MBBC is relaxed, the robustness is higher, and the flexibility and applicability of the model are high. According to the method, a scale-limited staggered search framework and three types of pruning optimization rules are provided, so that the search space can be remarkably reduced.

The embodiments of the present application relate to the field of artificial intelligence. Provided are a product recommendation method and apparatus, a model training method and apparatus, and a device. The product recommendation method in the present application comprises: acquiring a bipartite graph corresponding to at least one user and at least one product, and on the basis of a relational graph convolutional network of an attention mechanism, performing convolutional processing on the bipartite graph, so as to obtain a set of representations; determining a user-product pair distance corresponding to a user-product pair representation, and on the basis of the user-product pair distance and the bipartite graph, determining a structural feature corresponding to the user-product pair representation; on the basis of user-product pair representations and structural features, generating a set of predicted information; and on the basis of the set of predicted information, recommending products to the user. The method in the present application can precisely recommend products to users.

The application discloses a multi-view graph Transformer cognitive assessment system and method fusing explicit features and a storage medium. The system comprises a feature embedding module, which is used for generating basic representations for three types of entities, namely learners, questions and knowledge concepts, and fusing difficulty coefficients of questions and knowledge concepts into their representations respectively; an intra-view information aggregation module, which is used for constructing three heterogeneous bipartite graphs, namely learner-question, learner-knowledge concept and question-knowledge concept, and updating node representations based on graph Transformer; a multi-view feature fusion module, which is used for fusing representations of each entity in different views to generate specialized representations; and a cognitive assessment module, which is used for mapping the specialized representations of learners to a vector space with the number of knowledge concepts as the dimension, and the obtained vector is the cognitive assessment result and is used for predicting the probability of answering any question correctly. The application fuses explicit difficulty, models multi-view graph structure and interacts with graph Transformer in depth, thereby significantly improving the accuracy and interpretability of cognitive assessment.

The application provides a user screening method and device based on graph contrast learning, equipment and storage medium, which can be applied to the field of artificial intelligence or the field of finance. The method comprises the following steps: generating a plurality of weight binary tuples comprising users, financial products and user products according to user behavior records; constructing a bipartite graph network by using the plurality of binary tuples; taking the bipartite graph network as input, training the graph network model based on a supervised learning method and an unsupervised contrast learning method, and obtaining a trained graph network model; processing the bipartite graph network by using the trained graph network model, and obtaining a screening index of each first node; and taking the screening index as the basis for screening the users represented by each first node. According to the scheme, the bipartite graph network is constructed, the graph network model is analyzed, the screening index of each user is automatically determined, and therefore the high-value users can be automatically screened according to the screening index.

The invention relates to the technical field of financial science and technology and big data processing, and discloses an enterprise credit report automatic acquisition and credit business co-processing system, which converts a request into a context vector through a business request analysis module; the dynamic strategy route generation module calculates an acquisition priority based on the vector and generates a targeted task configuration file; the multi-source heterogeneous data acquisition module executes data acquisition; the data conflict arbitration module constructs a bipartite graph topological structure, eliminates conflicts among sources by using trust propagation iteration, and outputs cleaned feature vectors; the credit risk quantitative evaluation module calculates a credit score based on nonlinear mapping; and when the score falls into a gray interval, the gray closed-loop feedback module calculates decision ambiguity and feeds back an offset correction vector, and triggers supplementary collection of the uncollected data source. According to the method, on-demand data acquisition and truth value automatic discovery are realized, the boundary sample decision problem is effectively solved through a gray closed-loop mechanism, and the automation level of a credit system is improved.

An information processing system, which predicts an unknown binary relation between a treatment method and a biomarker based on a known ternary relation among the treatment method, the biomarker, and a disease, generates for each disease, based on the known ternary relation, a disease-specific bipartite graph that represents the binary relation between the treatment method and the biomarker, calculates, based on the disease-specific bipartite graph, a disease-specific inter-treatment-method similarity between treatment methods, a cross-disease inter-treatment-method similarity between the treatment methods, a disease-specific inter-biomarker similarity between biomarkers, and a cross-disease inter-biomarker similarity between the biomarkers, and calculates and outputs a disease-specific prediction score and a cross-disease prediction score of an unknown edge.

The application provides a task allocation method applied to a UAV platform based on sequential ordering collaborative bidding, which comprises the following steps: dividing the UAV platform and disaster relief sites into a first point set and a second point set; representing the contact between the first point set and the second point set by a bipartite graph; allocating target tasks between the first UAV platform and the first site based on a preset bidding algorithm; the first UAV platform is any UAV platform in the first point set; the first site is any disaster relief site in the second point set; determining a bidding priority parameter of the target task; generating a bidding sequence of the UAV platform based on the bidding priority parameter and a preset roulette method; adopting a preset bid bidding algorithm to bid for the target task in the order of the bidding sequence to determine the target task with the highest bid; and performing target value decay and tolerance constraint processing on the target task with the highest bid to obtain a target UAV platform allocated with multiple target tasks.

The invention discloses a social perception maximization method based on differentiatable graph optimization, and relates to the technical field of data mining and network analysis, and the method comprises the steps: obtaining social network cascade data, and constructing a weighted bigraph of a user perception event; constructing a social perception forward propagation model according to the weighted bigraph, defining an activation state variable for each user, defining a total activation amount for each event, and introducing a function to evaluate the perception quality of each event; the activation state variables of all the users form an activation vector, the process of searching the optimal activation vector is constructed into a differentiable optimization task, and the optimization target is to maximize the perceived quality sum of all the events; and according to the optimal activation vector, generating a final seed user set as an optimal solution for maximizing social perception. According to the method, the problems of local optimization and model simplification in the prior art are solved, global optimization is realized, and the user influence can be accurately modeled.

Embodiments of the present application provide an information prediction method and device, a prediction equipment, a program product and a storage medium, which are applied to the field of artificial intelligence technology. The method comprises: obtaining a target bipartite graph and initial embedding vectors of each node in the target bipartite graph based on original business data, the target bipartite graph being used to represent the association relationship between users and behavior characteristics; generating a weight value corresponding to each node based on an adaptive weight mechanism according to the initial embedding vectors of each node, the weight value being used to reflect the importance of the node to the prediction of user dissatisfaction; performing node aggregation based on the weight value corresponding to each node to generate a target embedding vector of each node; and obtaining a prediction result for the user dissatisfaction according to the target embedding vector of each node. By using the method, the problem of low prediction accuracy of existing information prediction methods for user dissatisfaction is solved.

This invention discloses a single-lane vehicle trajectory association method based on multiple geomagnetic sensors, primarily addressing the problems of complex current vehicle trajectory association algorithms, high cost of sensors such as cameras and radar, and susceptibility to weather conditions. This invention deploys multiple geomagnetic sensors at equal intervals along one side of a single lane, detecting vehicle arrival timestamps and uploading them to a computing device. The computing device calculates the predicted timestamp of the vehicle's arrival at the geomagnetic sensor, constructs a bipartite graph model of the predicted vehicle timestamp and the geomagnetic sensor measurement timestamp, and uses the KM algorithm to solve the bipartite graph model to be associated. This invention achieves the association between vehicle prediction data and geomagnetic sensor measurement data, thereby enabling vehicle state updates and achieving real-time association of single-lane vehicle trajectories, obtaining accurate single-lane vehicle trajectory information.

A system and method for training a recommender system (RS) are provided. The RS is used to make recommendations over a bipartite graph, the bipartite graph including multiple user nodes, multiple item nodes, and an observation graph topology defining edges connecting at least some of the user nodes to some of the item nodes. The RS includes an existing graph neural network (GNN) model configured with an existing parameter set. The method includes: applying a loss function to compute an updated parameter set of an updated GNN model, the updated GNN model being trained with a new graph using a first parameter set as initialization parameters; the loss function being used to distill knowledge based on node embeddings generated by the existing GNN model for the existing graph, wherein the new graph includes multiple user nodes and multiple item nodes also included in the existing graph; and replacing the existing GNN model of the RS with the updated GNN model.

The application belongs to the technical field of heliostat calibration, and particularly relates to a tower type optical thermal mirror field multi-tower cooperative heliostat calibration scheduling method, the core of which is to model the selection problem of each batch of heliostats to be calibrated as a bipartite graph maximum flow problem with capacity constraints, and iteratively solve it, so that the number of heliostats calibrated simultaneously in each batch is maximized, thereby globally minimizing the total batch number and accelerating the heliostat calibration efficiency.

Described are a method, system, and computer program product for generating robust graph neural networks using universal adversarial training. The method includes receiving a graph neural network (GNN) model and a bipartite graph including an adjacency matrix, initializing model parameters of the GNN model, initializing perturbation parameters, and sampling a subgraph of a complementary graph based on the bipartite graph. The method further includes repeating until convergence of the model parameters: drawing a random variable from a uniform distribution; generating a universal perturbation matrix based on the subgraph, the random variable, and the perturbation parameters; determining Bayesian Personalized Ranking (BPR) loss by inputting the bipartite graph and the universal perturbation matrix to the GNN model; updating the perturbation parameters based on stochastic gradient ascent; and updating the model parameters based on stochastic gradient descent. The method further includes, in response to convergence of the model parameters, outputting the model parameters.

The application discloses a knowledge graph recommendation method fusing GNN and ResNet, and comprises the following steps: embedding user representation in a user-item bipartite graph and item representation in a knowledge graph through a GNN model; adding a residual connection between output vectors of each layer of the GNN model, so that the node representations of each layer have distinguishability, thereby ensuring the uniqueness and uniqueness of each node; obtaining high-quality difficult negative samples containing rich information based on a combination of gated interpolation mixing and hierarchical sampling; performing addition and pooling operations on the user representation and the item representation to generate user embedding representation and item embedding representation for prediction; obtaining a predicted matching score through the user embedding representation and the item embedding representation, and performing recommendation based on the matching score. The application effectively alleviates the oversmoothing problem caused by the increase of the number of GNN layers, improves the generalization ability and robustness of the recommendation model, and makes the recommendation result more accurate and personalized.