20 results about "Relation discovery" patented technology

The invention provides a business modeling method, medium and system based on an enterprise data asset management platform, and belongs to the technical field of modeling methods.The business modeling method based on the enterprise data asset management platform comprises the following steps that an enterprise data asset catalog is read to generate a data asset list, the method comprises the steps of extracting a foreign key association relationship between data tables to construct an initial entity relationship table, calculating reference frequencies between the data tables to generate an entity association degree matrix, dividing a service domain by adopting a K-means clustering algorithm, counting data table fields in the service domain to generate a data table field mapping table, recording a data table operation sequence to construct a data table operation link diagram, and establishing a data table operation link diagram. The method comprises the following steps: generating a data table node evaluation matrix, dividing sub-domains and determining a core entity, and constructing and training a multilayer neural network model for constructing business affiliation, data table incidence relation and business interface definition of a new data table, so as to solve the defects in the aspects of data asset relation discovery, business modeling automation, service interface generation and the like in the prior art.

The invention relates to the technical field of people relationship mining, in particular to a multi-dimensional people relationship discovery method based on multi-modal information fusion, which comprises the following steps: extracting text, visual and audio modal features from a video; according to the Euclidean distance of the modal features, optimal alignment path optimization and nonlinear time adjustment are carried out, and the modal features of time step alignment are obtained; performing self-attention and cross-attention mechanism processing on the modal features to obtain multi-modal representations of texts, vision and audios; multi-modal representation of texts, vision and audios is carried out, and a graph structure network of a social relation, a standing relation and a co-occurrence relation is established; determining the social relationship closeness degree, the standing consistency degree and the co-occurrence strength of each person in the video according to the graph structures of the social relationship, the standing relationship and the co-occurrence relationship; according to the method, the understanding accuracy of the multi-dimensional character relationship can be improved.

The application discloses a kind of individual and group-oriented behavior internet modeling and prediction recommendation method, the method includes the following steps: step 1: establish the behavior preference model of behavior internet based on hypergraph;Step 2: extract the relationship between behaviors based on the model of step 1;Step 3: discover the behavior mode of user based on the behavior relationship of step 2;Step 4: predict the behavior trend of user based on the behavior relationship of step 2 and the behavior mode of step 3;Step 5: personalized service recommendation is carried out for user based on step 2, step 3, step 4.The model of behavior internet is proposed in the application, and the personalized behavior internet is obtained by cause-effect inference and relationship discovery, and the recommendation is carried out by using the deep learning method based on external knowledge, which improves the recommendation efficiency and effect for individual and group, overcomes the shortcomings that the traditional recommendation method based on deep learning is difficult to fully utilize personalized knowledge, and realizes the personalized and efficient recommendation of service.

The invention relates to a campus card swiping log-based student companion relationship discovery method, which comprises the following implementation steps of: firstly, preprocessing an original campus card swiping log data set, uniquely identifying students and places, and combining card swiping records in the same time period to obtain a dining log data set; then, according to a set time-space co-occurrence constraint condition, screening out a card swiping co-occurrence record meeting the time-space co-occurrence constraint condition, and counting the card swiping co-occurrence frequency of the student pair; screening to obtain a high-frequency card swiping co-occurrence student pair data set according to a set high-frequency co-occurrence threshold value; and finally, merging student pairs with a mutual connection relationship to obtain a high-frequency card swiping co-occurrence student group data set, performing visualization for a specified student group, and analyzing a dynamic change condition of a companion relationship among multiple students in the student group. The method can effectively discover the companion relationship among the students, tracks and analyzes the dynamic change process of the companion relationship along with time, and provides support for improving the education management level.

The invention relates to the technical field of knowledge graphs, and provides a semantic graph and structure graph fused entity relationship automatic generation method, which comprises the following steps: constructing a semantic graph according to text content, and constructing a structure graph according to a text structure; calculating the semantic association degree between each keyword vector and each text paragraph structure vector; performing semantic association degree sorting on keywords related to each text paragraph structure to construct an association degree matrix; calculating the semantic correlation between every two keywords, and constructing a plurality of undirected relation pairs according to the semantic correlation so as to automatically screen and generate an entity set in the semantic correlation knowledge graph; based on the text evidence and each undirected relation pair, performing relation calculation on two keywords in each undirected relation pair to construct a semantic association knowledge graph; a large model hint project for relationship discovery is improved based on a high quality relationship sample library. According to the method, entity relationship automatic labeling can be carried out on massive and dynamically changing text contents.

The application discloses a Web table abnormal data discovery method based on a text semantic mapping relationship. The application aims at discovering abnormal data with fuzzy or even wrong semantic information in a Web table. The method mainly comprises three parts: a semantic representation module, a column type inference module and an error discovery module. First, the semantic representation module represents the meaning of cell text. For a cell in a table, the string text in the cell is represented as a semantic vector according to context information. Then, the column type inference module infers the type of the column where the cell is located, and obtains the mode information of the column. Finally, based on the mapping relationship between the column type and the semantic vector of the cell text of the main column cell and the target cell, abnormal data in the table is discovered and labeled.

This invention relates to the field of causal relationship discovery technology for electromechanical equipment monitoring data, specifically to a method for causal relationship discovery based on hierarchical knowledge embedding in electromechanical equipment monitoring data. The method includes: collecting electromechanical equipment monitoring data; establishing a hierarchical knowledge constraint matrix based on the knowledge information of the electromechanical equipment; establishing a causal discovery model; weighting and adjusting the causal strength calculated by the causal discovery model based on the hierarchical knowledge constraint matrix, or masking and modulating the attention matrix of the causal discovery model based on the hierarchical knowledge constraint matrix; processing the electromechanical equipment monitoring data based on the knowledge-embedded causal discovery model to obtain a causal network adjacency matrix; performing conflict edge detection based on the hierarchical knowledge constraint matrix, updating the causal network adjacency matrix, and using the updated causal network adjacency matrix as the causal relationship analysis result. This invention can improve the performance of causal relationship identification in monitoring data.

The application discloses a user active relationship discovery method and system for information recommendation, and the method comprises the following steps: constructing a user relationship multi-attribute heterogeneous graph; calculating the relationship score between a target user and other users based on the relationship between the vertices in the user relationship multi-attribute heterogeneous graph; obtaining the active relationship group of the target user according to a preset score threshold and the relationship score between the target user and other users; determining the N users most similar to the target user from the active relationship group of the target user based on a TOP-N algorithm, and taking the N users most similar to the target user as the active relationship users for information recommendation of the target user. The active relationship group of the target user is calculated according to the association relationship between users, so that the discovery of the active relationship of the user is realized, and the accuracy of information recommendation is further improved.

The invention provides a causal relationship discovery method and system and related equipment, and the method comprises the following steps: the causal relationship discovery system divides a data set into a plurality of category groups according to the number of groups input by a user, enables objects in the category groups to be objects sharing a causal graph, and uses different category groups to train different sub-models, the causal diagrams of the observation data in the corresponding category groups can be mined by different sub-models, so that a plurality of causal diagrams corresponding to the data set are obtained, and the problem that a current causal discovery algorithm can only identify a single causal diagram can be solved.

The invention provides a training method of an archive data reconstruction model, an archive data anomaly detection method and device, equipment and a medium. The method comprises the steps of obtaining an archive record to be detected; vectorizing the archive record to be detected to obtain a corresponding input vector; inputting the input vector into an archive data reconstruction model obtained based on normal archive record unsupervised training to obtain a corresponding reconstruction vector; calculating a reconstruction error between the input vector and the reconstruction vector; and when the reconstruction error is greater than a preset error threshold value, determining that the to-be-detected archive record is abnormal data. Through the pre-trained archive data reconstruction model, the internal association and non-linear relationship of each field in the archive record can be identified, novel anomalies of unknown types and undefined rules are found, the limitation of traditional hard coding rules is broken through, and accurate detection of deep and hidden defects in archive data is realized.

The present application relates to the technical field of intelligent analysis of financial data, and particularly relates to a method for automatically discovering reconciliation constraints for high-sparse heterogeneous financial data tables. The present application identifies a locally consistent record set by record clustering and field sub-cluster generation based on a field non-empty mode, then discovers reconciliation relationships within a locally effective sample range, and then introduces an inverse frequency weight and enhances sparse fields, so that fields that only appear in specific business scenarios but play a key role in reconciliation structure identification are not easily covered by high-frequency fields. The common non-empty constraints and splitting of sufficient field candidate sub-clusters are performed to gradually shrink the candidate field set, avoid irrelevant fields from being mixed into the same linear solution space, and improve the concentration and interpretability of numerical reconciliation relationship discovery.

The invention relates to the technical field of networks, and discloses a micro-service dependency discovery method, device, equipment and medium, and the method comprises the steps: continuously collecting the communication direction of each communication among all application examples deployed in a container cluster in N time windows and the traffic characteristic data of the communication, according to the metadata of the application instance in any time window, the communication direction of each communication and the flow characteristic data, constructing a directed graph representing the communication relationship among all the application instances in the container cluster in the time window, and inputting the directed graphs corresponding to all the time windows into a trained model subjected to micro-service dependency relationship discovery so as to find the micro-service dependency relationship. And obtaining the dependency relationship among the micro-services corresponding to all the application examples in the N time window inner container clusters. The technical scheme provided by the invention can be applied to business management systems such as financial science and technology, medical treatment, health and pension and the like based on the micro-service architecture, dynamic discovery of the micro-service dependency relationship is realized, and the real-time performance and the accuracy of discovery of the micro-service dependency relationship are improved.

The invention aims to provide an automatic identification method for an asset network relationship and a device for realizing the method. The method is realized by using SNMP + ARP, a routing table of each network device is accessed through the SNMP, and a connection relation of the network devices is drawn according to a corresponding relation of IP network segments; and discovering the subnet relationship through the corresponding relationship between the IP and the MAC of the ARP table. Therefore, the survival assets can be searched, the connection relation and the asset information can be identified, the security management efficiency of the assets can be improved, and the network security threats can be reduced.

A task relationship determination method, device and equipment are used to determine the association relationship between tasks that can be automatically determined. In the present application, the relationship discovery device can first determine the sample attributes in each task and the corresponding attribute values. Then, according to the sample attributes and the corresponding attribute values of each task, the task attributes of each task and the corresponding attribute values are determined. After determining each task attribute and the corresponding attribute value, the hybrid attributes of each task are determined based on the sample attributes and / or the task attributes of each task. Then, the associated tasks, i.e. the association relationship between tasks, are determined based on the task hybrid attributes of each task. In the whole process, it does not depend on relevant personnel and can efficiently determine the associated tasks.

The invention provides an open type relation extraction method and device based on a large language model, and the method comprises the steps: constructing a corresponding example set and an instruction prompt for each training example in a training data set, and enabling the example set to comprise examples which belong to the same relation with the corresponding training example; inputting each training instance, the corresponding instance set and the instruction prompt into a first large language model for training; a corresponding example set and an instruction prompt are constructed for each training example, and the example set does not comprise examples belonging to the same relationship with the corresponding training example; inputting each training instance, the corresponding instance set and the instruction prompt into the second large language model, and adopting the trained relation predictor as a teacher to guide the second large language model to train; performing new relationship prediction on the to-be-tested instance by adopting the trained relationship discoverer and relationship tester; therefore, the accuracy of relation prediction is effectively improved.

The invention discloses a key information infrastructure association evaluation system and method based on a large language model, and the method comprises the steps: firstly extracting information from multi-source heterogeneous data through an entity extraction and relationship discovery engine based on large language model driving, and constructing and updating a unified risk knowledge graph; then, parallelly starting intrusion trace analysis and service logic security analysis, and writing an analysis result back to the knowledge graph for enriching; thirdly, calculating and generating an attack path from an attack entrance to a core target and a corresponding script based on the knowledge graph containing the threat analysis result; and finally, performing multi-mode empirical verification on the generated attack script, and feeding back a verification result to the knowledge graph to form a closed loop. According to the scheme provided by the invention, risk discovery, association analysis, attack simulation and closed-loop verification of the key information infrastructure can be automatically carried out, isolated security data points are converted into linked attack paths, and the evaluation depth, breadth and efficiency are remarkably improved.

The invention discloses a dynamic visualization system and method for a certificate asset relation topology. The system comprises a data input layer, a relation discovery module, a topology calculation engine, a visual rendering module, an interaction control module and a real-time monitoring module. The data input layer collects certificates, IT assets and configuration data; the relationship discovery module adopts a depth-first search algorithm to identify relationships of deployment, dependency, protection and the like of certificates and IT assets, and stores the relationships after triple verification of domain name matching, configuration analysis and network detection; the topology calculation engine realizes layout optimization based on an improved force-oriented layout algorithm; the visual rendering module adopts WebGLGPU acceleration and LOD technology to distinguish asset types and relationships through visual coding; the interaction control module provides interaction such as dragging and zooming. According to the method, dynamic visualization of the certificate and IT asset relationship is realized, the management efficiency is improved, the certificate change risk is reduced, the large-scale data visualization performance is improved, and the method is suitable for certificate asset relationship management of a complex IT environment of an enterprise.

The present invention provides a method and system for discovering a unidirectional common relationship of blockchain addresses, the method comprising the following steps: obtaining an initial address; obtaining a candidate address having a transaction relationship with the initial address based on the initial address, and constructing a first relationship graph based on the initial address and the candidate address; dividing the initial address and the candidate address in the first relationship graph into at least one node group based on the connection relationship in the first relationship graph; for each node group, obtaining an affiliated address having a transaction relationship with the initial address or the candidate address in the node group, and constructing a second relationship graph based on the initial address, the candidate address and the affiliated address of each node group; calculating the embedding vector of each node based on the basic index of each node in the second relationship graph, constructing the node corresponding to the initial address and the node corresponding to the candidate address in the second relationship graph into a node pair, calculating the similarity of the two nodes in the node pair based on the embedding vector, and determining whether the candidate address is a regulatory address.

The application relates to the field of marine information technology and data processing, and discloses a marine survey historical data correlation analysis and retrieval system based on a knowledge graph, which comprises a self-adaptive relation kernel compiling engine, a user query intention, a field semantic graph, a dynamically selected and compiled relation kernel encapsulating a field algorithm, a generated relation discovery execution plan, a dynamic hypergraph construction module executing the plan, a calculation of generated relation instances representing high-order correlations as hyperedges by calling the relation kernel to act on original data, a dynamically constructed dynamic hypergraph with space-time data field nodes as vertices and the relation instances as hyperedges, and a pattern matching on the dynamic hypergraph to retrieve complex correlation patterns and provide traceable analysis results. The application realizes the calculation-driven dynamic construction of data relations, can efficiently and flexibly discover and analyze high-order correlations in marine space-time data.

Embodiments of the present disclosure provides a fine-grained image recognition method and apparatus using graph structure represented high-order relation discovery, wherein the method includes: inputting an image to be classified into a convolutional neural network feature extractor with multiple stages, extracting two layers of network feature graphs in the last stage, constructing a hybrid high-order attention module according to the network feature graphs, and forming a high-order feature vector pool according to the hybrid high-order attention module, using each vector in the vector pool as a node, and utilizing semantic similarity among high-order features to form representative vector nodes in groups, and performing global pooling on the representative vector nodes to obtain classification vectors, and obtaining a fine-grained classification result through a fully connected layer and a classifier based on the classification vectors.