Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

52 results about "Collaboration graph" patented technology

In mathematics and social science, a collaboration graph is a graph modeling some social network where the vertices represent participants of that network (usually individual people) and where two distinct participants are joined by an edge whenever there is a collaborative relationship between them of a particular kind. Collaboration graphs are used to measure the closeness of collaborative relationships between the participants of the network.

Vent travel element universe virtual character cooperation system and method based on large space interaction

The invention relates to the technical field of computer systems and human-computer interaction, in particular to a text travel element universe virtual role collaboration system and method based on large-space interaction. The method specifically comprises the following steps: tracking a user spatial position in real time through a multi-source positioning technology, and carrying out identity authentication and equipment legality verification by fusing multi-modal features; constructing a virtual-real fused semantic point cloud structural body, binding role interaction nodes, dynamically generating a task collaboration graph by using a graph neural network, and driving multiple roles to collaboratively execute subtasks; calling a graph attention network and a GPT module to deduce personalized plot branches in combination with user multi-modal input and a historical track, and rendering an immersive interactive scene in real time; spatial preferences and content interests are mined by adopting DBSCAN clustering and depth feature coding, a dynamic portrait is constructed, and a task path and content presentation are optimized through a weighted recommendation algorithm. According to the method, cloud rendering, edge calculation and cross-module cooperative control are integrated, and the interactivity and immersion experience of the text travel element universe are remarkably improved.
Owner:SHANGHAI YUANMENGTANG VIRTUAL REALITY INFORMATION TECHNOLOGY CO LTD

Automatic driving safety key simulation scene generation method based on adversarial generation and co-evolution

The invention discloses an automatic driving safety key simulation scene generation method based on adversarial generation and co-evolution. The method comprises the following steps: receiving a basic traffic scene described by a natural language, generating an antagonistic element scene containing security threats by using a large language model in combination with a traffic safety knowledge base, and analyzing the antagonistic element scene into an executable scene script; constructing a multi-agent confrontation collaboration diagram based on the meta-scene, and recognizing a key background vehicle through a cross-timing attention mechanism in combination with a time mask and time decay mechanism; and performing disturbance optimization on the key background vehicle trajectory to generate an automatic driving test scene. According to the method, a scientific and systematic solution with engineering operability is provided for safety verification of the automatic driving system when the automatic driving system faces real traffic challenges such as multi-source intervention and dynamic collaborative threat, and the method has wide adaptation capability and important industrial popularization value.
Owner:BEIHANG UNIV

Graphite sales data prediction method based on artificial intelligence

The invention belongs to the field of sales data prediction, and particularly relates to a graphite sales data prediction method based on artificial intelligence. The method comprises the following steps: firstly, acquiring a sales data set containing sales time, region, quantity and price; secondly, extracting periodic fluctuation characteristics of each region, dividing and defining a period through an endogenous period driven by data, and generating a period influence factor in combination with multi-dimensional sales characteristics; then, constructing a sales collaboration graph by taking the regions as nodes and the sales fluctuation synchronization degree as an edge weight, and extracting an embedded vector by using an improved graph convolutional neural network; multi-dimensional data such as embedded vectors are input into a fusion prediction model, and the model outputs a prediction value through a bidirectional gating circulation unit and a feature cross fusion module; and finally, the credibility is quantified and predicted, and a risk interval is determined for sales analysis and inventory planning. According to the method, the limitation of a traditional method is solved, and the prediction precision and interpretability are improved.
Owner:QINGDAO PINGYIHONG TECHNOLOGY CO LTD

Collaborative planning method and device for emergency rescue heterogeneous unmanned equipment

The invention belongs to the technical field of collaborative planning, and relates to a collaborative planning method and device for emergency rescue heterogeneous unmanned equipment. The method comprises the steps that an emergency rescue scene is acquired, multiple types of unmanned equipment are obtained, each type of unmanned equipment comprises multiple agents, and an overall cooperation mechanism of all the unmanned equipment is generated; according to responsibilities of different agents in the emergency rescue scene, generating node representation of each agent; obtaining an isomorphic distribution diagram and a heterogeneous collaboration diagram of each agent according to the overall collaboration mechanism of all unmanned equipment; obtaining a task graph of each agent according to responsibilities of different unmanned devices in the emergency rescue scene; inputting the node representation of the intelligent agent into the graph neural network to obtain a node graph embedding vector; and obtaining a decision vector of the unmanned equipment according to the node graph embedding vector, the isomorphic distribution graph, the heterogeneous cooperation graph and the task graph, and taking the decision vector as a cooperation planning scheme. According to the invention, the collaborative planning effect of heterogeneous unmanned equipment in an emergency rescue scene can be improved.
Owner:NAT UNIV OF DEFENSE TECH

Enterprise digital capability operation detection and evaluation method

The invention relates to the technical field of enterprise evaluation, in particular to an enterprise digital capability operation detection and evaluation method, which comprises the following steps of: collecting data from digital systems of a plurality of departments in an enterprise, and generating data flow information of each department; based on the data flow information of all the departments, a cross-department digital cooperation graph is constructed by adopting a graph model cooperation graph, nodes in the cooperation graph represent all the departments of the enterprise, and edges in the cooperation graph represent cooperation relations and data flow among all the departments; analyzing the collaboration map by using data flow network analysis, and identifying a data island of cross-department data flow; and constructing a cross-department cooperation efficiency scoring model, and calculating cooperation efficiency scores of all departments and the whole enterprise. According to the invention, an enterprise can accurately know the cooperation efficiency and the overall digitization capability level of each department, so that a reliable data basis is provided for a management layer, low-efficiency departments and cooperation bottlenecks can be found, and an accurate promotion plan can be made.
Owner:XINJIANG QUANTUM COMM TECH CO LTD

Multi-agent service collaboration method based on collaboration diagram and large language model

The invention discloses a multi-agent service collaboration method based on a collaboration diagram and a large language model, and the method comprises the steps: S10, collecting environment perception information, man-machine collaboration relation information and user demands containing a plurality of agents, and carrying out the preprocessing; s20, performing collaboration diagram construction on the preprocessed data to obtain a collaboration diagram; s30, performing large language model planning on the collaboration diagram to obtain a high-level task plan; s40, performing reinforcement learning skill optimization on the high-level task plan to obtain a specific action or an execution instruction; and S50, sending the specific action or the execution instruction to a corresponding agent for execution. According to the method, high-level task planning is carried out through the large language model, and low-level skill actions are optimized by utilizing a reinforcement learning technology, so that different types of agents efficiently cooperate, and convenience and comfort are provided for family life.
Owner:BEIHANG UNIV

Electronic component interaction control method and device, electronic equipment and storage medium

The invention provides an electronic component interaction control method and device, electronic equipment and a storage medium, and the method comprises the steps: collecting the interaction data of an electronic component from a plurality of sensors, and fusing the multi-modal feature vectors into a comprehensive feature vector through a weighted splicing fusion algorithm; constructing a modal relation mapping network, using an adaptive fusion network to combine the features of different modals, and using a deep self-attention mechanism to capture the relation between the modals; constructing a user behavior causal inference model, and inputting the user intention vector y into the user behavior causal inference model to infer potential behaviors of the user; constructing an inter-device collaboration graph model, adjusting resource allocation between devices, and outputting a resource allocation strategy; constructing an equipment anomaly detection model; and building a reinforcement learning model to dynamically adjust model parameters of the resource allocation strategy ropt, and dynamically adjusting environment changes and user requirements. According to the invention, comprehensive upgrading of interaction control of electronic component equipment is realized.
Owner:YUNZHOU INNOVATION TECH (GUANGZHOU) CO LTD

Collaborative management authority rule calculation method and system based on community discovery

A collaborative management authority rule optimization system based on community discovery comprises a feature extraction module, an organization role screening module and a data attribute screening and authority distribution control module. Generating a standardized user feature vector and a data vector through structured coding and semantic modeling; the organization role screening module constructs a user collaboration graph based on a community division algorithm, and reasones user optimal role mapping by using a community portrait model; the data attribute screening module is combined with a data classification tree and a structure dependency relationship extraction algorithm to recognize a semantic category and upstream and downstream dependency chains of target data; and the permission allocation control module performs reasoning in combination with a permission knowledge base according to roles, data classification and a dependency relationship, and finally generates a permission allocation table, so that dynamic permission configuration and strategy matching oriented to a collaborative scene are realized. Through automatic role mapping, data semantic association analysis and authority inheritance reasoning, accurate authority distribution and real-time strategy adjustment in a ship collaboration scene are realized, so that data security and collaboration efficiency are improved, and core support is provided for digital transformation of the ship industry.
Owner:SHANGHAI JIAOTONG UNIV

Cross-domain recommendation method, system and device based on comparative learning and storage medium

The invention provides a cross-domain recommendation method, system and device based on comparative learning and a storage medium. The method comprises the following steps: constructing a user-article collaboration graph; performing enhancement and de-noising processing on the collaborative graph to generate an enhanced and de-noised view; dividing positive and negative prototypes, fusing multiple views with the positive and negative prototypes through comparative learning, filtering noise, optimizing user preference feature representation, and finally constructing a cross-domain user preference model. According to the method, a noise elimination mechanism based on graph enhancement and graph denoising is provided, Gaussian resampling is used for the global situation to reserve graph structure core information, random mask is used for the local situation to dynamically filter low-quality edges, and edge quality evaluation is combined to achieve double denoising. According to the method, a user-prototype similarity division mechanism is introduced for the first time, user preferences are mapped to positive / negative prototypes, feature selection is guided by combining prototypes of comparative learning, the preferences are aggregated to the positive prototypes and are far away from the negative prototypes, the real preferences of the users are accurately modeled, and the basic data quality is improved.
Owner:ANHUI INVERNESS TECH CO LTD

Cooperative scheduling optimization method for operator fusion and grouping and fragmentation

The invention relates to the technical field of distributed online reasoning cluster scheduling, and discloses an operator fusion and grouping fragmentation collaborative scheduling optimization method, which comprises the following steps: S1, collecting operator execution events under a unified time base, recording operator identifiers, topological relations, fragmentation positions and resource occupation, and forming a joint execution event flow; and S2, constructing an operator layer graph based on the joint execution event flow, and identifying and fusing candidate fragments according to adjacent calling and intermediate data multiplexing. A joint execution event flow is collected under a unified time base, a space-time collaboration graph for superposing fragmentation boundaries and node resource states is constructed, and an operator fusion mode, a grouping fragmentation scheme and a scheduling strategy are jointly determined on the graph by taking a collaboration execution unit as granularity. Therefore, the same fusion candidate slice paragraph is explicitly constrained in the same slice and the same node in a multi-model mixed operation scene, cross-slice data migration and cache jitter are reduced, the balance of cluster resource utilization is improved, and the tail delay fluctuation of a service request is reduced.
Owner:马小凡

A collaborative personalized federated learning method and system in an asynchronous environment

This invention discloses a collaborative personalized federated learning method and system in an asynchronous environment. It constructs a fully asynchronous "client-server" architecture and initializes it. The client trains and uploads the trained model locally. The server estimates the collaborative graph of the uploaded model. Based on the estimation results and the model's staleness, the server's buffer is refreshed. The latest buffer is multicast based on the model's staleness. The total error of the collaborative graph is estimated in real time to adjust the current training state. The model received by the client continues to be trained locally until a preset convergence condition is met. Through a server-side buffering mechanism and a staleness weighting strategy, the client uploads an encrypted, delayed model, effectively solving the problem of severe inaccuracies in collaborative graph construction and model aggregation in asynchronous scenarios in personalized federated learning. It achieves a balance between clock time, communication overhead, and prediction accuracy.
Owner:BEIHANG UNIV

Intelligent voice conference behavior analysis method and system based on multi-source perception

The invention relates to the technical field of multi-source data processing and analysis, in particular to an intelligent voice conference behavior analysis method and system based on multi-source perception. The method comprises the following steps: acquiring an audio signal, cooperative behavior data and terminal equipment data, and constructing a time slice index data structure, a voice space positioning sequence and a terminal activation recording sequence; constructing an activation state corresponding matrix, and constructing a multi-source collaborative affiliation graph structure based on the voice space positioning sequence and the terminal activation record sequence to obtain a dominant speaking affiliation result; constructing a conference cooperation graph structure, and calculating the structural contribution strength of the non-voice input behavior to the conference decision content by using a structural coupling degree function to obtain a recessive behavior recognition result; and constructing a conference behavior credibility matrix, and eliminating repeated mapping records of conference participant identity multi-mapping conference decision content under the same time slice to obtain a conference behavior credibility identification result. According to the invention, identification and credibility evaluation of dominant speaking and implicit cooperation behaviors of participants are realized.
Owner:JIAN XIANGE ACOUSTIC ELECTRONIC CO LTD

A Hybrid Dynamic Negative Sampling Method for Knowledge Graph Recommendation Models

This invention discloses a hybrid dynamic negative sampling method for knowledge graph recommendation models, comprising the following steps: acquiring user-item bipartite graphs and knowledge graph data; then combining the user-item bipartite graphs and knowledge graphs to construct a user-item collaboration graph; and using the node2vec algorithm to select a fixed-size candidate negative sample set for all users in the user-item collaboration graph; inputting each user's positive sample data and the candidate negative sample set data into a knowledge graph-based recommendation model; performing positive mixing on each negative sample in the set to obtain enhanced negative samples; and then selecting and aggregating the enhanced negative sample set to construct hard negative samples. Using the constructed negative samples in the training process of existing knowledge graph recommendation models can significantly improve recommendation performance. Compared with existing negative sampling methods for knowledge graph recommendation models, the improvement effect is more significant and the flexibility is also higher.
Owner:XI AN JIAOTONG UNIV

An Unsupervised Fault Diagnosis Method and System for Rolling Bearings Based on Multi-Level Feature Collaboration Graphs

A method and system for unsupervised fault diagnosis of rolling bearings based on multi-layer feature co-matrix is ​​disclosed. The method involves: acquiring vibration signals and extracting time-frequency fault feature vectors; using a multi-layer feature extraction network to obtain shallow, medium, and deep fault features; performing multiple base clustering operations on each level of features; constructing a co-co-matrix based on the frequency of sample pairs belonging to the same cluster to form a fault sample relationship graph; inputting the co-co-matrix and fault feature vectors into a graph attention network; and achieving unsupervised fault diagnosis through self-supervised clustering optimization. The system comprises a vibration signal acquisition and time-frequency fault feature extraction module, a multi-layer fault feature extraction module, a multi-layer base clustering and co-co-matrix construction module, a graph attention unsupervised clustering diagnosis module, and a diagnosis result output module, all connected to form a fault diagnosis system. This invention overcomes the problems of unstable clustering in a single feature space and susceptibility to noise interference in the relationship graph, and is suitable for intelligent fault diagnosis of rolling bearings under label-less operating conditions.
Owner:QINGDAO UNIV OF TECH

Personalized federal medical image segmentation method based on uncertainty collaboration diagram and double-layer aggregation mechanism

The invention discloses a personalized federal medical image segmentation method based on an uncertainty collaboration diagram and a double-layer aggregation mechanism. The method comprises the following steps: 1, initializing hyper-parameters of a server side and a client side; 2, preprocessing medical image data of a client; 3, training a basic segmentation network model of the client; 4, constructing a target function of the client and updating the client to obtain model parameters; 5, calculating an uncertainty value of each client, and uploading the uncertainty value and a client model parameter to a server; 6, constructing a target function used for optimizing the uncertainty collaboration diagram and updating an adjacent matrix of the uncertainty collaboration diagram; 7, on the basis of the optimized uncertainty collaboration diagram, constructing a double-layer aggregation mechanism at a server side; 8, the client receives the global model of the server and updates parameters; 9, repeating the steps 3-8, and performing iterative training; and 10, outputting a personalized segmentation model of each client, accurately screening a reliable client cooperation relationship, and improving segmentation precision and generalization ability.
Owner:SHAANXI UNIV OF SCI & TECH

Personalized federal learning training method based on construction of collaboration diagram between clients

The invention discloses a personalized federated learning training method based on construction of a collaboration diagram between clients, and belongs to the technical field of federated learning. The method comprises the core steps of 1) calculating a feature distribution distance between clients based on KL divergence, 2) constructing a directed collaboration graph based on normalization of a calculated adjacency matrix, 3) aggregating a model with structural information based on the directed graph, and 4) performing local optimization. The method mainly relates to challenges confronted by data distribution heterogeneity, in particular to potential relation modeling among client data distribution, the challenges of data heterogeneity and the like are relieved by constructing a client collaboration diagram and providing a collaboration diagram assisted personalized aggregation and training method, and knowledge sharing in personalized federal learning is promoted.
Owner:INSTITUTE OF INFORMATION ENGINEERING CHINESE ACADEMY OF SCIENCES

A multi-party collaborative graph learning method and system based on material data privacy protection

The application relates to a multi-party cooperation graph learning method and system based on material data privacy protection. In the learning method, multiple participants independently perform machine learning, and model parameters are uploaded to a server for joint training, so that material data is prevented from being leaked in the training process. Noise disturbance is performed on the model parameter transmission process, the leakage of the model parameters is avoided, and the graph node relationship matrix is set to participate in the training of the local graph learning model, the dependence relationship information of a material preparation process is fully considered, and the performance of the obtained model can be improved.
Owner:SHANGHAI UNIV

Three-pillar driven cloud collaborative multi-agent deep reinforcement learning decision system

The present application relates to a kind of three pillar drives cloud coordination multi-agent deep reinforcement learning decision system, belong to agent decision field.The system of the present application includes: end side execution module, cloud side teacher guidance module, cloud side action cooperation module, cloud side parameter sharing module and cloud side global optimization module.The present application utilizes cloud coordination mode, establishes dynamic cooperation graph based on self-attention mechanism, utilizes federal cooperation mechanism to realize the network parameter sharing of same task area, adopts width learning optimization strategy network, establishes the three pillar architecture of dynamic cooperation, parameter sharing and global optimization to guarantee the convergence of multi-agent deep reinforcement learning;The present application can accelerate the convergence speed of multi-agent deep reinforcement learning, enhance the coordination between agent, improve the stability of system.
Owner:BEIJING INST OF COMP TECH & APPL

Intelligent machine group collaboration method and system based on collaboration diagram

The invention discloses an intelligent machine group collaboration method and system based on a collaboration diagram, and is applied to the field of intelligent machine group collaboration and vehicle-road intelligent collaboration. According to the system, a sensing processing module, a cooperative control module and a cooperative cognition module are deployed on each intelligent machine participating in cooperation, and the cooperative control module comprises a cooperative diagram generation and management module, a cooperative diagram-based cooperative control module and a cooperative diagram-based feature segmentation module. According to the method, a cooperation capability diagram is updated in real time according to sensing data, a cooperation demand diagram is generated, effective data transmission between intelligent machines is carried out through a cooperation order diagram, the intelligent machines only focus on processing and transmission of cooperation feature data, and a demand side carries out multi-source feature fusion to obtain comprehensive features for subsequent decision making. According to the method, the collaboration diagram is innovatively introduced for interaction, the inherent limitations of a single intelligent machine on perception capability and communication resources are effectively overcome through a mode of screening and transmitting high-value sparse features, and the method has advantages in a communication limited scene.
Owner:BEIJING UNIV OF POSTS & TELECOMM

Scene-adaptive multi-vehicle cooperative sensing method

The invention provides a scene-adaptive multi-vehicle cooperative sensing method, which comprises the following steps: firstly, designing a dynamic reasoning BEV encoder, and adaptively adjusting a feature extraction mode by a laser radar branch and a camera branch according to scene complexity; and then, a modal consistency constrained Transform fusion module implicitly aligns laser radar and camera BEV features by using cosine similarity loss, and realizes cross-modal context sensing through self-attention. And finally, the flexibility of multi-vehicle cooperative sensing is improved by combining the edge weight coded cooperative diagram with pose sensing feature transformation. Experimental results on two authoritative data sets, namely, OPV2V and DAIR-V2X, show that the method provided by the invention realizes dual optimization of reasoning precision and speed in different complexity scenes, and effectively improves the precision and efficiency of 3D target detection.
Owner:NANJING TECH UNIV

Collaborative management permission rule optimization system based on community discovery

A collaborative management permission rule optimization system based on community discovery includes a feature extraction module, an organization role screening module, a data attribute screening and permission allocation control module, the feature extraction module generates standardized user feature vectors and data vectors through structured coding and semantic modeling according to user data, collaboration objects and data attribute information; the organization role screening module constructs a user collaboration graph based on a community division algorithm and infers the optimal role mapping of the user using a community portrait model; the data attribute screening module identifies the semantic category and upstream and downstream dependency chain of the target data in combination with a data classification tree and a structure dependency relationship extraction algorithm; the permission allocation control module infers in combination with a permission knowledge base according to the role, data classification and dependency relationship, and finally generates a permission allocation table, realizing dynamic permission configuration and strategy matching for collaborative scenarios. The present application realizes precise permission allocation and real-time strategy adjustment in the ship collaborative scenario through automatic role mapping, data semantic association analysis and permission inheritance reasoning, thereby improving data security and collaboration efficiency and providing core support for the digital transformation of the ship industry.
Owner:SHANGHAI JIAOTONG UNIV

Decentralized medical federal learning method based on dynamic collaboration diagram

The invention provides a decentration medical federal learning method based on a dynamic collaboration diagram, which comprises the steps of performing multiple rounds of federal training on a plurality of client models, and each round comprises the following steps: acquiring each client model and a global collaboration diagram which comprises all client nodes and a plurality of edges, each edge represents a directed edge that one client model can be aggregated by another client; dividing each client into a first class or a second class, and selecting a client model meeting a knowledge contribution degree requirement from the graph for a first class of client models; according to the client model selected for the class of clients, updating edges pointing to nodes of the class of clients in the graph; according to the updated global collaboration diagram, performing knowledge aggregation on all client models needing to be aggregated of each client to obtain an aggregation model; and replacing the classification layer of the client model with the classification layer of the aggregation model to obtain a new model, and updating the parameters of the new model based on the classification loss of the new model and the characterization difference loss between the new model and the aggregation model.
Owner:INST OF COMPUTING TECH CHINESE ACAD OF SCI

Student adaptive learning recommendation system based on deep learning

The invention belongs to the technical field of learning recommendation, and particularly relates to a student adaptive learning recommendation system based on deep learning, which comprises a data processing module, a learning integration module, a cognition judgment module and an adaptive recommendation module, according to the system, by designing an L2 normalized linear attention mechanism, learning resource information is effectively integrated, and a learning resource theme is clarified; a context perception collaboration graph is constructed, multi-aspect cognitive states of the learner are judged, learning data are deeply analyzed, and learning recommendation accuracy is improved; the method comprises the following steps: establishing an SASRec model as a learning model, introducing an expert mixed self-attention layer, decomposing learning preferences of learners, optimizing and training the learning model by using a global cross entropy loss function, introducing a gradient scaling algorithm to update a parameter set, providing personalized recommendation according to related features of learning resources, continuously optimizing a learning resource recommendation list, and finally, carrying out learning resource recommendation. Therefore, the learning resource recommendation list better fits the actual learning condition, and the learning efficiency of the learner is improved.
Owner:SHANDONG POLYTECHNIC COLLEGE

Trusted data space low-latency interaction platform fused with edge computing

The invention discloses a trusted data space low-delay interaction platform fused with edge computing, and relates to the technical field of edge computing, and the platform comprises the steps: constructing a hierarchical trusted execution environment in a target edge computing network based on a preset trusted execution environment layering rule; constructing an equipment cooperation relation graph of the target edge computing network, dividing the equipment cooperation relation graph into a plurality of real-time levels, and configuring a differentiated key pre-distribution strategy for each real-time level; based on behavior characteristic pre-screening and real-time connection performance evaluation, performing dual node screening on network nodes, determining federated learning participation nodes, and obtaining an elastic federated learning scheme; and collaboratively integrating a hierarchical trusted execution environment, a differentiated key pre-distribution strategy and an elastic federated learning scheme, and executing low-delay interaction management. The problems that in the prior art, the low-delay requirement and the safety of data interaction in the edge computing network cannot be effectively met are solved.
Owner:LINGSHU TECH CO LTD

Self-supervised learning multi-modal recommendation method and system based on diffusion model

The invention discloses a self-supervised learning multi-modal recommendation method and system based on a diffusion model. The method comprises the steps that loss functions adopted during training of a multi-modal recommendation model comprise the steps that a normalized interaction matrix of user articles is used for generating a user article collaboration diagram through a diffusion model, ID guiding is carried out, and modal semantic generality is extracted so as to obtain an ID-guided collaboration diagram and collaboration diagrams guided by all modals; and respectively calculating the loss represented by the ID guide mode and the loss for enhancing semantic consistency, and summing to obtain the self-supervised loss. The invention aims to generate interaction data by using a diffusion model so as to inject modal information and ID information, compare the injected data to meet modal alignment and ID-guided feature representation, and inject modal perception cooperation signals in self-supervised learning. Therefore, the recommendation accuracy of the multi-modal recommendation system is improved by obtaining effective self-supervision loss.
Owner:HUNAN NORMAL UNIVERSITY

Multi-role data access control system for industry collaboration

PendingCN121966972AImplement data access controlavoid safety hazardsSecuring communicationData streamData interface
The invention relates to a multi-role data access control system for industrial collaboration, and discloses a strategy generation mechanism driven by a main body relation collaboration graph and a collaboration contract, a joint judgment model and a unified executable constraint packet and multi-execution-point forcing mechanism. And the multi-role data access control system realizes fine control on business data, report data, process data, interface calling and log data. The system is characterized by comprising a main body and identity module, a resource registration and object modeling module, a collaborative contract and strategy instance module, a strategy and control module, a process linkage and life cycle module and a full-link audit and evidence storage module, and a collaborative relation graph between the main body and the resource entity is constructed, and the resource registration and object modeling module outputs a structured resource object.
Owner:安徽恒源煤电股份有限公司

Encrypted knowledge graph

A computer-implemented method for managing access rights to a knowledge graph is provided. The method comprises splitting, for each user system, its respective portion of the knowledge graph into a plurality of knowledge subgraphs, encrypting each of the knowledge subgraphs, and generating a plurality of private summary graphs. The method also comprises maintaining a collaboration graph comprising one vertex per user system and edges representing collaborations between the users, mapping all private subgraphs of all user systems to one public summary graph, each vertex of the public summary graph comprises less data than the related vertex of the related private summary graphs and wherein none of the vertices of the summary graph comprises any encryption or decryption key, and granting access to a selected knowledge subgraph from a first user system to a second user system.
Owner:KYNDRYL INC

A personalized federated learning training method based on constructing inter-client collaboration graph

The application discloses a kind of based on the personalized federated learning training method of constructing inter-client cooperation graph, belong to federated learning technical field.The core steps of the application include:1) based on the feature distribution distance between client based on KL divergence calculation,2) based on the construction of normalized directed cooperation graph after adjacency matrix calculation,3) the model with structure information based on the aggregation of directed graph,4) local optimization.The application is mainly related to the challenge faced by data distribution heterogeneity, especially related to the modeling of potential relationship between client data distribution, by constructing client cooperation graph and proposing cooperation graph assisted personalized aggregation and training method, alleviate the challenge such as data heterogeneity, promote knowledge sharing in personalized federated learning.
Owner:INSTITUTE OF INFORMATION ENGINEERING CHINESE ACADEMY OF SCIENCES

A collaborative method and system for intelligent machine swarm collaboration based on collaboration graphs

This invention discloses a collaborative graph-based method and system for intelligent machine swarm collaboration, applicable to the fields of intelligent machine swarm collaboration and vehicle-road intelligent collaboration. The system deploys perception processing, collaborative control, and collaborative cognition modules on each participating intelligent machine. The collaborative control module includes a collaborative graph generation and management module, a collaborative graph-based collaborative control module, and a collaborative graph-based feature segmentation module. The method updates the collaborative capability graph in real time based on perception data, generating a collaborative demand graph. Effective data transmission between intelligent machines is achieved through the collaborative ordering graph, allowing intelligent machines to focus solely on processing and transmitting collaborative feature data. The demanding party performs multi-source feature fusion to obtain comprehensive features for subsequent decision-making. This invention innovatively introduces a collaborative graph for interaction, effectively overcoming the inherent limitations of individual intelligent machines in perception capabilities and communication resources by filtering and transmitting high-value sparse features, thus offering advantages in communication-constrained scenarios.
Owner:BEIJING UNIV OF POSTS & TELECOMM

A recommended method, system and medium for handling power distribution network faults

This invention discloses a method, system, and medium for recommending fault handling measures in power distribution networks, belonging to the field of intelligent operation and maintenance technology for power systems. The method includes: constructing a fault-handling bipartite graph, a fault-fault collaboration graph, and a handling-handling collaboration graph; using a pre-trained language model to complete semantic encoding of fault text and handling text, obtaining initial semantic embeddings of entities; using a graph convolutional network with residual connections to propagate homogeneous node features in the fault-fault collaboration graph and the handling-handling collaboration graph, and using a bidirectional graph attention network to adaptively allocate neighbor node weights to aggregate heterogeneous node features in the bipartite heterogeneous graph; introducing a multi-head attention mechanism to adaptively fuse heterogeneous features from multiple graphs to obtain a unified entity feature representation; pooling and aggregating the fault set features, calculating the matching probability of each handling measure using a bilinear matching function, and outputting a fault handling recommendation scheme sorted in descending order of matching probabilities, thereby achieving accurate recommendation of fault handling measures in power distribution networks.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO LTD RESEARCH INSTITUTE +1