56 results about "Shared knowledge" patented technology

The invention discloses a multi-agent cooperation enhancement method, system and equipment based on a knowledge graph, and the method comprises the steps: obtaining original data in an external environment, carrying out the preprocessing and feature extraction of the original data, generating a knowledge triple, storing the knowledge triple in a local knowledge graph, and submitting the knowledge triple to a shared knowledge graph for knowledge updating; when a to-be-executed task is received, decomposing the to-be-executed task by utilizing the large language model and querying global knowledge in the shared knowledge graph and local knowledge in the local knowledge graph to obtain a plurality of sub-tasks; a bipartite graph minimum cost matching algorithm is adopted to match a plurality of sub-tasks with the capability and availability of each agent to generate a preliminary task allocation scheme, and a large language model is utilized to optimize the preliminary task allocation scheme to generate an optimal task allocation scheme; and sending each task allocation knowledge fragment in the optimal task allocation scheme to a corresponding agent for collaborative execution through a semantic communication protocol.

The invention relates to the technical field of lithium battery safety management, and discloses a lithium battery fault diagnosis method based on a BMS, and the method comprises the steps: obtaining multi-dimensional battery operation data and real-time data, firstly extracting a local feature vector, generating a preliminary fault signal indication, then extracting an abnormal feature vector according to the preliminary fault signal indication, and carrying out the fault diagnosis according to the abnormal feature vector; and if the preset threshold is exceeded, compressing and transmitting to the adjacent management unit to form a shared data packet. According to the local feature vector and the shared data packet, updating a diagnosis model parameter to obtain an optimized fault recognition model for analyzing real-time data and calculating a fault matching degree, and determining a potential fault type if a threshold value is exceeded; and generating a collaborative query request to the distributed network to obtain a historical fault empirical data set, integrating the historical fault empirical data set, refining parameters to obtain an accurate fault probability, activating an alarm and recording a log if a warning threshold is exceeded, and finally updating the global shared knowledge base. According to the method, the problem of insufficient lithium battery fault diagnosis accuracy in a distributed scene is solved, and collaborative optimization of diagnosis accuracy and distributed collaboration is realized.

The invention provides an incomplete multi-modal learning method based on prompt distillation, and aims to improve the robustness and generalization ability of a multi-modal model in a modal missing scene. According to the method, the advantages of knowledge distillation and prompt learning are fused, a collaborative framework composed of a hierarchical prompt generator, a teacher network and an inference network is constructed, and efficient migration of inter-modal shared knowledge and sample-level fine-grained information is realized. The method comprises the following steps: firstly, constructing and preprocessing a multi-modal data set; based on a prompt distillation mechanism, an efficient knowledge transmission mechanism is established between the teacher network and the reasoning network; and finally, inference is performed on the incomplete modal sample by using the trained prompt generator and the inference network, so that the accuracy and the adaptive capacity of downstream tasks are remarkably improved. According to the method, through collaborative optimization of the prompt generator and the double-distillation mechanism, the knowledge migration efficiency is remarkably improved while the light weight of the model is kept, and an efficient solution is provided for multi-modal learning in a complex real environment.

The invention provides a septal tumor accurate diagnosis system based on artificial intelligence, a multi-scale attention fusion module receives multi-modal data, focuses a septal region, dynamically adjusts the weight and outputs a key feature vector, an adversarial self-supervision pre-training module digs tumor morphological features based on the vector, robustness and generalization ability are enhanced, and the accuracy of diagnosis is improved. And the causal intervention diagnosis decision-making module constructs a decision-making tree by using enhanced features to perform preliminary diagnosis and transmits causal information to the meta-learning enhanced diagnosis module, and the meta-learning enhanced diagnosis module adjusts parameters by using a small number of samples for rare subtypes, outputs a final diagnosis result and feeds back the final diagnosis result. According to the system, multi-mode and dynamic attention, cooperative confrontation self-supervision pre-training and causal reasoning are fused, rare tumor diagnosis is optimized through meta-learning, a closed-loop feedback and shared knowledge graph is formed, and the diagnosis precision, interpretability and rare case diagnosis capacity are integrally improved.

The invention provides a self-adjusting dialogue type multi-agent system and interaction method based on dynamic topology, and the system comprises a scheduling module which is used for receiving user input and routing a task to a corresponding field task agent in a conventional state; the domain task intelligent agents are used for processing the received tasks, evaluating the cognitive states in real time and sending out assistance signals when the cognitive states meet preset conditions; the shared knowledge management module is used for maintaining a dynamically updated distributed semantic knowledge graph; wherein the distributed semantic knowledge graph is used for sharing, associating and persisting structured knowledge entities and context information generated in the session process among all modules in real time; the topology control module is used for dynamically switching the communication control topology of the system from a first topology structure to a second topology structure in response to the received assistance signal; and when the assistance signal is eliminated, the communication control topology is recovered from the second topology structure to the first topology structure.

The invention relates to the technical field of model-driven system engineering modeling, in particular to an MBSE-oriented multi-agent collaborative automatic modeling method and system. The method comprises the following steps: acquiring a natural language modeling demand of a user, retrieving in a shared knowledge environment to obtain a knowledge sub-graph related to the demand, constructing a modeling context, and calling a generation agent to generate a current SysML v2 model; calling a verification agent to perform grammar verification and semantic verification based on the domain knowledge graph on the model, and outputting a verification result; when a verification result represents that grammar does not pass or semantic inconsistency exists or an engineering constraint rule is violated, calling a repair agent to execute deterministic repair and iterative verification according to the domain knowledge graph until a preset condition is met, and outputting a final SysML v2 model; notch marking or pocket bottom repairing can be carried out under the condition that the notch is covered. According to the method, the correctness, interpretability and convergence efficiency of complex system modeling are improved.

The application discloses a robot cognitive development method based on ontology semantics, and comprises the following steps: constructing a robot article identification professional knowledge base based on attribute function and ontology information representation of article definition; information determination based on attribute discrimination and semantic search; and robot cognitive development based on attribute information addition. The application simulates the process of human memory, learning and cognition of articles based on an ontology semantic knowledge base, and through machine learning and sensor attribute information, the robot can actively cognize, learn, expand and accumulate learned knowledge and experience according to information data; the robot can continuously develop its cognitive ability through learning, automatically construct a robot article identification professional knowledge base of unknown articles, and based on a semantic structure of triplets, share knowledge and exchange operation logic between man and machine, so that the cognitive level of the robot is improved and the operation experience of an operator is improved.

The invention discloses a multi-modal knowledge base system construction method oriented to a large language model, which comprises the following steps: S1, creating a knowledge base instance, including setting a knowledge base name, description and visibility parameters; s2, adding a multi-modal document to the knowledge base, wherein the multi-modal document comprises a text, a PDF (Portable Document Format), an image and a table format; s3, configuring large language model parameters including a model identifier, an API key and a basic API address; s4, realizing a knowledge-based question and answer function, and enhancing a generation technology by combining a standardized API (Application Program Interface) with a retrieval; s5, realizing a dynamic scoring function of knowledge entries, and performing multi-dimensional scoring on the entries through a large language model; and S6, managing knowledge base authority, distinguishing a personal knowledge base mode and a shared knowledge base mode, and setting document visibility. According to the method, a multi-modal analysis engine is adopted, unified storage of multi-modal data is supported, lightweight document indexing is adopted, automatic scoring is performed through LLM, manual auditing dependency is reduced, and meanwhile high-quality knowledge sharing is ensured in combination with permission control.

The embodiment of the invention provides a business processing system based on multiple agents, and relates to the technical field of data processing. The system comprises an agent layer which is integrated with a plurality of professional agents, the professional agents comprise a business agent and a task scheduling agent, and the task scheduling agent is used for receiving a task request, dynamically allocating a task to a service agent corresponding to the task based on the task request, monitoring a processing state and a system load of each service agent, optimizing a task execution path, and coordinating cooperative work of a plurality of service agents to enable the service agents to process the task in parallel; the collaboration layer is used for providing information sharing and collaborative decision for a plurality of professional agents; and the resource layer is used for providing shared knowledge and tool support for a plurality of professional agents. Through one embodiment of the invention, the problem of low efficiency of the business process in the related technology is at least solved, and the effect of improving the efficiency of the business process in the related technology is further realized.

The invention belongs to the field of man-machine interaction, and particularly discloses an intelligent AI dialogue interaction method and system based on an RICH normal form, and the method comprises the steps: obtaining current dialogue context data and a user input signal, processing the data through an intention recognition model, and obtaining a user intention vector and a scene label; according to the user intention vector and the scene label, acquiring hierarchical state information in a preset multilayer interaction framework, and if the matching degree of the vector and the current hierarchical state is lower than a preset threshold value, activating an information sharing mechanism to obtain an updated shared knowledge graph; through the shared knowledge graph, a dynamic scene adaptation module is fused, a potential switching path and intention extension of a conversation are determined, and an optimized interaction path sequence is obtained; the objective of the invention is to solve the problems of inaccurate intention recognition, stiff interaction path switching and incoherent response logic of a traditional dialogue system in a complex scene in the prior art.

The invention relates to the technical field of electric vehicle charging infrastructure intelligent management, and discloses a charging pile load balancing method and system based on a neural network, and the method comprises the steps: analyzing the state, environment and user behavior data of a charging pile through a multi-mode auto-encoder network; generating a unified scene representation; recognizing a typical scene by using a variational auto-encoder and a density clustering algorithm; decomposing the charging load balancing knowledge into cross-scene shared knowledge and scene specific knowledge; automatically generating a neural network architecture adaptive to the current scene based on the scene similarity; using a meta reinforcement learning algorithm to train a meta strategy network to generate a load balancing decision; scene smooth transition is realized through gradual model switching; according to the method, multi-scene adaptation, knowledge sharing and smooth switching are realized, the problems of system redundancy, knowledge isolation, scene switching performance fluctuation and the like in a traditional method are effectively solved, and the operation efficiency of the charging infrastructure is improved.

The present disclosure relates to utilizing a dynamic knowledge graph enrichment system to dynamically and automatically maintain knowledge graphs shared between groups of user identifiers with up-to-date findings and discoveries. In particular, the dynamic knowledge graph enrichment system changes static shared knowledge graphs into dynamically evolving ones utilizing statistical guarantees that automatically incorporate new edge connections into a shared knowledge graph after verifying the reliability and veracity of the proposed edge connections being offered. Further, the dynamic knowledge graph enrichment system facilitates forming new connections between different shared knowledge graphs that previously went undetected by flexibly facilitating exploration over multiple knowledge graphs and providing synergistic knowledge graph updates.

We provide the system. [Solution] A means of acquiring knowledge from elderly people as voice or text data using speech recognition technology, A method for analyzing the acquired data using natural language processing algorithms and classifying it into general themes, A means of storing classified data on a digital information storage medium and providing it to users as needed, A means of providing an interface for sharing knowledge about the elderly among caregivers, A system that includes this.

The invention relates to the technical field of intelligent questions and answers, in particular to a medicinal and edible component query method and system based on a knowledge graph enhanced large model, and the method comprises the steps: receiving multi-modal heterogeneous data input by a user through the large model, and carrying out unified standardization processing on the heterogeneous data to obtain standard input; mapping a key entity in the standard input to a pre-constructed medical health knowledge graph node, and generating a plurality of reasoning paths; performing information fusion, conflict resolution and semantic sorting on the plurality of reasoning paths to generate a comprehensive report; a reinforcement learning strategy is introduced, and the answer content and the recommendation strategy are optimized according to feedback or questions of the user; knowledge sharing is carried out in a cross-mechanism mode, medical health knowledge maps from different sources are communicated and combined, and the medical health knowledge maps are continuously updated. According to the invention, accurate, efficient, personalized and safe food and medicine component query service can be provided for the user.

The invention belongs to the technical field of text recognition, and provides a multi-granularity multi-view multi-task incremental multi-language text recognition method and system, and the method comprises the steps: extracting visual features and semantic features of a target image through a pre-trained feature recognizer; carrying out shared knowledge discovery of two granularities of characters and words, determining a shared instance, and defining a corresponding language tag to which the shared instance belongs; a language adapter is utilized to predict language scores of characters and words by modeling visual features and semantic features, dynamic weighting is performed on visual dependence and semantic dependence by utilizing a gating mechanism, a global language score fusing visual and semantic information is obtained, cooperation of feature recognition is guided, and recognition efficiency is improved. Sequence decoding of character level and word level is completed, and multi-language text recognition is achieved. According to the invention, the forgetting of old language knowledge in the incremental learning process can be reduced, so that the continuous learning ability of the text recognizer is improved.

The invention discloses an agent interaction memory cooperation method and device based on memory grading and a medium, and the method comprises the steps: obtaining the original interaction data of a user, and carrying out the context interaction analysis of the original interaction data, so as to obtain an interaction state variable; based on the interaction state variable, determining user interaction multiplexing state data through cross-session multiplexing of the user unique identifier; obtaining an iteratable shared knowledge base through high-frequency multiplexing knowledge precipitation according to the user interaction multiplexing state data; performing data interaction control on the interaction state variable, the user interaction multiplexing state data and the iteratable shared knowledge base to determine a directional calling state of the memory data; and based on the directional calling state of the memory data, through multi-level cache matching of agent demand analysis, obtaining agent interactive memory collaborative storage data. Through the method, the technical problem that an artificial intelligence interaction memory function cannot meet real-time interaction dynamic and personalized requirements in the prior art is solved.

The application provides a large model-oriented distributed computing power scheduling method and related device, the method comprises: obtaining an agent task request and task data, the request comprising computing power demand, data sensitivity level and memory calling demand; based on the data sensitivity level and the computing power demand, determining the computing power source as edge computing power or public cloud computing power; if it is edge computing power, issuing a task scheduling instruction to the edge computing power node to control its exclusive allocation and release the computing resources after completion; if it is public cloud computing power, routing the task to a public cloud node; based on the memory calling demand, controlling the execution node to write the task execution memory with a timestamp into a shared knowledge base to trigger the knowledge base to perform time sequence sorting and gradient degradation processing. The application realizes the shunting of heterogeneous computing power, safeguards data privacy while preventing resource preemption of edge multi-task concurrency, and realizes the sharing and storage release of context features through the memory degradation mechanism.

This application provides a self-regulating dialogic multi-agent system and interaction method based on dynamic topology. The system includes: a scheduling module, used to receive user input and route tasks to corresponding domain task agents under normal conditions; several domain task agents, used to process the received tasks and evaluate the cognitive state in real time, and issue assistance signals when the cognitive state meets preset conditions; a shared knowledge management module, used to maintain a dynamically updated distributed semantic knowledge graph; wherein, the distributed semantic knowledge graph is used to share, associate and persist the structured knowledge entities and context information generated during the conversation in real time among all modules; and a topology control module, used to dynamically switch the communication control topology of the system from a first topology to a second topology in response to the received assistance signal; and to restore the communication control topology from the second topology to the first topology when the assistance signal is eliminated.

The invention relates to the technical field of blade rolling, and provides a rolling fault prediction method and device, a storage medium and computer equipment, and the method comprises the steps: receiving and fusing multi-dimensional data from a plurality of sensors of a rolling production line, and obtaining to-be-predicted data; inputting the to-be-predicted data into a pre-trained prediction model, and obtaining rolling fault prediction information based on the to-be-predicted data and the prediction model; the prediction model is a model constructed and updated based on an improved efficient lifelong learning algorithm and comprises a shared knowledge matrix and task specific parameters, the task specific parameters are associated with task identifiers determined by rolling static feature data, and the task identifiers are used for identifying different rolling condition tasks. According to the embodiment of the invention, the multi-dimensional data is fused, and the prediction model based on the improved lifelong learning algorithm is used for predicting the rolling fault, so that the adaptability of the model is improved, the prediction capability can be dynamically updated and optimized, and a more accurate prediction result is provided under different working conditions.

This invention discloses a satellite-based individual identification method under varying modulation conditions based on multi-task decoupling learning, primarily addressing the poor performance of existing technologies in extracting individual features under varying modulation conditions. The implementation scheme involves: receiving downlink communication link signals passing through an observation orbit and dividing them into training and test sets; establishing two knowledge extraction blocks and one shared knowledge extraction block, each composed of a complex-valued multi-scale embedding unit and an attention-gated unit, and connecting these three knowledge extraction blocks in parallel to form a feature extraction module; stacking the feature extraction modules and connecting them to a classifier to form a multi-task decoupling network; inputting the training set data into the multi-task decoupling network for training; inputting the test set data into the trained multi-task decoupling network, and outputting the modulation type and satellite identity results. This invention mitigates feature loss caused by modulation variations, can extract rich individual information under varying modulation modes, enhances the recognition performance of complex signals, and can be used for electronic reconnaissance and identification of source satellites.

The invention relates to an image classification method based on multi-source federal prompt tuning, which is applied to a multi-client collaborative learning scene, each client holds a local data set and cannot be directly accessed by a server or other clients, and in the tuning process, the local data set is not directly accessed by the server or other clients. Each client relieves the distribution difference between the clients through a local feature extraction and sharing mechanism, meanwhile, training of a visual basic model comprising a prompt module is carried out through federated learning, the clients and a server, and the prompt module is composed of global prompt and local prompt and captures cross-domain shared knowledge and individual specific features; and after tuning is completed, an image classification task is completed by using the visual basic model. Compared with the prior art, the method has more excellent classification accuracy and robustness, and is suitable for federal image classification learning tasks of efficient communication and privacy protection.

The application belongs to the technical field of wireless communication, and discloses a phase shift feedback method of a reconfigurable intelligent surface. The method is applied to a sending device, and the phase shift feedback method of the reconfigurable intelligent surface comprises the following steps: obtaining a to-be-transmitted phase shift matrix; determining a feature serial number according to the to-be-transmitted phase shift matrix, a target encoder and a shared knowledge base, wherein the shared knowledge base is common to the sending device and a reconfigurable intelligent surface device; and sending a bit stream of the feature serial number to the reconfigurable intelligent surface device, so that the reconfigurable intelligent surface device determines a received phase shift matrix according to the bit stream of the feature serial number, the shared knowledge base and a target decoder. In the foregoing manner, the feature serial number of the to-be-transmitted phase shift matrix that needs to be fed back is determined, and only the bit stream of the feature serial number is transmitted during transmission, so that the communication resources occupied by the feedback of the reconfigurable intelligent surface and the feedback data volume are greatly reduced, the time resources consumed during the feedback are reduced, and the phase accuracy of the feedback is ensured.

To provide a system for realizing secure and safe mountain climbing by predicting a risk in mountain climbing and proposing a safety measure.SOLUTION: The system includes a data collection section, a big data generation section, a risk prediction section, a safety measure proposal section, a real-time situation grasping section, a warning section, a knowledge sharing section, a community formation section, and a mountain climbing plan optimization section. The data collection unit collects past mountain climbing data. The big data generation unit accumulates data as big data. The risk prediction unit predicts a risk. The safety measure proposal unit proposes a safety measure. The real-time situation grasping unit collects and analyzes real-time data during the mountain climbing. The warning unit issues a warning based on the analyzed data. The knowledge sharing unit shares knowledge based on the collected data and the analysis result. The community forming unit manages the formed community. The mountain climbing plan optimization section optimizes the mountain climbing plan.SELECTED DRAWING: Figure 1

The invention discloses a multi-agent collaborative large model data distillation method and system under an edge cloud architecture. According to the method, firstly, training and inference are executed locally through an edge student Agent, and local knowledge information is extracted to obtain generator model parameters; secondly, the edge student Agent uploads generator model parameters to the cloud teacher Agent; then the cloud teacher Agent aggregates knowledge of each edge student Agent and updates teacher model parameters; then the cloud teacher Agent issues global knowledge to the edge student Agents to help distillation of cloud large model knowledge; and finally, the performance of each edge student Agent is improved through collaborative reasoning and collaborative strategies. According to the method, a teacher-student relationship network of a cloud large model and an edge small model is established, and knowledge is shared through a soft label and a feature alignment distillation mechanism; by using uncertainty weighting and distillation temperature self-adaption methods, the model precision and the training stability are ensured.

The invention discloses an obstacle segmentation method based on a shared knowledge strategy particle swarm two-dimensional Ostu algorithm, which solves the technical problems of poor obstacle segmentation effect and low efficiency in a planetary scene, and obtains an obstacle segmentation result by preprocessing an obtained image, calculating a two-dimensional gray histogram of the preprocessed image and initializing particle swarm parameters. Calculating an inter-class discrete matrix of all the particles, calculating fitness function values of all the particles according to a set fitness function, updating particle local extremum and particle swarm global extremum according to the function values, updating particle speed and position according to the extremum, and combining with a shared knowledge strategy to obtain a shared knowledge strategy; and continuously iteratively searching an optimal solution, and outputting the optimal solution as an optimal segmentation threshold value to carry out obstacle segmentation, carrying out morphological processing optimization on the segmented image, and outputting a final obstacle segmentation map.

The invention provides a multi-model fusion question and answer interaction method and system and related equipment. The method comprises the steps that a data set containing question and answer field related knowledge is constructed; respectively inputting the data set into an online general large model and at least one online subdivision domain model for fine tuning training; carrying out knowledge base sharing on the obtained general large model on the fine tuning line and the fine tuning field subdivision model on the fine tuning line; obtaining questions and answers input by the user through the local model; respectively inputting the question-answering question into the fine-tuning on-line general large model and the fine-tuning on-line subdivision domain model, and obtaining a first question-answering result and a second question-answering result by calling a shared knowledge base; and analyzing and summarizing the first question-answering result and the second question-answering result through a local model, and outputting a target result of the question-answering question. According to the technical scheme, the professional ability of the local model is improved by integrating the professional ability of the online model, the hardware requirement of the local model is reduced, and then the customization cost of the local model is reduced.

The application discloses a human-machine collaborative task shared knowledge generation method and device and a medium. The method comprises the following steps: human-machine knowledge difference analysis; human-machine knowledge sharing criterion establishment; shared knowledge graph construction; and shared knowledge generation. The shared knowledge generation method can generate the knowledge required for each Agent in a multi-Agent team to make corresponding decisions, can effectively improve information transparency, can make the method have explainability, can effectively improve the trust degree of a human, can reduce the operation load of the human, and can efficiently and reasonably complete the human-machine interaction and collaboration process.

The invention belongs to the technical field of artificial intelligence, and discloses a cross-domain recommendation method and system fusing a large language model and hierarchical comparative learning, and the method comprises the steps: carrying out the domain decoupling of graph volume out-of-domain specificity and shared knowledge, generating a fine-grained semantic vector through the LLM, and designing the hierarchical comparative learning to achieve the alignment of semantic and structural features, and finally, outputting a recommendation result through two-stage adaptive fusion. According to the method, semantic and structural information is effectively fused, accurate migration of cross-domain knowledge is realized, the accuracy of cross-domain recommendation is effectively improved, cross-domain interference is inhibited, and the generalization ability of the model in a data sparse and cold start scene is enhanced.

The application relates to the technical field of model-driven system engineering modeling, in particular to a multi-agent collaborative automatic modeling method and system for MBSE. The method obtains user natural language modeling requirements, retrieves and obtains a knowledge subgraph related to the requirements in a shared knowledge environment, constructs a modeling context, calls a generation agent to generate a current SysML v2 model, calls a verification agent to perform syntax verification and semantic verification based on a domain knowledge graph, and outputs a verification result; when the verification result represents that syntax is not passed or there is semantic inconsistency or violation of engineering constraint rules, a repair agent is called to perform deterministic repair according to the domain knowledge graph and iterative verification until a preset condition is met to output a final SysML v2 model; in the case of coverage gaps, gap marking or bottom repair can be performed. The application improves the correctness, explainability and convergence efficiency of complex system modeling.

A system for crowd-sourced media insights and predictive analysis is disclosed. Electronic devices capture media representing a real-world environment and associate it with user-generated insights. A processing unit converts the media into unique identifiers to store and retrieve these insights from a database, enabling users to access shared knowledge by capturing similar content. Additionally, the system analyzes live-streamed video using machine learning to predict subsequent events and transmit alerts or predictive insights to the user in real-time.