183 results about "Knowledge structure" patented technology

The invention discloses a declarative knowledge construction method and device for semantic and situational knowledge collaborative modeling, computer equipment and a readable storage medium, and relates to the field of data processing.The method comprises the steps that firstly, a multi-modal document is analyzed, a chapter abstract is extracted, and structured content is obtained; entities, events and multi-modal knowledge points are extracted from the structured content, and cross-modal fusion is carried out on the entities, the events and the multi-modal knowledge points; carrying out anaphora resolution based on the fused knowledge points, and constructing a double atlas containing a knowledge atlas, a affair atlas and a four-dimensional relation triple; clustering the double maps to obtain a theme community, and performing association mapping on the community, the triple and the entity event, the chapter abstract and the multi-modal knowledge point to form association knowledge; vectorizing the associated knowledge and establishing a vector knowledge index; and carrying out compression ratio and accuracy evaluation on the knowledge through an evaluation system, and feeding back and optimizing the whole knowledge construction process. According to the method, multi-modal knowledge deep fusion and semantic scene collaborative modeling are realized, and the knowledge structuring degree and the application reliability are improved.

The invention relates to the technical field of building engineering cost intelligent estimation, and discloses a building engineering cost intelligent estimation and checking method fusing a domain knowledge graph and a large language model, and the method comprises the following steps: S1, building a building engineering domain knowledge graph (KG), integrating a quota library, a material price library and historical project data, and carrying out the calculation of a building engineering domain knowledge graph (KG); and defining an entity relationship and a dynamic updating rule. According to the intelligent building engineering cost estimation and checking method fusing the domain knowledge graph and the large language model, a closed loop of knowledge structuring (KG) + semantic comprehension (LLM) + multi-modal reasoning is realized for the first time, engineering cost management is promoted to be transformed from experience driving to data-knowledge dual driving, a reusable technical normal form is provided for the field of intelligent construction, and the engineering cost is estimated and checked. The method solves the problems of efficiency, precision, dynamics and interpretability of traditional cost management, has technical innovation and practical value, and provides core technical support for digital transformation of constructional engineering.

The invention relates to the field of intelligent automatic driving, in particular to a knowledge-driven end-to-end automatic driving method based on a sparse expert mechanism and a diffusion model. Comprising the following steps: S1, sensing information processing and state coding; s2, sparse expert module construction and multi-task training; constructing a sparse expert module composed of a plurality of experts, and obtaining a reusable driving skill through multi-task behavior cloning training; s3, generating a diffusion strategy network and an action sequence; and on the basis of a diffusion model, a future multi-step control action sequence is generated from the current state condition, and a continuous and stable driving decision is formed. And S4, a continuous learning and task migration mechanism. According to the method, a combinable and explainable modular driving knowledge structure is constructed, so that the strategy modeling capability is remarkably improved; a diffusion generation mechanism effectively improves the smoothness and stability of the decision process; and a continuous learning and task migration mechanism of structural decoupling improves the long-term adaptability and deployment efficiency of the system.

The invention discloses a large-model-driven automatic knowledge graph construction method based on a confidence feedback mechanism, and aims to improve the structural accuracy and semantic consistency in a structured triple generation process, and perform structural constraint guidance by using a few-sample prompt mechanism and a cross validation mechanism of a heterogeneous large model. And the control capability of the large language model on the triple format is enhanced, so that format offset and semantic redundancy in the generation process are reduced. And meanwhile, a multi-dimensional confidence evaluation system is constructed, model consensus judgment, semantic rationality analysis and knowledge consistency verification are fused, and refined quantification and screening of triple quality are realized. According to the method, a confidence backtracking feedback strategy is introduced, a generation-verification-optimization closed-loop process is constructed, the expression and correction capability of the system on a complex knowledge structure is enhanced, the dependence on an external API is effectively reduced, the consumption of computing resources is reduced, and the operation efficiency of the system and the feasibility of engineering deployment are remarkably improved.

The invention relates to a new media AI marketing content creation method and device, equipment and a medium. The method comprises the steps of obtaining user demand configuration, and obtaining a creation intention vector through natural language analysis; based on the platform characteristic knowledge base, extracting a corresponding structure specification and a propagation mechanism according to the target platform, and encoding to obtain a platform characteristic vector; obtaining historical content interaction data corresponding to the audience group, and generating a user-content interaction vector by adopting collaborative filtering and a label similarity algorithm in combination with the content keyword; and calling an artificial intelligence large model, and performing content generation according to the creation intention vector, the platform feature vector and the user-content interaction vector to obtain content creation data. By adopting the method, the goal of automatic, high-quality and personalized new media marketing content creation in a multi-platform environment can be realized by means of natural language analysis, knowledge structure extraction, large model generation constraint and the like.

The invention relates to the field of semiconductors, in particular to an answer generation method and device based on a semiconductor knowledge base and a medium. The method comprises the steps of performing data association on structured knowledge according to a semiconductor domain knowledge structure to construct a class of knowledge bases; performing data recombination on the unstructured knowledge according to the layout so as to construct a second-class knowledge base; a query question is received, retrieval is executed in the first-class knowledge base and the second-class knowledge base based on a preset retrieval model, and a plurality of related knowledge fragments are obtained; determining a first fusion weight between the knowledge fragments of the same knowledge type based on the relevancy between each knowledge fragment and the query question and the update time of each knowledge fragment; determining an optimal knowledge type according to the group feature information, and determining a second fusion weight between different knowledge types according to the optimal knowledge type; and based on the first fusion weight and the second fusion weight, obtaining a reference answer according to the plurality of knowledge fragments. And the requirements of the semiconductor field on information comprehensiveness and answer accuracy are considered.

The invention provides a multi-source heterogeneous data dynamic updating and agent knowledge base construction method, and relates to the technical field of data processing, and the method comprises the steps: organizing heterogeneous data through a three-dimensional feature matrix, constructing feature mapping through singular value decomposition and cross decomposition, and executing recursive tensor completion to generate a fusion feature space; extracting multi-scale features and determining a stable knowledge entity based on comprehensive measurement; constructing a network structure and dividing knowledge clusters; and performing differentiation fusion of the knowledge clusters based on the life cycle parameters. According to the method, efficient integration of heterogeneous data, accurate extraction of knowledge entities and dynamic optimization of knowledge structures are realized, and the intelligent level of knowledge management is improved.

The invention provides a water conservancy knowledge structured extraction and verification method and device, and belongs to the technical field of artificial intelligence, and the method comprises the steps: carrying out the differential text processing of different formats of files, and generating an intermediate file; classifying the intermediate file into a regulation class or a non-regulation class based on a preset rule base; performing hierarchical title identification on the regulatory files to form entry knowledge blocks, and converting table contents into HTML (Hypertext Markup Language) knowledge blocks; performing semantic segmentation on the non-regulation file to generate knowledge blocks; performing knowledge block checking and filing, and marking an abnormal alarm block; converting the table knowledge blocks into natural language description by utilizing a large model; and positioning the context of the original text of the alarm knowledge block, and performing intelligent correction through a large model. According to the method, a traditional semantic analysis model and a large language model are creatively fused, a closed-loop process of preprocessing, extraction, verification and correction is formed, the problems of structured analysis and error correction of complex texts in the water conservancy field are solved, and the knowledge processing efficiency and accuracy are remarkably improved.

The invention relates to the technical field of online education, and discloses an online course learning management method based on a knowledge graph. The method comprises the following steps: acquiring multi-modal learning behavior data of a learner, and extracting a deep learning state vector reflecting knowledge understanding depth, learning input degree and cognitive confusion through semantic fusion; and dynamically calculating and updating the logical relationship strength among the knowledge points in the course knowledge graph by using the vector, so that the knowledge structure can adaptively evolve along with the actual cognitive state of the learning group. And generating a real-time personalized learning path based on the updated knowledge graph and the current state vector of the learner. Meanwhile, according to cognitive confusion features in the state vector, intervention measures such as pushing of remedial resources, adjusting of content sequence or starting of self-adaptive testing are triggered in real time. According to the method, the dynamic optimization of the knowledge graph and the accurate and immediate response of learning intervention are realized, and the adaptability and management efficiency of online learning are improved.

The invention provides a knowledge structured extraction method based on a multi-modal large model, and relates to the technical field of data processing. Comprising the steps of determining to-be-extracted attribute information according to a data extraction requirement, and generating a structured data model according to the to-be-extracted attribute information; according to the structured data model and a preset extraction strategy, performing data extraction on a to-be-processed original multi-modal file to obtain a plurality of extraction results; merging the plurality of extraction results to obtain a merged extraction result; and verifying the combined extraction result to obtain a target extraction result. According to the method, in knowledge extraction through a multi-modal large model, through an extraction strategy of combining multi-round extraction with staged extraction, the data volume of one-time extraction of the model can be reduced, data extraction omission is avoided, meanwhile, staged extraction enables the model to be gradually progressive from coarse to fine and from global understanding to field-level fine extraction, context loss is avoided, and the knowledge extraction efficiency is improved. And the precision of the extraction result is improved.

System and method for using knowledge fabric based on knowledge ontology, designed for deep language understanding and mechanistic causal reasoning, and meta-knowledge repository for auditable question answering. The method includes receiving an input text from a user, building a knowledge graph that represents real world facts and associations in the form of contextually tagged and weighted knowledge propositions, in multiple knowledge domains. The knowledge graph in combination with causal path knowledge and metadata describing digital sources containing answers constitutes the knowledge fabric. The method includes resolving ambiguity and determining actual intent of the user for the input text, from a plurality of interpretations of intent for sentences using the knowledge graph in conjunction with logical inference to achieve deep natural language understanding. The method includes finding / delivering response to the input request as to why / how unknown factors resulted in known outcome, or what outcomes are likely given known causal factors.

The invention relates to the technical field of talent management and data analysis, in particular to a multi-source data driven scientific and technological talent ability portrait accurate matching method. The method comprises the following steps: collecting and preprocessing multi-source talent data, and mining text information by applying a natural language processing technology; constructing a talent knowledge structure graph by identifying the relationship between the key entities and the extracted entities; determining a plurality of capability dimensions, and establishing a quantitative model or rule for each dimension to measure the capability; calculating scores of all dimensions by comprehensively considering personal differences of talents and working situation factors, and generating a visual ability portrait; a demand standard is determined by analyzing the situation of a demand side, the matching degree between talents and the demand standard is calculated after the talents are screened from a talent pool, the recommended talents are ranked according to the matching degree, and related parameters and rules are optimized according to feedback of the demand side. The talent ability can be comprehensively and accurately described, efficient and accurate matching is achieved, the method adapts to dynamic changes, and talent configuration in the science and technology service field is effectively optimized.

The invention provides an educational resource sequence recommendation method and a related device, and relates to the technical field of resource recommendation. After student learning behavior data, learning resource attribute data and knowledge graph structure information are obtained, a regularization matrix decomposition technology is adopted to carry out potential factorization modeling on a student-resource interaction scoring matrix, and basic preference embedding features of students and resources are extracted. Aiming at diversity and complexity of student learning behaviors, a multi-view feature coding module is designed, learning sequence and short-term interest features are extracted from a behavior sequence view, static attribute features of students and courses are extracted from an attribute information view, and knowledge point pre-correction dependence and association relationship features between courses are extracted from a knowledge structure view. In order to improve the consistency and discrimination of feature representation, on the basis of multi-view fusion features, a comparative learning mechanism is introduced, and the discrimination and migration modeling ability of the model to student interests and preferences is enhanced by constructing a positive and negative sample comparison loss function.

The invention relates to the technical field of education science and technology recommendation systems oriented to personalized learning path optimization, and particularly discloses an education science and technology recommendation system oriented to personalized learning path optimization. The method aims at solving the problems that an existing recommendation system is difficult to dynamically perceive a cognitive state, knowledge structure semantics and dependency relationships are ignored, and the recommendation precision is low due to data sparsity. The system comprises a multi-source data acquisition and fusion module, a dynamic cognitive state evaluation module, a knowledge graph construction and semantic enhancement module, a path generation and optimization decision module and a self-adaptive execution and feedback adjustment module. Through multi-source data fusion, real-time cognitive state quantification, knowledge graph semantic modeling, multi-target optimization path generation and closed-loop feedback adjustment, accurate recommendation of personalized learning paths is realized, and the continuity, rationality and educational effectiveness of the paths are effectively improved.

The invention relates to the technical field of intelligent education, in particular to an intelligent Agent heuristic question and answer teaching system based on knowledge graph driving, and the system comprises a four-layer progressive architecture including a data layer, a knowledge layer, an Agent layer and an interaction layer: the data layer is responsible for the collection, preprocessing and privacy protection of multi-dimensional teaching data, and provides high-quality input for the construction of a knowledge graph; the knowledge layer is used for converting data into a structured knowledge graph and storing a multidisciplinary knowledge network in an entity-relationship-attribute triple form; in the scheme, a'data-knowledge-Agent-interaction 'four-layer architecture is constructed, knowledge structured association and dynamic updating are realized based on a knowledge graph, weak points of students are accurately positioned by means of sensing, reasoning and decision-making modules of an intelligent Agent, a personalized problem chain is generated, and a multi-modal interaction and privacy protection mechanism is combined, so that the knowledge knowledge association and dynamic updating are realized. Systematization of a knowledge system, personalization of question and answer guidance, precision of data utilization and dynamic knowledge updating are achieved, and the defects of a traditional teaching system are effectively overcome.

The invention provides a cloud-edge collaborative heterogeneous graph neural network vehicle re-identification method, and relates to the technical field of intelligent traffic. The method comprises the following steps: constructing a knowledge structure through a multi-modal heterogeneous graph; extracting a multi-modal semantic relationship through a heterogeneous graph neural network to construct a cloud teacher model; in the method, a dynamic knowledge distillation mechanism driven by feature clustering is introduced to carry out knowledge fine-grained migration to generate a lightweight student model. Experiments are verified on VeRi-776, CityFlow-ReID and a self-built traffic data set, and results show that under the conditions that model parameters are compressed by 63% and a video memory is reduced by 64%, the total precision loss of a student model does not exceed 5%, and the reasoning speed reaches 213FPS. Compared with an existing traditional baseline method, the HGKDF has significant statistical advantages in RMSE and MAE indexes, is superior in training duration and video memory overhead, and is suitable for constructing the deployment of an integrated traffic large model system oriented to city-level intelligent interactive decision and safety monitoring.

The invention relates to the technical field of artificial intelligence and data fusion, in particular to a knowledge graph-based multi-modal data fusion computing system, which comprises a multi-modal data source interface module, a preprocessing module, a fusion alignment module, a knowledge graph module and a reasoning module, different modal features are mapped to a unified semantic space based on a Riemannian manifold theory, cross-modal information interaction is realized through a geometric perception attention mechanism on manifold, and uncertainty is quantified based on manifold entropy. The knowledge graph module provides semantic constraint to guide a fusion process and receives a fusion result to update a knowledge structure, the reasoning module combines fusion features and uncertainty evaluation to perform reliability sorting and interpretation information generation, the system improves the accuracy, reliability and interpretability of multi-modal data fusion, the fusion accuracy is improved by 15%-20%, and the system is suitable for popularization and application. And the cross-modal semantic alignment error rate is reduced by more than 30%.

The invention relates to the technical field of knowledge tracking, and discloses a graph enhanced double-memory collaborative knowledge tracking model based on an ACT-R cognitive architecture. Comprising a static knowledge structure coding module based on hypergraph projection, a batch-level dynamic learning track construction and coding module, a cross-graph gating fusion mechanism, a sequence modeling module and an expert hybrid prediction module. According to the method, long-term stable structured semantic association between concepts in declarative memory is modeled through a static knowledge structure diagram, a dynamic learning trajectory diagram based on batch reconstruction is designed to accurately capture an evolution rule of a behavior sequence in programmed memory, and on the basis, a cross-diagram gating fusion mechanism and a hybrid expert mechanism are introduced, so that the evolution rule of the behavior sequence in the programmed memory is accurately captured. And self-adaptive fusion and multi-path decision of double-graph features are realized.

The invention provides an engineering carbon emission factor dynamic calculation and traceability analysis method and system, and belongs to the field of engineering construction. The method comprises the following steps: acquiring multi-source heterogeneous data of the whole engineering construction process, and preprocessing the multi-source heterogeneous data to form a data set supporting knowledge modeling and computational analysis; constructing a semantic knowledge structure in the field of carbon emission factors in the whole engineering construction process by utilizing a knowledge graph technology; carrying out dynamic calculation on the carbon emission factor and correcting the weight; and carrying out visual analysis on a dynamic calculation result and a correction result to complete dynamic calculation and traceability analysis on the engineering carbon emission factor. According to the method, the defects in a carbon emission accounting system in the existing engineering construction field are overcome.

The invention provides a clinical decision knowledge graph construction method and system, and the method comprises the steps: extracting standardized entities corresponding to diseases, symptoms and diagnosis and treatment elements from medical knowledge data; constructing a clinical concept knowledge graph for representing a medical concept logic relationship and a causal relationship according to the semantic association relationship and the causal dependency relationship among the standardized entities; a diagnosis event, an examination event and a treatment event related to the patient are extracted, link evidences among the events are determined based on the event chain relation among the events, and a clinical event knowledge graph used for representing the disease course evolution process of the patient is constructed according to all the link evidences; and performing knowledge element fusion based on an entity association relationship between the clinical concept knowledge graph and the clinical event knowledge graph, and generating a target knowledge graph for clinical decision analysis. By adopting the scheme of the invention, the cross-map fusion of the static medical concept knowledge and the dynamic disease course event chain relationship can be realized, and the clinical decision knowledge structure with the reasoning ability can be constructed.

The invention discloses an intelligent talent allocation management system and method for chain enterprises, and the method comprises the steps: collecting the flow data and knowledge transmission records of each store, recognizing organization memory nodes through network analysis, and constructing a knowledge fracture risk conduction diagram; frequency features are extracted based on shock wave simulation and Fourier transform, and a cascade influence path of personnel loss is predicted; obtaining a capability-load mismatch coefficient through phase analysis, and generating a deployment opportunity window; a personnel flow potential field is constructed, capability diffusion simulation is carried out, and a multi-dimensional deployment decision space is generated by using tensor operation; carrying out gradient search in the decision space to identify an optimal balance point, generating a ripple allocation path set, and optimizing to obtain an optimal sequence; detecting tissue knowledge density change after deployment is executed, and triggering knowledge structure recombination based on a phase change critical point. According to the method, conversion from passive response to active prediction is realized, the risk of knowledge fracture can be effectively prevented, and talent resource allocation is optimized.

The invention is suitable for the technical field of education, and provides a content recommendation method and system based on children cognition, and the method comprises the steps: obtaining historical learning data, carrying out the construction of a knowledge graph based on the historical learning data, and generating a knowledge structure graph; performing vector mapping processing on the plurality of knowledge nodes and the edges connecting the plurality of knowledge nodes by using a graph embedding algorithm to generate a knowledge vector set; according to the knowledge vector set, calculating a semantic association degree among the plurality of knowledge nodes, and if the semantic association degree is greater than a preset threshold value, extracting a key knowledge node from the plurality of knowledge nodes by using a hierarchical clustering algorithm; determining a learning sequence corresponding to the key knowledge nodes according to a preset knowledge learning strategy, and generating a knowledge progressive sequence; generating a target knowledge node sequence based on the knowledge progressive sequence and the historical learning data by using a preset reinforcement learning algorithm; and the target recommendation content is generated according to the target knowledge node sequence and recommended to the user, so that more appropriate learning content can be recommended.

The invention provides a homework correction mark leaving method based on AI. The method is applied to the technical field of teaching and artificial intelligence, and comprises the following steps: S1, obtaining an original input image of a paper job; s2, adopting an OCR model based on a cross attention mechanism to perform accurate text recognition and structured cleaning and preprocessing of text data on the writing content in the original input image to obtain high-dimensional text features; s3, constructing an answer judgment and error recognition model, introducing an improved discrete laying hen algorithm to perform feature selection and model training, and outputting an error classification result; s4, generating a simulation correction trace and a traceable scoring log according to the error classification result and the original homework image; and S5, mapping the error type with a standard knowledge structure in the knowledge point library to generate a customized lecture. According to the invention, the accuracy, interpretability and teaching suitability of homework correction are obviously improved.

The invention relates to the technical field of intelligent education, in particular to a personalized learning path generation method and system driven by a dynamic knowledge graph. The method comprises the steps of collecting and processing multi-modal learning behavior data of a learner to construct and update a dynamic knowledge graph reflecting a knowledge mastering state and knowledge point association in real time; a double-engine diagnosis mechanism combining large model deep reasoning and knowledge graph real-time verification is adopted, and cognitive weak points and knowledge structure defects of learners are accurately recognized; and on the basis of a diagnosis result, a personalized learning path adaptively matched with the cognitive state of the learner is generated through multi-agent collaborative decision, and closed-loop optimization is performed on a knowledge graph and a path planning strategy according to real-time feedback of a path execution effect. According to the invention, the defects of the traditional adaptive learning system in the aspects of diagnosis accuracy, individuation degree and dynamic adaptability are effectively overcome, and accurate, efficient and continuously optimized individualized learning experience can be provided for students. According to the method, through deep fusion of double-engine diagnosis and the dynamic knowledge graph, the accuracy and reliability of cognitive state diagnosis are remarkably improved, and the illusion problem of a large model in the STEM field is solved; through a multi-agent collaborative decision-making mechanism, high personalization and dynamic adaptability of a learning path are realized; finally, a teaching closed loop with a self-optimization capability is formed, and the intelligent level and the teaching efficiency of the self-adaptive learning system are essentially improved.

Provided is a method and system (108) for building and leveraging a knowledge fabric (110) in a Software Development Lifecycle (SDLC). A plurality of SDLC artifacts are received from a plurality of heterogeneous data sources (102). The plurality of SDLC artifacts are then correlated to build an end-to-end correlation and are clustered to generate an SDLC knowledge fabric (110). This includes extracting semantic and contextual data from the plurality of SDLC artifacts using Natural Language Processing (NLP) and deep text analytics and transforming the extracted semantic and contextual data to knowledge graphs. One or more actionable items (112) are then derived using the SDLC knowledge fabric (110) and the one or more actionable items (112) are used to improve overall process efficiency and accelerate software delivery in the SDLC.

The invention relates to the technical field of deep learning, in particular to an intelligent prospecting model construction method based on a knowledge graph and data deep learning, which comprises the following steps: collecting multi-source geological data to perform time decomposition and spatial stratified sampling to extract features, calculating an attachment weight through semantic coding to construct a geological knowledge structure, and constructing a geological prospecting model; the method comprises the steps of extracting spatial-temporal features in a convolution mode, combining with semantic deviation degree weighted fusion to generate a hierarchical feature coupling result, carrying out aggregation analysis on spatial-temporal correlation to extract consistent components, judging a metallogenic response relation through conditional probability reasoning, fitting a model, calculating deviation, adjusting weights, analyzing semantic consistency, and carrying out classification and aggregation to generate an intelligent prospecting optimization result. The feature precision is improved through time decomposition and spatial stratified sampling of multi-source geological data, data association is enhanced through semantic coding, time-space consistency is enhanced through convolution extraction and semantic fusion, multi-scale features are balanced through hierarchical coupling, the ore-forming relation judgment accuracy is improved through probabilistic reasoning, and the result precision and stability are remarkably improved.

One or more computing devices, systems, and / or methods for providing adaptive explainability for machine learning models are provided. A knowledge structure, representing entities with nodes and relationships between entities as edges between the nodes, is processed to create knowledge system entity embeddings. A dimensionality of the knowledge system entity embeddings is reduced to create dimensional embeddings. The dimensional embeddings and relationships are processed using an optimal transport plan to generate feedback. The feedback is used to modify the knowledge structure for generating adaptive explainability information that explains predictions generated by the machine learning models.

The invention discloses a document processing method and a related device, and relates to the technical field of data processing. Comprising the following steps: analyzing an original document to obtain a text sequence corresponding to the original document; processing the text sequence through a large language model to generate a directory list corresponding to the text sequence; segmenting the text sequence to obtain a plurality of text blocks; based on the plurality of text blocks, a preset fusion rule and a large language model, determining a plurality of target semantic text segments and titles, starting point positions and ending point positions of the target semantic text segments; constructing a hierarchical knowledge structure based on the target list, the plurality of target semantic text segments, and titles, starting point positions and ending point positions of the target semantic text segments; the hierarchical knowledge structure has the characteristics of structuralization, traceability and semantic coherence, and the knowledge retrieval accuracy and efficiency can be improved by storing the hierarchical knowledge structure into a knowledge base.

A hierarchical personalized federal learning method based on cross-client prototype clustering relates to the field of artificial intelligence and distributed machine learning, and comprises the following steps: a client maps a local sample to a semantic prototype space for information compression, and generates a category prototype matrix; the server receives the prototype matrix uploaded by each client, dynamically clusters and aggregates the prototype matrix, constructs a multi-level knowledge structure including a cluster-level expert model and a global semantic prototype, and maintains the global semantic prototype through index moving average; the client receives the cluster-level expert model, the global model parameters and the global semantic prototype from the server, and carries out a new round of local training on the basis; the client integrates local data and multi-level knowledge issued by the server to optimize three types of loss updating models; through loop iteration, the local model of the client gradually realizes personalized adaptation and global alignment. According to the method, the communication efficiency, the privacy protection safety, the model convergence speed and the generalization ability are improved.

The invention belongs to the technical field of emotion recognition, and particularly discloses a big language model enhanced dialogue causal emotion implication method and system, and the method comprises the steps: extracting common knowledge corresponding to the identity of a speaker of each dialogue, and carrying out the coding and aggregation of the speech-level feature representation and the common knowledge through a graph structure, generating a first utterance feature representation with enhanced common knowledge; inputting the utterance-level feature representation into a multi-head self-attention module with relative position perception, and generating a second utterance feature representation with an enhanced dialogue structure; and splicing the utterance-level feature representation and the knowledge structure interaction feature to obtain a final utterance complementarity feature representation, and inputting the utterance complementarity feature representation into a multi-layer perceptron to obtain a prediction result of the emotion causal relationship. According to the method and the device, the emotion causal relationship prediction performance of the model in a complex dialogue scene can be improved, and the emotion causal relationship prediction accuracy is improved.