87 results about "Knowledge element" patented technology

The invention discloses a document analysis and query method based on a knowledge graph, and the method comprises the steps: carrying out the part-of-speech tagging of a received to-be-analyzed document, and obtaining a part-of-speech tagging result; extracting knowledge element information from the part-of-speech tagging result based on a preset power grid domain ontology knowledge base, and constructing an initial knowledge graph based on the knowledge element information; combining nodes in the initial knowledge graph to obtain a fused knowledge graph; constructing a mapping table according to the fused knowledge graph, and generating a candidate query template set based on the mapping table; receiving a natural language query statement input by a user, selecting a target query template from the candidate query template set based on the natural language query statement, and generating a target query statement; querying from the fused knowledge graph by using the target query statement, and outputting a query result; according to the method, the accuracy and comprehensiveness of information analysis can be effectively improved, the query intention of the user can be accurately understood, and the accurate query and analysis requirements of professionals on project documents are met.

The invention relates to the technical field of artificial intelligence, in particular to a self-organizing knowledge base construction method, system and device based on a multi-agent system and a storage medium, and the method comprises the steps: S1, monitoring a dialogue information flow, calculating the self-reply confidence coefficient of an agent, and if the self-reply confidence coefficient exceeds a preset dynamic threshold value, judging that a potential knowledge value exists and triggering an extraction request; s2, in response to the request, performing semantic coding on a dialogue fragment, extracting knowledge elements, and generating a structured knowledge unit by utilizing graph neural network modeling and relation calibration; s3, through a pre-training semantic coding model, mapping the multi-dimensional semantic coding model into a multi-dimensional semantic vector; s4, carrying out hierarchical clustering comparison and combination, and if a threshold value is exceeded, establishing a new classification and adjusting a knowledge base index; and S5, knowledge fingerprints are generated and compared, and fusion updating is carried out based on the reputation scoring model when semantics conflict or redundancy occurs. According to the method, unstructured dialogue high-precision knowledge extraction is realized, so that the knowledge base structure is dynamically self-organized according to semantics, multi-source knowledge is coordinated, and the knowledge management automation level and quality are improved.

The invention provides an intelligent scheduling method and system applied to cable operation and maintenance, and the method comprises the steps: firstly deconstructing migration knowledge elements in a cable operation and maintenance migration knowledge base, dividing resource adaptive, process adaptive and risk adaptive knowledge elements, and generating a knowledge element classification set; constructing an operation and maintenance scene dynamic portrait of the to-be-scheduled cable, and integrating the operation characteristics, the operation and maintenance task requirements and the historical operation and maintenance performance data to generate a three-dimensional scene portrait; calculating an adaptation coefficient of each migration knowledge element and the portrait, and screening out a target knowledge element combination; constructing an initial operation and maintenance scheduling scheme framework based on the target knowledge element combination, and integrating various adaptation rules; and updating and adjusting a scheme framework through multi-turn portrait iteration, optimizing resource configuration, a step sequence and an abnormity coping plan, obtaining a final scheme, generating an operation and maintenance scheduling instruction, and sending the operation and maintenance scheduling instruction to an operation and maintenance execution terminal. According to the invention, intelligentization, precision and high efficiency of cable operation and maintenance scheduling are realized.

The invention provides a knowledge graph-driven medical data agent collaborative governance method, system and device and a medium, and belongs to the technical field of medical data governance, and the method comprises the following steps: constructing a knowledge graph containing a quintuple, and defining mapping relationships among medical data entities, attributes and standards; extracting multi-modal medical data knowledge elements through a structured / unstructured engine, and converting the multi-modal medical data knowledge elements into a quintuple for storage; executing table / field / value domain mapping based on the knowledge graph to generate a data mapping result; binding the quality control rules and the medical specifications to form a quality control quintuple; arranging the task into a directed acyclic execution sequence through an agent group cooperation task; and dynamically updating the knowledge graph and performing version management. Medical data full-link governance, knowledge graph unification heterogeneous data semantics, intelligent agent cooperation automatic processing complex tasks, update guarantee knowledge timeliness and quality control rule binding guarantee data quality are achieved, and medical data governance efficiency and accuracy are improved.

The invention belongs to the technical field of electric power intelligent operation and maintenance, and relates to an electric power equipment fault prediction knowledge graph updating method which comprises the following steps: continuously receiving multivariate heterogeneous information from an electric power system operation environment, and applying at least one technology of natural language processing, time sequence analysis and statistical pattern recognition. Extracting new knowledge elements to form a to-be-verified new knowledge set, performing semantic consistency verification on the to-be-verified new knowledge set and an existing knowledge graph, detecting logic conflicts between the to-be-verified new knowledge set and existing knowledge triples and rules through logical reasoning, and obtaining a to-be-verified new knowledge graph based on a context environment where the logic conflicts occur. A conflict resolution strategy is automatically selected and executed by evaluating evidence reliability of new and old knowledge, and an optimized knowledge rule is generated, so that an existing knowledge graph is dynamically updated, the problems that an updating mechanism of the existing knowledge graph is lagged and evolution of context-related knowledge is difficult to process are effectively solved, and the reliability of the knowledge graph is improved. And the knowledge graph is pushed to be converted from static storage to dynamic evolution intelligent agents.

The invention relates to the technical field of textbook construction and optimization, in particular to an automatic textbook construction and optimization method and device.The method comprises the following steps that original resources of multiple types of textbooks are collected, and core knowledge elements in the original resources are recognized; constructing a structured mapping relationship between the knowledge elements and the teaching resources based on the core knowledge elements; on the basis of the structured mapping relationship, constructing a knowledge graph covering multidisciplinary knowledge points and a preposition dependency relationship, and automatically identifying knowledge faults and complementing missing association so as to form a knowledge network with self-evolution ability; based on the knowledge network, utilizing a hybrid generation model to cooperatively work, and quickly generating a structured textbook framework containing knowledge veins, teaching cases and exercise resources; learning condition data including classroom behaviors, learning feedback and knowledge mastering conditions are obtained, and a quantity and chemical condition report is obtained after data processing; and updating data based on the quantitative chemical situation report and the knowledge graph.

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 provides an automatic construction and semantic understanding enhancement method and device for a large model knowledge base. Internet multi-type data sources are integrated through multi-source data collection, and text data related to the large model application field is obtained; a knowledge extraction and fusion technology is utilized to extract key knowledge elements from the collected data and eliminate redundant contradictory information to form a unified knowledge representation system; based on the unified knowledge representation system, automatically constructing a knowledge base with rich semantic association according to a predefined structure mode, and introducing a data quality evaluation mechanism to guarantee the reliability of the knowledge base; and based on the constructed knowledge base, the deep semantic analysis technology is used to enhance the semantic understanding ability of the large model, including semantic coding, knowledge graph embedding, semantic reasoning and knowledge completion. According to the scheme, efficient, accurate and automatic construction of the knowledge base can be achieved, the semantic understanding ability of a large model is remarkably improved, the large model can understand and apply knowledge more accurately, and powerful support is provided for artificial intelligence application.

The invention relates to the technical field of data generation, and discloses a training content generation method and system based on a deep learning model and a storage medium, and the method comprises the steps: obtaining basic content data, and extracting text, table and format content to generate structured data; identifying a content structure and extracting training knowledge elements through deep learning model semantic analysis, and constructing a knowledge graph in combination with an association relationship between the elements; calculating the occurrence frequency, the content chapter and the core degree of the training knowledge elements to obtain a difficulty coefficient; obtaining training parameters, calculating a knowledge coverage coefficient and a knowledge depth coefficient, and generating initial content in combination with the first text framework and the large language model; and performing full-content verification and correction through the second text framework to obtain training content. According to the method, the training content is automatically generated, the problems of low efficiency, high cost, unstable quality and the like of manual question setting are solved, and the scientificity and pertinence of the training content are improved.

The invention discloses an autonomous evolution type 5G protocol fuzzy test system based on a large language model, and the system comprises a protocol extraction unit which employs the large language model to analyze a 5G protocol specification document, and extracts structured protocol knowledge elements containing a message format, state machine logic and parameter constraints; the vulnerability analysis unit generates a potential vulnerability description list through big language model reasoning based on the knowledge elements; the test strategy generation unit drives a large language model to generate a targeted test case based on protocol knowledge, vulnerability description and historical test knowledge in the dynamic test knowledge base; the test execution unit executes the use case and captures a result; and feeding back an analysis result of the internalization unit, extracting a causal rule, generating new test knowledge and feeding back the new test knowledge to the knowledge base. According to the system disclosed by the invention, test targets and resources can be autonomously scheduled according to continuous updating of the knowledge base, so that the test process has an iterative optimization capability, and full-process automation and intelligentization of 5G protocol fuzzy testing are realized.

The invention relates to the technical field of knowledge graph construction, and provides a knowledge graph construction method and system based on a traditional Chinese medicine classical prescription, and the method comprises the steps: obtaining multi-source heterogeneous data composed of a digital text of a classical ancient book, a teaching material, a traditional Chinese medicine authority dictionary, an academic journal literature and structured clinical data; the standard text data is subjected to mixed strategy knowledge extraction, discrete knowledge elements such as entities, relations and attributes are separated out, and the discrete knowledge elements form an original fact unit of a knowledge graph; the extracted discrete knowledge elements are subjected to ontology modeling and formalized representation, and a multi-dimensional semantic network with the prescription-syndrome-disease-machine as the core is constructed; and performing semantic similarity comprehensive judgment by calculating the character string similarity of the entity names and combining context features of the entity names. The system comprises a heterogeneous data processing module, an entity recognition module, a semantic network construction module and a similarity judgment module. Structured organization, deep association and intelligent utilization of classical prescription knowledge are realized.

The invention discloses an academic publication knowledge service intelligent management system and method based on digital twinning, and relates to the field of knowledge service and technology, and the system comprises a data storage module, a dimension weight quantification module, a weight distribution difference analysis module, an effective trigger distribution model construction module and a response priority analysis module. According to the method, the initial weight of each dimension is quantified through six dimensions including multi-dimensional intelligent retrieval, covering knowledge elements, semantic understanding and user portraits, the weight can be dynamically allocated according to user retrieval content features such as whether keywords contain limits or not, new and old attributes of the users and keyword expansion capacity, the weight is updated based on subjective operation contents of the users, and the user experience is improved. The retrieval strategy is deeply matched with the individual requirements of the user, the problem of low retrieval efficiency is greatly reduced, and the operation cost of repeatedly modifying retrieval conditions by the user is reduced.

The invention discloses an intelligent decision-making method based on a road material design knowledge graph system, relates to the field of intelligent design of traffic infrastructure materials, and solves the technical problems that design and optimization of the traffic infrastructure materials depend on artificial experience and a large number of tests for a long time, and the design effect is poor. The method comprises the following steps: extracting knowledge elements from material related data and carrying out standardization processing; constructing a knowledge graph by taking the knowledge elements as nodes; embedding the trained prediction model in the knowledge graph as an executable model node; constructing a combined reasoning mechanism in the knowledge graph; constructing a multi-target optimization model according to a material design target set by a user, and solving according to a performance prediction result generated by a plurality of executable model nodes in the knowledge graph; outputting an interpretable material design decision result based on the solving result; according to the method provided by the invention, the number of experiments and the design cost are greatly reduced, and the efficiency, accuracy and reliability of material design are improved.

The invention discloses a cultural transmission knowledge graph construction method based on game video content analysis, and the method combines deep learning, natural language processing and multi-modal data fusion technologies, and is dedicated to deep mining and structured representation of cultural knowledge elements and transmission association thereof from mass and multi-modal game video data. And constructing a knowledge graph serving cultural transmission research and practice. The method not only provides an important theoretical tool for understanding a new mechanism of cultural transmission in the digital era, but also opens up an innovative technical path for actively, accurately and efficiently bearing and transmitting the cultural value by using game media, and has remarkable theoretical value and application potential.

The invention relates to the technical field of data processing, and particularly provides a multi-source heterogeneous operation and maintenance data analysis method and system based on an industry subject portrait, and the method comprises the steps: collecting industry subject data from a multi-heterogeneous data source, and carrying out the cleaning and standardization processing to form standard data; based on this, constructing an industry knowledge graph comprising industry subjects, knowledge elements and association relationships thereof; meanwhile, in combination with the standard data and the knowledge graph, the quantitative score of the industry subject on the multi-dimensional portrait is calculated and mapped into a portrait label; and realizing anomaly detection, root cause positioning, prediction and early warning or multi-dimensional dynamic analysis by utilizing the knowledge graph and the main body portrait. And finally, generating an industry health state and risk situation conclusion based on an analysis result. According to the invention, the efficiency and reliability of operation, maintenance and management are obviously improved.

The invention discloses a fair competition review case library construction and expansion method based on knowledge modeling and large language model generation, and the method comprises a case library construction stage and an intelligent expansion stage. Collecting data from the heterogeneous data source, analyzing the data into structured case data through an information extraction technology, and finally constructing case knowledge elements for describing case types, subjects and law article application relationships; in the expansion stage, based on existing case knowledge elements, a'case-legal provision 'relation model with variable scene parameters is constructed, diversified candidate case texts are generated by utilizing a large language model, and the candidate case texts are stored after being subjected to validity and logic consistency rechecking, so that efficient and high-quality automatic expansion of a case library is realized.

The invention relates to the technical field of computers, and discloses a table metadata processing method and system, electronic equipment and a storage medium, and the method comprises the following steps: carrying out regional semantic segmentation through weighted fusion based on frame geometric features, character coordinates and language features of a table; constructing a structural feature vector based on the row number, the column number and the row list header cell style of the table, performing geometric semantic integrity reconstruction, executing element standardization and semantic extraction, and performing geometric semantic integrity reconstruction on the basis of a region coding library, a unit conversion chain and a knowledge element rule library by utilizing normalized index names, time and region information. And calculating a regional code, a standard measurement value, a knowledge element ID and a retrieval field, matching the normalized index with an existing index system, and outputting a structured record. According to the method, the requirements of high accuracy and traceable structuralization during data processing in a big data environment can be met.

The application discloses a carbon footprint tracing system fusing a knowledge graph and space-time attention, which comprises a multi-source data fusion module, a power grid, product and auxiliary data are used to obtain space-time characteristics and a knowledge graph original data set; an electric carbon factor dynamic prediction module, preliminary prediction is carried out through a fusion model, and a dynamic error correction model is used to calibrate to obtain a final electric carbon factor; a knowledge graph construction module, based on preset entities and relations and attributes, extracts and fuses knowledge elements from original data, and constructs a special knowledge graph; a carbon footprint accurate accounting module, based on the correlation of the special knowledge graph, automatically matches the final electric carbon factor with the power consumption of each production process of the product, combines non-power carbon emission data, and obtains the total amount of product carbon footprint through an accounting model; and a carbon footprint tracing module, based on the special knowledge graph, traces the production link of the total amount of product carbon footprint. The application improves the product carbon footprint accounting accuracy and full-link tracing capability.

This invention relates to the field of virtual reality simulation technology, and more particularly to a method for generating immersive virtual reality training content. The method includes: a step of constructing a domain knowledge graph: extracting atomic knowledge elements representing operational actions, equipment entities, state parameters, and safety rules from standard documents and case libraries of the target training domain, and establishing multi-dimensional logical relationships between these atomic knowledge elements, including causal, temporal, hierarchical, and probabilistic relationships, to form and store the domain knowledge graph; a step of real-time collection of trainee data; a step of constructing and updating a trainee state model; a step of dynamic content decision-making; and a step of generating and injecting non-linear training content. By collecting trainee data in real time and matching it with the knowledge graph, the system can provide trainees with precise feedback after each training session, helping them identify and overcome their knowledge gaps.

The invention discloses a learning plan analysis system based on beauty teaching, which comprises an education server, a learning analysis module and a learning strategy generation module, and is characterized in that the education server receives a learning progress result of a learner, and the learning analysis module identifies vulnerable knowledge elements by analyzing the result and generates a learning strategy; the learning strategy generation module generates supplementary learning content in a targeted manner; and meanwhile, functional modules such as a learning psychological analysis unit, a learning career analysis unit and an art learning evaluation unit are integrated, and algorithms such as deep learning, natural language processing and reinforcement learning are combined, so that functions such as learning mode recognition, beauty ability multi-dimensional evaluation, personalized learning strategy generation and career path recommendation are realized. According to the system, knowledge weak points and learning requirements of the learner can be accurately positioned, customized learning planning and feedback are provided in combination with beauty teaching characteristics, the learning efficiency and beauty literacy are effectively improved, and support is provided for comprehensive development and occupational planning of the learner.

The application provides a relationship triple construction method, a knowledge enhancement method and related devices, which can fine-tune a pre-training model using knowledge corpus of an application scenario of a knowledge graph to obtain a language model. The language model learns the knowledge of the application scenario and can distinguish whether a target sentence meets the part-of-speech requirement of a relationship triple, whether it is a knowledge element meeting the knowledge of the application scenario of the knowledge graph, and then construct a relationship triple according to the target sentence meeting the above requirements. Manual annotation is not required, which avoids the problem that manual annotation may not cover all possible entities and relationships, resulting in low quality of the knowledge graph.

This application discloses a method, apparatus, device, and storage medium for constructing UAV technical data, relating to the field of computer technology. The method includes: extracting target cognitive behavior data of UAV aviation support personnel during mission execution to construct target triples and establishing a cognitive process model using a target BERT model; clustering knowledge elements related to UAV use and maintenance to obtain a knowledge hierarchy framework; extracting unstructured data from the UAV domain and constructing a target knowledge model using knowledge graph technology; determining the cognitive vectors at each stage of the cognitive process model; determining the matching relationship between the cognitive process and knowledge based on the similarity between the knowledge vectors and cognitive vectors corresponding to the target knowledge model; and constructing UAV technical data by combining the cognitive process model and the target knowledge model. While ensuring the completeness of the UAV technical data construction, the method also considers the cognitive characteristics and knowledge needs of users at different stages.

The application discloses an intelligent medicine operation quality file generation system and method based on big data, and belongs to the technical field of medicine operation quality management information. The application discloses an intelligent medicine operation quality file generation system and method based on big data, and belongs to the technical field of medicine operation quality management information. The compliance big database module adopts a three-level hierarchical rule base architecture, decomposes medicine operation quality file benchmark items into knowledge elements in the form of five tuples of subjects, actions, objects, conditions and evidence, and constructs a knowledge element dependency relationship topology graph based on reference, refinement, supplement and conflict relationship. The dynamic rule engine module loads the knowledge element set with the operation scene as the first dimension and the enterprise operation location as the second dimension, and performs hierarchical conflict detection and priority applicability processing. The application solves the problems in the prior art that a template solidification cannot adapt to a new version benchmark, local benchmark differences lack systematic management, and a flying inspection response efficiency is low.

The invention relates to an operation difficulty dynamic adjustment method and system based on a cognitive flow field, and belongs to the field of teaching assistance. The method comprises the following steps: constructing a three-dimensional cognitive space, and obtaining coordinate positions of knowledge elements in the three-dimensional cognitive space; capturing a time sequence behavior data flow in a user problem solving process and mapping the time sequence behavior data flow to a three-dimensional cognitive space to obtain a cognitive flow trace function; constructing a cognitive flow dynamic model based on the cognitive flow trace function, and outputting cognitive resource density; constructing a multi-dimensional difficulty field based on the cognitive resource density, and outputting a difficulty potential energy field; extracting problem solving track topological features based on the difficulty potential energy field to obtain topological feature values; difficulty field reconstruction is carried out based on the cognitive resource density and the topology characteristic value, and a reconstruction difficulty potential energy field is obtained; and dynamically adjusting the operation difficulty based on the reconstruction difficulty potential energy field. According to the method, modeling and difficulty prospective adjustment of a learning process are realized by constructing a three-dimensional cognitive space and establishing a cognitive flow dynamic model and a difficulty potential energy field.

Typically, in order to preserve their specializations, AI agents operate with isolated knowledge bases to learn skills independently from other AI agents. This results in a waste of computational resources and reduces the adaptability of the AI agents and overall system. Accordingly, disclosed embodiments enable AI agents to learn from each other, while retaining their specialized capabilities. A knowledge manager may identify knowledge elements, within a knowledge repository, that are compatible with a receiving AI agent and do not diminish the specialization of the receiving AI agent. An incremental transfer process may be executed, with performance monitoring and rollback capabilities, to incorporate the knowledge elements into the receiving AI agent, while ensuring that the knowledge transfer does not diminish the specialization of the receiving AI agent. A cross-agent collaboration interface is also disclosed for knowledge sharing and collaboration between AI agents.

The invention provides an aircraft system fault diagnosis method based on knowledge expression. The method comprises the following steps of 1, collecting fault description files; step 2, constructing a fault relation element category library; step 3, constructing a fault knowledge element library; step 4, constructing a fault relationship of the fault description file; and 5, constructing a fault relation reasoning model, and diagnosing the aircraft system fault through the fault relation reasoning model. According to the method, various fault description files such as various maintenance manuals and fault trees are collected, a text processing mechanism and a knowledge graph mode are adopted, the incidence relation between faults is extracted from a fault description text, a fault reasoning relation graph is constructed, and a fault diagnosis reasoning model for fault relation reasoning oriented to knowledge expression is constructed; the method has the capability of reasoning unknown faults and relations based on the existing faults and relations, so that complementation of a fault reasoning model and prediction of unknown fault relations can be realized.

This invention provides a knowledge graph and user intent task decomposition method for satellite mission planning. Addressing the problems of existing knowledge storage methods leading to severe fragmentation of knowledge elements, existing remote sensing knowledge graph ontologies focusing on specific application contexts and failing to achieve multi-source heterogeneous data fusion, as well as poor scalability, the constructed knowledge graph divides entities into multiple functional libraries and relationships into intra-library entity relationships and inter-library entity relationships. This results in a massive network structure where entities within each functional library are closely connected, and entities between libraries also have close relationships, thus coupling the entire graph into a whole and achieving the fusion of multi-source heterogeneous data such as user needs and resource characteristics.

The invention relates to an intelligent science popularization content generation method and system fusing multi-dimensional disaster data, and the method comprises the steps: obtaining original disaster prevention data of a target region in a preset monitoring period through a data collection node disposed in a disaster monitoring network, and obtaining a standardized disaster data set through data cleaning and format standardization processing; performing spatial-temporal feature extraction processing on the set to generate a multi-dimensional feature vector set, inputting the multi-dimensional feature vector set into a dynamic disaster scene generation model for fusion calculation and scene simulation, and outputting a three-dimensional dynamic simulation result of a disaster development process; based on the result, science popularization knowledge element analysis and recombination are carried out, a structured science popularization content framework is generated, self-adaptive rendering is carried out according to personalized demand parameters submitted by a user terminal, customized disaster science popularization content is generated and pushed to a target user terminal, and the cognition and coping ability of the public to disaster risks can be improved.

This invention provides an intelligent scheduling method and system for cable operation and maintenance. First, the migration knowledge elements in the cable operation and maintenance migration knowledge base are deconstructed, dividing them into resource adaptation, process adaptation, and risk adaptation knowledge elements, and generating a knowledge element classification set. A dynamic portrait of the operation and maintenance scenario of the cable to be scheduled is constructed, integrating operational characteristics, operation and maintenance task requirements, and historical operation and maintenance performance data to generate a 3D scene portrait. The adaptation coefficient between each migration knowledge element and the portrait is calculated, and target knowledge element combinations are selected. Based on the target knowledge element combinations, an initial operation and maintenance scheduling scheme framework is constructed, incorporating various adaptation rules. Through multiple rounds of portrait iterative updates and adjustments, the scheme framework is adjusted, optimizing resource allocation, step sequence, and anomaly response plans, resulting in the final scheme and generating operation and maintenance scheduling instructions which are sent to the operation and maintenance execution terminal. This invention achieves intelligent, precise, and efficient cable operation and maintenance scheduling.

The invention discloses a disaster knowledge element extraction method driven by cross-disaster dynamic prompt and a related device, and relates to the field of artificial intelligence application and disaster emergency information extraction, and the method comprises the steps: building an entity category set and a relation set corresponding to disaster categories; constructing an entity and category two-tuple example set and an entity relationship triple example set by utilizing historical samples; on the basis of task categories, cue words are added into the example set, and a cue word template set is constructed; preprocessing the to-be-processed text, dividing the to-be-processed text into a plurality of blocks, and extracting a disaster category of each block; generating dynamic cue words of the blocks according to the disaster category of each block and the cue word template set; the dynamic cue word comprises a cue word for named entity recognition and a cue word for relation recognition; and the dynamic cue word of each block is utilized to guide the large language model to perform entity relationship extraction to obtain the entity recognition result and the relationship recognition result of the blocks, and efficient extraction of knowledge elements in the disaster field can be realized without large-scale data labeling.