65 results about "Knowledge generation" patented technology

The invention discloses an ancient poetry multi-mode teaching system based on HSK classification, and belongs to the field of intelligent teaching. Comprising a learning evaluation portrait module, a learning path generation module, a physiological emotion monitoring module, a cognitive load analysis module, a vocabulary grammar enhancement module, a modal semantic analysis module, an artistic conception analogy generation module, a holographic scene generation module, a feedback interaction experience module, a content dynamic adjustment module, a recitation pronunciation correction module and an expansion knowledge generation module. A performance analysis feedback module and a review memory enhancement module; according to the method, learning tasks can be updated in real time along with the change of the ability of students, a one-step teaching mode is avoided, the learning efficiency and adaptability are improved, the learning experience and continuity are greatly improved, cultural faults can be effectively reduced, non-native Chinese language persons can understand the deep significance of ancient poetry more easily, and the learning efficiency is improved. The memory depth of the learner to the poetry artistic conception and vocabularies is greatly improved, and the learning motivation is also improved.

The invention relates to the technical field of computers, and discloses a fragmentation block chain federal learning method based on a large language model multi-agent, and the method comprises the steps: S1, initializing a client; s2, dynamic fragmentation scheduling and distribution; s3, generating and uploading local knowledge; s4, intelligent agent collaborative routing and knowledge acquisition; s5, knowledge fusion and model updating; and S6, repeatedly executing the steps S3 to S5 until the model converges or reaches a preset number of iterations. According to the invention, under a decentralized and fragmented federated learning architecture, the complex reasoning ability of a large language model and the autonomous cooperation mechanism of three multi-agent systems, namely a fragmented scheduling agent, a fragmented knowledge state agent and a global knowledge routing agent, are deeply fused; according to the mechanism, dynamic optimization of a bottom layer fragment structure and intelligent routing of high-value knowledge are achieved in an intelligent mode, and therefore the overall efficiency and model performance of a system under the condition of heterogeneous data and heterogeneous equipment are remarkably improved.

A multi-generation RAG process generates a first-generation prompt for input to LLMs. The first-generation prompt may specify a concept, raw data, and a first-generation request to draw inference of information related to the concept using the raw data. The process provides the first-generation prompt for execution by the LLMs and receives a first-generation response. The process iteratively updates the inference of information using the LLMs. The iteratively updating includes using the inference of information related to the concept that are received from a previous-generation response as contextual information for a subsequent-generation RAG process. The process receives a subsequent-generation response generated by executing the LLMs and stores the iteratively updated inference of information related to the concept for retrieval by the machine-learned language model.

The invention discloses a building industry multi-subject collaborative full life cycle carbon emission management system and method, and relates to the technical field of building industry carbon emission management. The system comprises a carbon perception and data collaboration module, a carbon simulation and deduction module, a carbon optimization and collaborative regulation and control module, a carbon accounting and auditing traceability module and a carbon decision and knowledge generation module. Multi-source heterogeneous data is converted into a carbon flow unit with time, space and responsibility subject attributes, data checking, locking and compliance sharing are achieved by means of a multi-subject data collaboration pool, carbon data barriers among all responsibility subjects are broken through, the problems of fragmentization of carbon data collection and lack of collaboration are solved, and the data collection efficiency is improved. And meanwhile, the carbon accounting and audit traceability module completes multi-dimensional accounting based on the carbon flow unit, and establishes a digital carbon chain pointing to an original data source and a responsibility subject, so that full-link traceability of carbon emission data is realized.

The invention discloses a knowledge-intensive task-oriented thinking chain prompt optimization method and system, and belongs to the field of natural language processing. When the model is used for reasoning knowledge-intensive tasks based on thinking chain prompts, the generated thinking chain often has the problems of knowledge missing, irrelevance, even errors and the like. Therefore, a prompt optimization process is improved in a manner of combining knowledge generation and knowledge evaluation, and a thinking chain prompt which more pays attention to knowledge quality is obtained, so that the knowledge integrity, correlation and correctness of a model generated thinking chain are improved. The knowledge generation means that the guide model explicitly generates knowledge fragments in the thinking chain; according to knowledge evaluation, a loss function containing answer accuracy and multi-dimensional knowledge indexes is designed, and gradient information of the loss function is utilized to select an optimal candidate word to update and optimize prompts. The system comprises a thinking chain knowledge generation module, a knowledge evaluation module, a candidate word initialization module, a candidate word gradient calculation module and a prompt updating module.

The invention discloses a multi-modal health knowledge generation and integration system, which is applied to the technical field of fusion and application of health data, and comprises a multi-modal data acquisition and preprocessing module used for acquiring multi-modal health data and preprocessing the multi-modal health data; the multi-level fusion framework design sub-module comprises data-level fusion based on a time alignment algorithm, feature-level fusion based on a Transform cross-modal attention mechanism and decision-making-level fusion based on an ensemble learning algorithm; and the intelligent knowledge generation sub-module comprises health knowledge graph construction based on a graph neural network, personalized health risk report and intervention scheme generation based on a conditional generative adversarial network, visual image feature contribution degree based on gradient weighted class activation mapping and text data key factor analysis based on an SHAP value. According to the method, the bottleneck of a data island is fundamentally broken through, and the generation quality and the application value of health knowledge are improved.

The invention discloses a retrieval enhancement generation method and system based on long-term memory and multi-source knowledge iterative fusion. The method comprises the following steps: firstly, after user inquiry is received, triggering long-term memory database retrieval, external knowledge base retrieval and large language model internal knowledge generation in parallel to construct a multi-source candidate knowledge base, and improving result correlation by adopting a self-adaptive retrieval method based on a dynamic top-k value; secondly, inputting user query and multi-source knowledge into the large language model for iterative fusion, and generating a final answer through conflict detection, fusion processing and confidence evaluation; and finally updating the long-term memory database. The method can improve the accuracy and stability of the generated answers in a multi-round interaction and complex task scene, reduces the risk of answer one-sidedness, inference incompleteness or fact deviation, enhances the multi-source knowledge utilization ability and long-term learning ability of the system, and improves the user experience. Therefore, the comprehensive performance and reliability of the system in applications such as intelligent question answering, information retrieval assistance and text generation are obviously expanded.

The invention discloses an electronic archive intelligent data management method and system based on a large model. The method comprises the following steps: firstly, acquiring an electronic archive file, performing unified access, identifying a text through an OCR (Optical Character Recognition) technology, and reserving original layout coordinate information; analyzing the document page by page to generate a page-level structured object containing text and position mapping; then, on the basis of a preset domain prompt template, calling a large language model to extract file metadata from the page-level object, recognizing entities and relationships in combination with file knowledge ontology, and generating a semantic triple containing an evidence source page number; and finally, fusing the metadata and the triple to construct an electronic archive knowledge graph with a page-level traceability function. According to the invention, through a whole-process semantic governance mechanism from access, analysis to knowledge generation, the problems of difficult analysis of unstructured data, weak semantic understanding and lack of traceability in traditional archive management are effectively solved, and the intelligent level and data credibility of archive governance are significantly improved.

The invention discloses a 6G-oriented cross-domain knowledge-driven network intelligent scheduling system and a 6G-oriented cross-domain knowledge-driven network intelligent scheduling method. The architecture comprises a cross-domain knowledge collaborative evolution module, a network closed-loop management module and a multi-normal-form learning decision module, the cross-domain knowledge collaborative evolution module is used for disassembling a network state into an environment domain, a network domain and a user behavior domain, and online incremental fusion of three-domain knowledge is realized through a graph neural network; dynamically updating the global knowledge base based on knowledge distillation; the network closed-loop management module is used for constructing a function closed loop of perception-reasoning-knowledge generation-decision issuing-verification optimization-memory retrieval; and the multi-normal-form learning decision module is used for fusing meta learning, reinforcement learning, federal learning and self-supervised learning to realize rapid migration and continuous evolution of the strategy. According to the method, the network data throughput, the fault recovery speed, the flow prediction precision, the resource fairness, the path efficiency and the task success rate are improved, and the method is suitable for complex 6G scenes such as air-ground integration and low-altitude traffic control.

The invention discloses a domain knowledge content generation method and system based on retrieval enhancement generation, and belongs to the technical field of natural language processing and knowledge generation. Performing fuzzy semantic retrieval in the structured knowledge base to obtain semantic matching entries and calculate semantic relevancy; constructing a semantic fusion representation vector to form an enhanced representation set; performing context alignment with the query vector, and extracting a causal path fragment with the maximum weight; inputting the causal path and the query vector into a large language model, and executing a condition generation operation; analyzing and judging whether the generated content has term jump or logic chain scission or not based on the term dependency graph, if so, adjusting a context window and regenerating, and finally outputting professional content meeting dependency consistency; according to the method, the accuracy and logic consistency of professional field text generation are effectively improved, and the method is suitable for knowledge generation tasks in complex semantic scenes.

The invention belongs to the technical field of hole part process decision and computer aided process design (CAPP), and relates to a body-based hole part geometric tolerance assessment method. The method is specifically realized through the following steps: step 1, constructing a knowledge generation ontology in the geometric tolerance assessment domain of the hole parts; 2, establishing an SWRL inference rule for geometric tolerance evaluation of the hole parts; 3, extracting part structure and surface information, and establishing assertion formula sets AC and AS; 4, extracting geometric tolerance specification information of the hole parts, and constructing an assertion formula set AG; and step 5, constructing an instantiated ontology knowledge base generated by geometric tolerance assessment domain knowledge of the hole parts, and reasoning the instantiated ontology knowledge base by using a reasoning engine according to the SWRL reasoning rule base to generate geometric tolerance eligibility of the hole parts.

The application discloses a brain map drawing electronic pen and a handwriting knowledge generation and database construction device and method, and relates to the technical field of knowledge generation and database construction, and specifically relates to a brain map drawing electronic pen and a handwriting knowledge generation and database construction device and method.

This disclosure relates to knowledge generation and implementation. A knowledge graph system comprises at least one processor, at least one database communicatively connected to the at least one processor, and a memory storing executable instructions. When executed, the instructions cause the at least one processor to aggregate data associated with a plurality of entities, the aggregated data reflecting one or more relationships between two or more of the plurality of entities. Attribute data identifying loan amounts, property values, and appraisal sources may be extracted from the aggregated data. A knowledge graph data structure may be generated having a plurality of subject notes corresponding to the extracted attribute information. Statistical distributions of attributes associated with one or more appraisal sources may be generated and an anomaly in a first statistical distribution may be detected based on a comparison of the first statistical distribution with a second statistical distribution.

The present invention provides a method, device, and robot for generating knowledge for solving applied problems, comprising: obtaining input text corresponding to a target applied problem input by a user; segmenting the input text into sentences and assigning a corresponding sentence number to each acquired text sentence; obtaining the noun content associated with each text sentence to form noun content recognition knowledge; and matching noun content extension knowledge associated with the input text from a pre-constructed knowledge base based on the noun content associated with each text sentence and its corresponding sentence number. The method, device, and robot for generating knowledge for solving applied problems provided by the present invention can generate noun content extension knowledge that is not contained in the applied problem itself but is of great value in solving the problem, thereby more effectively assisting the development of various systems related to applied problems and improving the accuracy of problem solving in various systems to a certain extent.

This application provides a method and system for dynamic knowledge updating of a large-scale model for operation and maintenance (O&M) technical services. It relates to the field of O&M technology, and models the knowledge generation triggering factors, iteration impact range, and knowledge association strength in multi-source O&M knowledge update sources to obtain a knowledge full-link evolution set including knowledge generation trigger records, iteration impact assessment results, and cross-knowledge association networks. Then, dynamic coupling rules are generated based on the real-time scenario characteristics of O&M tasks and this knowledge full-link evolution set. These rules are used to filter and integrate the knowledge full-link evolution set to obtain a scenario-coupled knowledge set. This set is embedded into the knowledge architecture of the large-scale O&M technical service model, and storage node parameters and retrieval logic are adjusted to obtain the basic updated model. Finally, the basic updated model is dynamically adapted in multiple dimensions, and the knowledge call priority sequence and storage node parameters are adjusted in conjunction with changes in scenario characteristics to generate an adapted updated model, thereby improving the adaptability of the large-scale O&M technical service model.

The application relates to a medical field knowledge fusion method based on label and content hierarchical fusion, and belongs to the technical field of knowledge fusion and artificial intelligence. Medical knowledge of different knowledge graphs is fused to construct a medical field knowledge base and provide large-scale medical industry knowledge services. The method comprises three steps of defining a meta label of an entity in a medical knowledge graph, data label alignment, and label content fusion. The method has the following characteristics: firstly, in the data label alignment, a data label alignment method based on integration of string measurement and semantic measurement is designed, and data label alignment is performed by using string features and semantic features. Secondly, in the label content fusion, a joint mechanism of content aggregation verification and unstructured knowledge generation is proposed, which aims to extract common knowledge of different granularities in different knowledge graphs, verify contradictory knowledge, and convert common triple knowledge into unstructured sentences to construct a medical knowledge base.

The embodiment of the invention discloses a medical knowledge base knowledge generation method and electronic equipment. The method comprises the steps of determining a knowledge template of a medical knowledge base under the condition of obtaining input data; and obtaining a knowledge generation model, and according to the input data and the knowledge template, through the knowledge generation model, generating target knowledge to be stored in the medical knowledge base so as to store the target knowledge in the medical knowledge base. According to the technical scheme, the problem that the knowledge generation efficiency of the medical knowledge base is low can be solved.

The embodiment of the invention provides a cross-data-source knowledge generation method and device and electronic equipment, and relates to the technical field of information, and the method comprises the steps: obtaining a configuration template, declaring a knowledge template in the configuration template, and enabling the knowledge template to comprise at least one piece of first to-be-filled data, the configuration template also declarates a first service type, a first interface identifier and a first query statement corresponding to each piece of first to-be-filled data, calling a first interface through a first service to access a first data source for each piece of first to-be-filled data, and querying in the first data source according to the corresponding first query statement, and obtaining a query result of the first to-be-filled data, and filling the query result of each first to-be-filled data into a knowledge template to obtain knowledge. According to the embodiment of the invention, the cost of multi-source access in large model application is reduced.

The invention relates to a rapid knowledge-based method for highway engineering investigation and design data based on a large model, and belongs to the technical field of digitization of highway engineering investigation and design data. The method comprises the following steps: performing structured processing on unstructured data in a PDF file, then performing multi-level decomposition to obtain a highway engineering investigation and design information depth definition table, and constructing a highway engineering investigation and design domain ontology based on the table; performing alternate interaction by utilizing prompt and the large model to obtain an entity and attribute preliminary result in the document; performing entity alignment and ambiguity elimination on the preliminary result data so as to create a knowledge graph of the specific engineering project; and after the vector database and the graph database are constructed, retrieving from the two databases by using a large model and a retrieval enhancement technology to obtain corresponding answers of the questions. The problems of difficult investigation and design data management and low retrieval speed are solved, and compared with the prior art, the retrieval efficiency and fineness are improved by at least 10 times or above, and the user question answering process is easy and friendly.

The invention relates to a code knowledge generation method and device, a computer readable storage medium and a computer program product, and the method comprises the steps: extracting an internal code of each layer of method called by each service method and a quoted data query statement from a project code, and obtaining a first code extraction file; determining a first text length of the first code extraction file, and taking a service method of which the first text length exceeds a first preset threshold value as an entry method; for each data access method in the project code, reversely tracing a call chain corresponding to each entry method, and extracting codes necessary for the operation of the data access method from codes of each layer of method included in the call chain to obtain a second code extraction file; and performing code analysis on the second code extraction file and the first code extraction file of which the first text length does not exceed a first preset threshold value through the large model to generate code knowledge. According to the method, the accuracy and reliability of code knowledge extraction can be remarkably improved.

The invention provides a moving behavior prediction method and device based on a natural language large model agent, and relates to the technical field of artificial intelligence, the method is applied to an LLM agent, and the method comprises the following steps: converting moving behavior track data of a target user into a user moving behavior prediction task described by a natural language; calling a space-time memory module, and extracting long-term memory, short-term memory and user portrait information of the target user based on the moving behavior track data; calling a geographic knowledge generation module, aligning the multi-scale urban space structure based on the movement behavior trajectory data to generate a first candidate place, and calling a sharing mode mining module to generate a second candidate place according to the trajectory data of similar users; and a comprehensive reasoning module is called, and a final prediction result of the user movement behavior is generated through a natural language large model based on the long-term memory, the short-term memory, the user portrait information, the first candidate place and the second candidate place of the target user, so that the prediction accuracy is improved.

The invention relates to the technical field of knowledge bases, can be applied to the field of science and technology finance / digital medical treatment, and discloses a knowledge base knowledge generation method and device, electronic equipment and a storage medium. The method comprises the steps that after knowledge generation process parameters configured by a user are obtained, the knowledge generation process is organized and managed in a unified mode, and the process parameters comprise knowledge data source types, knowledge generation structure requirements, knowledge generation rules and knowledge storage strategies. Collecting to-be-processed data based on the configured knowledge data source, processing the to-be-processed data under the constraint of the generation rule, and outputting structured knowledge content according to a preset structure; then manual auditing, intelligent auditing or combined auditing is executed according to the auditing configuration; after the audit is passed, the knowledge content is written into the target knowledge base according to a storage strategy, and vectorization processing is completed. According to the method, integrated automatic execution of knowledge generation regularization control, quality auditing and warehousing processing is realized.

The invention relates to the technical field of tobacco industry roll package production quality control, in particular to a roll package quality intelligent knowledge base model construction method and system and a storage medium, and the method comprises the steps: obtaining and processing multi-modal data in the roll package quality field; constructing a roll quality knowledge base model based on a large language model; performing multi-modal semantic alignment on the multi-modal data by adopting the roll quality knowledge base model so as to obtain structured data; according to the structured data, a knowledge question-answer pair is generated by adopting the roll quality knowledge base model; and performing dynamic index optimization deployment on the roll quality knowledge base model. According to the embodiment of the invention, cross-modal semantic fusion of text description and equipment parameters is realized by constructing a unified semantic space, the semantic fault problem of a traditional method is fundamentally solved, a self-guiding knowledge generation mechanism gets rid of dependence on manual annotation, and automatic expansion of question and answer pairs is realized.

The invention discloses a context dynamic tracking and course learning method oriented to multi-round dialogue perception. The method comprises the following steps: S1, performing utterance feature extraction in an utterance feature extraction module: extracting a semantic feature vector of each utterance in a dialogue through a first pre-training language model; the first pre-training language model is a RoBERTa model; s2, performing common sense feature extraction in a common sense feature extraction module: extracting common sense feature vectors related to emotion dynamic from the semantic feature vectors through a second pre-training knowledge generation model; the second pre-training knowledge generation model is a COMET model; s3, performing dynamic tracking modeling through a context dynamic tracker, and performing speaker information modeling through a speaker information module; and S4, course learning training: introducing a dynamic weight distribution mechanism and a weight loss function to carry out adaptive training on the dialogue sample. According to the method, dynamic contexts, multi-dimensional speaker features and common knowledge can be deeply fused, and the understanding of complex dialogue emotions is improved.

The invention provides a large-model knowledge dynamic updating method and system for operation and maintenance technology service, relates to the technical field of operation and maintenance, and aims to generate a trigger factor, an iterative influence range and knowledge association strength modeling for knowledge in a multi-source operation and maintenance knowledge updating source. Obtaining a knowledge full-link evolution set comprising a knowledge generation trigger record, an iteration influence evaluation result and a cross-knowledge association network; a dynamic coupling rule is generated based on operation and maintenance task real-time scene features and the knowledge full-link evolution set; screening and integrating the knowledge full-link evolution set by utilizing a dynamic coupling rule to obtain a scene coupling knowledge set; embedding the model into a knowledge architecture of an operation and maintenance technology service large model, and adjusting storage node parameters and retrieval logic to obtain a basic updated model; and finally, carrying out multi-dimensional dynamic adaptation on the model after basic updating, adjusting a knowledge calling priority sequence and storage node parameters in combination with scene feature changes, generating an adaptive updated model, and improving the adaptability of the operation and maintenance technology service large model.

The application belongs to the technical field of industrial digitization and artificial intelligence, and particularly relates to a kind of automobile parts enterprise knowledge generation and intelligent question answering system based on large model technology, through integrating multiple function platforms, the knowledge retrieval and generation path of "plug-in-knowledge base-large model" is constructed, and the large model technology is combined with the actual knowledge data of enterprise deeply, in the form of multi-modal of voice / text input / output + digital person audio picture synchronous dynamic interaction, the full-link intelligent service of "intention recognition-knowledge retrieval-business execution" is realized, the goals of efficiency improvement and digital operation agile iteration and continuous upgrading of manufacturing, research and development, operation and other businesses are achieved, the technical problems of poor adaptability of general model industry knowledge, split of multi-modal interaction and business system, low utilization rate of enterprise massive heterogeneous data and low efficiency of internal knowledge acquisition are solved.

The invention discloses an air and space big data intelligent data management system and method based on four-database collaboration and knowledge generation. A core layer adopts a temporary storage database, a controlled database, a product database and a knowledge database to form a four-database architecture; and the service layer adopts a data governance engine, a productization engine, a knowledge generation engine and an intelligent enabling engine. The system combines a data governance engine and a productization engine to carry out collaborative linkage to carry out data governance. Depositing knowledge assets into a knowledge base by relying on a knowledge generation engine; performing intelligent enabling application of each link of the data governance process by means of an intelligent enabling engine; and an intelligent enabling application effect is fed back, and knowledge assets in the knowledge base are optimized. According to the intelligent closed-loop data governance system, data governance implicit knowledge is converted into dominant knowledge and reused, the self-adaptability and the intelligent level of the data governance process are improved through knowledge reverse enabling, and then an intelligent closed-loop data governance system capable of self-learning, self-optimizing and continuously adapting to service requirements is formed.

According to the multi-modal personalized content generation and interpretation method and device for interaction between the user and the expert agent, the medium, the program product and the terminal provided by the invention, a traditional single text output mode is broken through, personalized generation of multi-modal contents such as texts, voices and videos is realized in combination with user knowledge preferences and user cognitive features, and the user experience is improved. The multi-modal personalized content adapts to receiving habits of different users. According to the method, the domain knowledge graph and the interaction context are fused, hierarchical disassembly, cross-domain association and personalized application suggestion of knowledge are realized, a user is helped to quickly master core knowledge and application scenes through deep interpretation, and the problem that interpretation of a traditional scheme is shallow is solved. According to the method, a closed-loop feedback optimization mechanism of user feedback, model optimization and strategy adjustment is established, it is ensured that knowledge generation and interpretation ability is dynamically upgraded along with user requirements, the method adapts to continuously-changing user requirements and domain knowledge updating, and therefore the service quality is continuously improved.

The invention provides a large-model intelligent document generation and optimization method and system fused with external retrieval, and relates to the technical field of artificial intelligence and document generation, and the method comprises the steps: obtaining theme information, constructing retrieval query, forming a knowledge fragment set, removing redundancy, and constructing a structured knowledge unit; according to the method, a dynamic routing mechanism is used for calculating the adaptation degree of knowledge units and paragraphs to generate an alignment index, a language model is controlled to selectively reference knowledge generation content, consistency verification and correction are carried out, and the accuracy, correlation and logic consistency of document generation are improved.

The application discloses a kind of food additive knowledge generation method and system based on multi-modal, it is related to food additive knowledge generation technical field, by regulation rule slot disassembly module and product scene slot construction module, regulation side and scene side are respectively transcribed into slot chain, so that regulation constraint object, food category, use condition, label requirement and other information and ingredient table, label identification, process specification, detection specification and question and answer target form the structure basis that can be aligned item by item, avoid only stay in the level of article original display or keyword hit.Multiple modal semantic module sorts out regulation text, ingredient table text, label image identification text, detection item specification, process specification text and question and answer request, so that subsequent processing is carried out under the same semantic fragment system, reduce the alignment fracture caused by the difference of expression from different sources, so that scene input has sustainable acceptance ability.