61 results about "Metaknowledge" patented technology

The invention relates to the technical field of artificial intelligence, and discloses a multi-agent collaborative data question-answering system and method based on a large model, and the system comprises an agent cluster module which is composed of five special agents, namely a data question-answering agent, a data extraction agent, a data analysis agent, a visualization agent and a quality examination agent, and achieves the task decomposition and collaborative execution through dynamic scheduling; the knowledge management module comprises a business knowledge base, a data element knowledge base and a user feedback base, and adopts a hierarchical knowledge fusion technology to provide domain knowledge support for the intelligent agent; and the supporting function module covers a front-end dialogue component and a verification and execution engine and is responsible for interactive interface rendering and result reliability verification. According to the method, the fine tuning requirement on the large model is remarkably reduced, the illusion of the large model is effectively intercepted through a dual verification mechanism, and the accuracy and reliability of question and answer results are improved.

The invention relates to the technical field of deep learning and small sample learning, in particular to a small sample capacity training method based on deep learning, which comprises the steps of 1, cross-domain data adaptation and feature alignment, 2, meta-knowledge distillation and prototype enhancement, 3, attention-guided small sample fine adjustment, and 4, model uncertainty quantification and iterative optimization. According to the small sample capacity training method based on deep learning, through cross-domain feature alignment, meta-knowledge distillation, prototype enhancement and dynamic iterative optimization, the problems of model overfitting and weak generalization ability in a small sample scene are solved, high-precision model training when the sample size is less than or equal to 50 is realized, and the training efficiency is improved. The method is suitable for data scarce scenes such as medical images and minority language processing.

The invention discloses a marketing content generation method and system based on big data, and relates to the technical field of big data processing and artificial intelligence, and the system comprises a multi-source data collection and fusion module which is used for receiving original multi-source data from different behavior sources, content sources and business scene sources, and performing structured processing, time relationship coordination and feature fusion on the original multi-source data to generate a fused portrait data set for describing the relationship between the user and the product. In the invention, through the multi-source data acquisition and fusion module, deep integration of heterogeneous multi-source data and unified representation of multi-modal features are realized, a high-quality data basis is provided for subsequent knowledge graph construction and content generation, and the problems of isolated island and insufficient feature utilization of traditional marketing data are solved; through a ternary knowledge graph construction module, information such as users, products, scenes and the like is constructed into a structured knowledge graph, and the semantic understanding ability and relevance of content generation are greatly enhanced.

The invention relates to the technical field of intelligent recommendation, and discloses an AI intelligent matching method based on a knowledge graph, and the method comprises the steps: constructing a quaternary knowledge graph containing a time dimension; identifying legal entities and semantic relationships by adopting an entity identification and relationship extraction technology; multi-level semantic features are extracted through a two-layer progressive semantic matching algorithm of a grammar layer, a semantic layer and a reasoning layer; constructing lawyer ability portraits based on a heterogeneous graph neural network and a time sequence perception graph convolution technology; progressive matching calculation is adopted, the optimal matching weight is learned through a multi-layer attention mechanism, and dynamically optimized intelligent matching is achieved. The technical problems of cold start, insufficient semantic understanding ability and poor timeliness processing ability in the existing legal consultation matching system can be solved, and the matching precision and the user satisfaction are 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 intelligent teaching, and provides a self-adaptive personalized teaching system based on AI and a knowledge graph, and the system comprises a knowledge graph construction module, a student portrait module, a self-adaptive recommendation module, a teaching interaction module, and a management module. The knowledge graph construction module comprises a data processing unit, a knowledge extraction unit and a graph storage unit; the data processing unit is used for carrying out preprocessing such as word segmentation and stop word removal on the teaching text; and the knowledge extraction unit is used for extracting knowledge point entities and relationships between the entities from the preprocessed text through a natural language processing technology. The knowledge points are subjected to fine-grained modeling through the knowledge graph, the AI algorithm is combined to analyze student learning behaviors and evaluation data, knowledge vulnerabilities and learning characteristics of students can be accurately captured, the error rate is lower than that of a traditional method, the self-adaptive recommendation module generates personalized learning paths based on the reinforcement learning algorithm, and the learning efficiency is improved. And repeated learning of students on invalid knowledge points is avoided.

The invention relates to the technical field of artificial intelligence, in particular to a software component analysis system based on artificial intelligence, which comprises a data acquisition and preprocessing unit, an AI-driven component identification and analysis unit, a risk detection and evaluation unit, a knowledge base and dynamic updating unit and a result visualization and disposal suggestion unit. According to the method, through a collaborative mechanism of multi-modal semantic fingerprint generation, cross-language component matching and dynamic and static feature dual verification, the bottleneck that traditional software component analysis is insufficient in recognition precision of high-confusion, cross-language and secondary packaging components is effectively broken through, and precise recognition and analysis of the components and key attributes thereof are achieved; therefore, a more reliable technical support which better meets actual business requirements is provided for software component security analysis.

The invention relates to the technical field of intelligent coal mine safety production, and discloses an intelligent coal mine safety early warning method and system based on deep learning, and the method comprises the steps: constructing a difficulty evaluation function, and achieving the progressive learning from simple to complex; based on a difficulty assessment result, a model-independent meta-learning algorithm is realized, so that the model quickly adapts to new mining area characteristics; constructing a privacy protection federated learning framework by using the meta-learning model, and realizing multi-mining-area cooperative training; a continuous learning module is constructed, and original experience is reserved when new knowledge is learned; constructing a meta-knowledge evaluation module to realize cross-mining-area safety knowledge sharing; according to the invention, the security risk identification accuracy is improved; multi-mining-area cooperative training is realized on the premise of protecting data privacy; the method has continuous optimization capability and effectively solves the problem of model drift; and efficient sharing and migration of cross-mining-area safety knowledge are realized.

The invention provides a construction method of a business element knowledge base and a local data analysis method based on the business element knowledge base. The business meta-knowledge base comprises table structure definitions for a plurality of business scenes, business rule documents and analysis templates. The local data analysis method based on the business meta-knowledge base comprises the following steps: receiving a user demand; generating a prompt instruction based on the user demand and the business meta-knowledge related to the user demand; sending the prompt instruction to the large language model, and receiving a structured data operation statement which is returned by the large language model and is generated based on the prompt instruction; and executing the structured data operation statement locally to obtain an answer to the user demand. According to the method, the business meta-knowledge base is constructed, then interaction between the local data and the large language model is achieved through the business meta-knowledge base, efficient data analysis is achieved through the large language model, and meanwhile the privacy security of the local data can be protected.

The invention discloses a battery health state detection method and device and a medium, and relates to the technical field of battery health management, and the method comprises the steps: collecting historical operation data and full-band EIS scanning data of a battery, constructing a battery data set, and independently collecting state parameters; training a deep reinforcement learning DRL intelligent agent through the battery data set, performing scanning decision on state parameters of the battery by using the trained DRL intelligent agent, and outputting a sparse frequency band decision vector; utilizing the sparse frequency band decision vector to control the EIS equipment to carry out sparse scanning on the target battery to obtain sparse frequency band EIS data; the method comprises the following steps: constructing a meta-knowledge base through various types of batteries, and training the meta-knowledge base through a meta-learning algorithm to obtain a reference migration model; and carrying out few-sample adaptation processing on the sparse frequency band EIS data by using the reference migration model. According to the method, sparse EIS scanning is carried out by dynamically selecting frequency points, and efficient real-time diagnosis is achieved.

The invention discloses an information retrieval method of a multivariate knowledge base in the field of constructional engineering based on a large model, and belongs to the technical field of information retrieval, the knowledge base in the field of constructional engineering including multivariate knowledge such as texts, drawings, specifications and standards, cases and the like is constructed, and cross-modal and cross-type accurate retrieval is realized by adopting a large model technology; the method specifically comprises the steps of preprocessing and structured representation of multivariate knowledge, fine-tuning training of a large model adapted to the field, semantic analysis of a user retrieval request, collaborative retrieval of multi-modal knowledge, intelligent sorting and optimization of retrieval results and dynamic updating and maintenance of a system. According to the method, the problems of limited knowledge types, insufficient semantic understanding, poor professional adaptability and the like in traditional constructional engineering information retrieval are solved, the retrieval accuracy, comprehensiveness and efficiency are remarkably improved, and professional and intelligent knowledge support is provided for constructional engineering employees.

The embodiment of the invention provides a dynamic model training optimization method and device, processing equipment, a product and a medium, and is applied to the technical field of model training. The method comprises the following steps: collecting at least one environment variable in a target scene in real time; according to the at least one environment variable, generating an environment change intensity index corresponding to the target scene, the environment change intensity index being used for quantifying an environment change degree in the target scene; under the condition that the environment change intensity index is larger than or equal to a preset threshold value, fast adaptation priori is obtained, and the fast adaptation priori is meta-knowledge generated based on an MAML algorithm in advance; based on the fast adaptation priori, the strategy of the intelligent agent is optimized, the intelligent agent can perceive the environment in the target scene and autonomously take action to achieve the preset target, and the fast adaptation priori is used for guiding and adjusting the input characteristics and strategy updating speed of the strategy network of the intelligent agent. The method solves the problem that the strategy of the intelligent agent in the non-stationary environment is easy to fail quickly.

The invention discloses a multi-modal intelligent garbage detection method based on dynamic adaptive meta-learning, and the method comprises the steps: constructing a framework based on a Faster R-CNN two-stage detector fusing uncertainty estimation and multi-modal features, fusing a dynamic adaptive meta-learning mechanism, and supporting and querying a set to achieve robust adaptation through environment interaction; an enhanced garbage proposal module is designed, multi-modal feature fusion is adopted, and deep interaction with query features is supported; and constructing an advanced garbage classification module, and introducing a dynamic soft attention mechanism to realize space-semantic alignment. In the training stage, mixed meta-learning is adopted, an extended FSOD data set is combined with self-supervised pre-training, N-way K-shot subtask training is performed, meta-knowledge is acquired, and dynamic fine tuning is performed by using a labeled junk image in combination with real-time environment data. In the optimization stage, real-time detection of various types of garbage is realized in the test stage through a multi-task joint loss function supervision framework.

A topic for building a new intent on which to train a chatbot can be received. A database of chatbot training data can be searched for a candidate intent having meta-knowledge similar to the received topic. Utterances associated with the candidate intent can be extracted. The received topic and the extracted utterances can be input to a trained machine learning model. The trained machine learning model generates example utterances for the new intent. The new intent with the generated example utterances can be used as training data for training the chatbot.

A knowledge-creation assistance device (100) comprises: a knowledge-structuring unit (111) that, on the basis of a measurement analysis result of a sample outputted from an analysis device, generates pre-complementary structured knowledge that is configured to include one or more items pertaining to measurement; a missing-knowledge estimation unit (113) that adds an item to the pre-complementary structured knowledge to generate post-complementary structured knowledge; and a knowledge-editing unit (114) that receives an editing instruction for the post-complementary structured knowledge and generates confirmed structured knowledge. The missing-knowledge estimation unit (113) adds the item to the pre-complementary structured knowledge by using an estimation model that is a machine learning model in which an item included in the pre-complementary structured knowledge is used as an explanatory variable, and an item included in the confirmed structured knowledge is used as an objective variable.

The invention discloses a sewage treatment lift pump AI intelligent regulation and control method and system based on working condition perception, and relates to the field of sewage treatment.The sewage treatment lift pump AI intelligent regulation and control method comprises the steps that multi-source heterogeneous working condition data are collected and preprocessed, and a regular real-time perception data sequence is output; carrying out calculation and feature extraction on the regular real-time sensing data sequence, and outputting a structured feature vector; inputting the structured feature vector into a preset lightweight causal graph model for causal inference, and outputting a causal task descriptor; taking a causal task descriptor as a retrieval key, performing matching in a historical operation case library, and outputting a similar historical strategy-result pair set; and jointly inputting the causal task descriptor, the similar historical strategy-result pair set and recent real-time interaction data into a meta-learner online adaptation module. According to the method, asynchronous updating is carried out on a historical operation case library, a lightweight causal graph model and a meta knowledge base of a meta learning device based on full-process data of a current control period.

The application provides an intelligent substation anomaly detection method based on meta-learning technology, first, a multi-working condition task library is constructed for meta-training. Secondly, a double meta-knowledge learning framework is designed to learn task-level model initialization parameters suitable for rapid adaptation. Subsequently, online reconstruction error driven anomaly detection is deployed, and the adapted model is used to calculate data reconstruction error in real time, and combined with the adaptive threshold to accurately distinguish and alarm grading. Finally, the system supports a dynamic updating mechanism, which continuously collects verified samples and periodically triggers meta-updating. The application significantly reduces the data dependence and deployment cost in new scenarios, improves the model's generalization ability to multiple working conditions and sensitivity to rare anomalies through meta-knowledge sharing, and ensures the long-term effectiveness of the detection system through online learning mechanism, providing an efficient, adaptive and evolutionary security protection solution for intelligent substations.

The application relates to the technical field of data mining, and discloses an intelligent mining system for research data, which comprises a heterogeneous data acquisition module, a semantic association analysis module, a decision graph generation module, a time sequence feature correction module and a knowledge distillation optimization module. The heterogeneous data acquisition module captures data features and constructs a multi-layer topology through a multi-source data perception node and a dynamic dimension fusion network; the semantic association analysis module analyzes semantic association by using a concept topology modeling unit, a knowledge vector clustering unit and a multi-modal switching link; the decision graph generation module generates a core decision benchmark framework based on a strategy optimization node and a rule reasoning engine; the time sequence feature correction module performs time sequence correction and noise compensation on semantic association; and the knowledge distillation optimization module detects deviation through an entity relationship evaluation network and feeds back an optimized decision framework. The system realizes intelligent acquisition of multi-source data, dynamic analysis of semantics and adaptive generation of a decision graph, and improves the precision and efficiency of research data mining.

The invention discloses a language model retrieval data processing method and device, a server and a storage medium, relates to the technical field of integrity verification of a model system knowledge base, and constructs a self-adaptive architecture from automatic verification of retrieval data consistency, isolation of inconsistent retrieval data and restoration of the retrieval data. Comprising the steps of performing risk assessment on dynamic knowledge vectors retrieved in real time; when the risk assessment reaches a preset risk level, calling a data verification contract from a preset distributed account book according to the unique identifier of the dynamic knowledge vector; the data verification contract is generated by executing an encryption process based on a meta-knowledge vector corresponding to the dynamic knowledge vector; the verification data verifies whether the contract is consistent with a dynamic data contract, wherein the dynamic data contract is generated by executing an encryption process based on the dynamic knowledge vector; and when the data verification contract is inconsistent with the dynamic data contract, calling a database interface to carry out failure marking on the dynamic knowledge vector in a database, and isolating the dynamic knowledge vector.

The application relates to the technical field of artificial intelligence and industrial simulation, in particular to an evolvable industrial simulation question and answer intelligent agent system and a simulation question and answer method thereof. The evolvable industrial simulation question and answer intelligent agent system comprises a knowledge base module, a tool calling module, an intelligent agent core module, a user interaction module and a log and audit module; the knowledge base module comprises a multi-element knowledge collection unit, a document loading and cutting unit, a vectorization embedding and storage unit, a semantic retrieval unit, a knowledge evolution unit, a rule base and a case base; the application has the beneficial effects that (1) knowledge is dynamically evolved and updated autonomously and timely; (2) the system has the capability of autonomously refining knowledge from simulation results; (3) tool calling is intelligentized, and the system realizes "question and answer as calculation"; (4) the system has strong capabilities of complex problem disassembly and iterative solution; and (5) the system has excellent adaptability and expandability.

The invention provides a federated learning method and system for proxy feature collaborative synthesis and meta-knowledge evolution oriented to model isomerism, and the method comprises the steps: repeatedly executing multiple iterations of global operation and local operation until a federated learning model meets a convergence condition; the global operation of the server side comprises the following steps: receiving element gradients uploaded by each client, performing security aggregation, and updating global sharing module parameters; the local operation of the client comprises the following steps of: decoupling a local model into a private module and a shared module based on local real data, and generating agent characteristics which are similar to the statistical distribution of the local real data and are decoupled in semantic information by a generator; local heterogeneous real features and proxy features are mapped to a shared potential space of a unified dimension; executing double-layer optimization of internal loop local adaptation and external loop element gradient calculation, and calculating element gradients for shared module parameters; only the element gradient is uploaded to a server side, and local joint optimization, feature mapping and element gradient calculation uploading operation are iteratively executed.

The application discloses an unsupervised meta-learning method based on visual causal intervention, which comprises an unsupervised meta-training stage and a supervised meta-testing stage. Specifically, in the meta-training stage, the method uses unlabeled support set samples to learn the causal relationship among input data, prior knowledge and predicted output and the causal relationship between hidden variables in the prior knowledge, eliminate the negative influence of context bias in the input data, and finally obtain unbiased meta-knowledge. In the meta-testing stage, the method uses the above obtained meta-knowledge as the initialization parameters of the model, combines supervised and unsupervised learning under the condition of a small amount of labeled support set samples, further eliminates the influence of data bias, and can accurately predict the label of the query set sample. The method can accurately predict the category of the output image in the few-shot image classification problem of the downstream task.

The application discloses an e-commerce API interface security vulnerability detection method based on a knowledge graph, which comprises the following steps: collecting API gateway logs and API document data of an e-commerce system, and constructing a business knowledge graph; obtaining an initial vulnerability detection rule library, and forming a rule knowledge graph; constructing a meta-knowledge graph; calling rules in the rule knowledge graph for matching detection; storing conflict events in the meta-knowledge graph; creating an embryo node, splitting a new rule node when the aggregation amount reaches a maturity threshold, and storing the new rule node as a birth event in the meta-knowledge graph; periodically calculating an effectiveness index, and storing the effectiveness index as a death event in the meta-knowledge graph when the effectiveness index is lower than an apoptosis threshold; analyzing the correlation mode between events, detecting API calling requests, and generating a vulnerability detection result. The application constructs business, rule and meta-knowledge graphs to realize API vulnerability detection self-evolution, and has the advantages of wide coverage, high accuracy and continuous optimization.

The application discloses a kind of based on Dirichlet process's continuous meta-learning method and device, comprising: at least 1 independent meta-knowledge distribution is constructed as initial meta-knowledge distribution;Buffer current time sequential data, based on initial meta-knowledge distribution, the number of new meta-knowledge distribution is dynamically added to the data of current time using Dirichlet process, and new meta-knowledge distribution is randomly initialized;Through Bayesian continuous learning method, the posterior of previous time meta-knowledge distribution is regarded as the prior of current time meta-knowledge distribution, and according to Bayes rule, the iteration update of meta-knowledge distribution is carried out in the data stream buffered, wherein, meta-knowledge distribution includes initial meta-knowledge distribution and the number of new meta-knowledge distribution increased at each time;The corresponding relationship between updated meta-knowledge distribution and target task is constructed, and the meta-knowledge initialization is carried out for the corresponding model of target task according to the corresponding relationship, then a small amount of labeled data is used to learn the model, and the model capable of realizing prediction is obtained.

The application relates to a causal DAG discovery method for an online service system based on fusion of soft priori. The method comprises the following steps: obtaining observation data and text meta-knowledge of the service system, preprocessing to form a standardized sample set, identifying variable types and semantics and outputting; generating a natural language description according to the variable semantics, querying a large language model for an ordered variable pair, analyzing to obtain three types of causal probability vectors, and calibrating to obtain an edge-level priori probability. An appropriate conditional independence test method is selected, high-confidence independent / dependent sentences are divided, and weights are assigned. A candidate directed acyclic graph is selected as an initial structure, parameters are estimated by linear regression, and data fitting scores are calculated, language priori scores, conditional independence penalty terms and counterfactual self-consistency penalty terms are calculated. Fusion is carried out into a hybrid score function, discrete optimization is carried out under the constraint of a directed acyclic graph, and a causal graph structure with the optimal score is output. The method can improve the efficiency and accuracy of a smart operation and maintenance system.

The application relates to the technical field of knowledge graph construction and intelligent information retrieval, and discloses a retrieval method based on a knowledge graph and related equipment. The method comprises the following steps: constructing a triple basic unit knowledge graph of entity-relation-entity; receiving a query sentence, and performing deep semantic segmentation on the query sentence to obtain semantic units; analyzing the query intention of a user according to the segmented semantic units and the triple basic unit knowledge graph, and mapping the query intention of the user to the triple basic unit knowledge graph; according to the mapped query intention, finding the optimal retrieval result on the triple basic unit knowledge graph based on a mapping relationship, and feeding back the optimal retrieval result to a retrieval end, thereby completing the retrieval work based on the knowledge graph. Through a recommendation algorithm, the possible results are compared according to the sorting of a specific algorithm, and a higher-precision retrieval result is obtained, so that accurate result matching is further realized.

The present application relates to the field of complex equipment design, and specifically relates to a complex equipment modular configuration design knowledge graph construction method, comprising: constructing configuration design module meta-knowledge ontology, proposing a configuration design event class knowledge multi-element representation method and a constraint rule class constraint expression and design parameter strong association knowledge representation method, and designing a constraint rule expression and design meta-parameter matching fusion algorithm to integrate module meta-knowledge into the configuration design knowledge graph. With the help of shape constraint language, an algorithm for automatically converting constraint rules into entity path constraints and design parameter attribute value constraints is designed to generate a constraint resource description architecture diagram; all project configuration design instances are extracted and converted into project resource description architecture diagrams; the constraint resource description architecture diagram is used for integrity checking of the project resource description architecture diagram, and the project configuration design instances that pass the checking are integrated into the knowledge graph to realize the integration of project instance knowledge and module meta-knowledge.

The invention discloses an industrial fault diagnosis-oriented multi-source semantic alignment adaptive meta transfer learning method. The method comprises the following steps of: 1, constructing a task of a meta training stage by adopting a leave-one method; step 2, pre-training; step 3, meta-training; step 4, fine adjustment of elements; and 5, calculating a test sample for testing. The key problems of fault sample scarcity and multi-source domain knowledge migration in the field of industrial fault diagnosis are effectively solved; more refined meta-knowledge extraction is realized, and cross-domain element loss is remarkably reduced; according to the method, prototype instance calibration and a micro-regression reprojection method are adopted, the influence of distribution difference among multiple domains on diagnosis performance is effectively reduced, in the fine adjustment stage, a reprojection prototype is replaced with a learnable matrix, a self-adaptive dynamically-adjusted prototype is obtained, optimization of meta-learner parameters is achieved through a semantic alignment bidirectional embedding module, and the accuracy and accuracy of diagnosis are improved. And the diagnosis precision and generalization ability of the model are further improved.

The invention provides a DDoS attack real-time prediction method and system based on a dynamic heterogeneous distillation network, and the method comprises the steps: constructing a dynamic heterogeneous topological graph, carrying out the space-time convolution calculation and meta-knowledge distillation of the dynamic heterogeneous topological graph, obtaining a space-time distillation prediction network, and carrying out the multi-task driving to capture a network attack behavior; performing pulse frequency domain analysis on the network attack behavior, adding a Hamming window to the traffic data time sequence and executing fast Fourier transform to extract a frequency domain component, generating an adversarial sample and injecting disturbance, and detecting whether a low-frequency pulse attack exists in the network attack behavior; and blocking network attack behaviors in real time, performing incremental training on the space-time distillation network, and updating parameters of the space-time distillation network. According to the method, the defect of high omission ratio of a static mode can be overcome, so that zombie host migration has no place to hide, the bottleneck of real-time response of network attack behaviors is broken through, the robustness of a space-time distillation prediction network is improved, a prediction blind area is filled up, and a DDoS defense core pain point is solved.

The invention discloses a multi-granularity learning path recommendation and optimization method based on a large language model. The method comprises the following steps: preprocessing a learning data set; constructing a multi-granularity knowledge structure; constructing a multi-level incidence matrix; constructing a double-layer concept lattice; constructing a knowledge unit and a skill multi-granularity dependency graph; generating a multi-granularity learning path; and recommending and optimizing the path. A three-level hierarchical knowledge structure of a large unit, a knowledge unit and a fine-grained skill is constructed through a large language model, and coarse-grained knowledge unit overall planning and fine-grained skill collaborative recommendation are realized. Compared with the prior art that multiple granularities are defined only according to the number of knowledge nodes of learning resources and clear hierarchical collaboration is lacked, the method ensures that the learning path conforms to an internal logic system of subject knowledge, enables the recommended path to directly guide the practical operation practice of learners through accurate mapping of fine-grained skills and specific topics, and improves the learning efficiency. The problem that a knowledge system and practical operation are separated in the prior art is solved, and the practicability and the performability of the path are remarkably improved.