1791 results about "Graph generation" patented technology

The invention provides an advertisement recommendation method and system fused with user dynamic behavior modeling. The method comprises the following steps: acquiring a cross-device operation record of a user on multiple terminal devices, and synchronously acquiring a device unique identification code, high-frequency click area data and an operation timestamp; and constructing a space-time relation graph based on combined analysis of the equipment switching time interval and the click behavior, and generating a feature set reflecting the behavior correlation degree. The trajectory similarity is calculated at the edge calculation node through hardware acceleration, and migration correlation parameters are generated. And inputting the space-time relation graph into a behavior analysis model to extract coherent behavior characteristics, and forming a multi-dimensional scene portrait in combination with geographic position change and interface element distribution. And finally, through identifying a user cross-device interest migration mode, dynamically screening advertisement contents matched with the current scene, and realizing accurate pushing strategy optimization based on behavior continuity. According to the technical scheme provided by the invention, the accuracy of cross-terminal advertisement recommendation and the scene response speed are remarkably improved.

The invention discloses a network attack AI detection analysis method and system based on the smart Internet, and belongs to the technical field of network security protection, and the method comprises the steps: building an attack feature library through distributed edge nodes in a cooperative manner, generating feature parameters of each node based on a local attack event, and transmitting the feature parameters to a central server for dynamic fusion through encryption; constructing a multi-modal interaction graph, identifying a potential attack link based on association strength among graph nodes, and deducing an attack intention to generate a defense strategy; deploying a virtualized network environment, dynamically injecting induction characteristics, and adjusting an induction strategy in real time according to the interaction behavior of an attacker; and monitoring an abnormal mode of the user behavior sequence, triggering an AI interaction verification process and storing a defense strategy. According to the method, rapid collection and fusion of network attack features are realized, the detection delay of network attacks is reduced, the accuracy of attack prediction is improved, the flexibility and effectiveness of network attack confrontation are enhanced, and the defense intelligence and adaptive ability of the whole network are improved.

The invention discloses an agent question-answering method, device, equipment and product based on knowledge graph recall, and relates to the technical field of large models, agents, artificial intelligence and knowledge graphs. The agent question-answering method comprises the steps that a target question input by a user in an intelligent interaction page is obtained, keywords in the target question are extracted, and the keywords are extracted; a target knowledge graph node matched with the keyword is determined in a target knowledge graph, a target sub-graph is determined at least based on the target knowledge graph node, and the target knowledge graph comprises a plurality of knowledge graph nodes; and generating an answer corresponding to the target question at least based on the target question and the target sub-graph through the question and answer large model. By introducing the knowledge graph, in the process of generating the answer by the question and answer large model, deep reasoning can be carried out based on the input question along the path in the knowledge graph by using the multi-level and structured information of the knowledge graph, so that the processing capability of the question and answer large model on the complex question is improved, and the accuracy and comprehensiveness of outputting the answer are effectively improved.

The invention provides a conference activity execution task decomposition, arrangement and management method based on an AI technology, and relates to the technical field of conference management, and the method comprises the steps: obtaining conference demand information, converting the conference demand information into digital representation data, employing a bidirectional recursive decomposition strategy to carry out task decomposition, and generating subtask data; calculating a task similarity matrix and a conflict matrix to construct a task execution directed graph to generate initial arrangement data; collecting an execution state to generate feedback data, calculating a risk assessment score, and adjusting a task execution scheme when the risk assessment score exceeds a threshold value. According to the invention, the execution efficiency of conference activities can be improved, the resource conflict risk is reduced, and dynamic optimization adjustment is realized.

The invention provides an enterprise intelligent diagnosis method, system and device based on a large model and a medium, and belongs to the technical field of enterprise diagnos.The method comprises the steps that enterprise heterogeneous data are collected through a multi-source data interface, cleaning, feature extraction and cross-modal alignment fusion are carried out, and enterprise real-time data are obtained; constructing a knowledge graph through a graph attention network on the basis of an industry index to which an enterprise belongs, and updating association weights among nodes at regular time; inputting enterprise real-time data into the pre-trained multi-modal large model for preliminary analysis, and outputting a risk thermodynamic diagram; key abnormal indexes are identified from the risk thermodynamic diagram, sub-graphs related to the key abnormal indexes are extracted from the knowledge graph, structured prompt words are generated from the sub-graphs, then the structured prompt words and standardized enterprise real-time data are jointly input into a multi-modal large model for joint reasoning analysis, and then a visual diagnosis report is generated. Accurate identification and intelligent diagnosis of enterprise risks are realized, and decision-making efficiency is improved.

The invention discloses a multi-modal-based software architecture intelligent design and optimization system, which relates to the field of intelligent design and optimization, and comprises the steps of performing modal perception and preprocessing on architecture design demand information input by a user, converting the architecture design demand information into structured semantic data, and optimizing and enhancing semantic expression by introducing a reinforcement learning strategy, forming a structured user intention vector set; through a semantic mapping and reasoning processing unit, user intention vectors in the user intention vector set are constructed into a semantic-component alignment graph, and graph structure modeling and confrontation generation algorithm combination are adopted to generate a candidate structure graph set; a structure diagram generation step in the structure generation module effectively improves the rationality, diversity and adaptation capability of an automatically generated structure, and provides core support for realizing automatic construction of a target-demand-oriented architecture structure.

The invention discloses a phase residual error-based shallow defect detection method and device and a storage medium, which are used for improving the detection precision of shallow defects. Obtaining a reflection fringe pattern of the display screen to be detected; performing phase recovery on the reflection fringe pattern to generate first phase recovery data; calculating and generating a first phase gradient amplitude diagram according to the first phase recovery data; constructing a local average background gradient at each pixel according to the first phase gradient magnitude image; generating a first gradient residual image according to the first local average background gradient image and the first phase gradient magnitude image; performing visual contrast enhancement processing on the first gradient residual image; self-adaptive threshold judgment is carried out on the first anomaly enhancement graph, and a first structure defect mask graph is generated; performing feature fusion on the first phase gradient amplitude image, the first anomaly enhancement image and the first structure defect mask image; and inputting the first feature fusion image into a target shallow layer defect identification model for shallow layer defect detection, and generating a shallow layer defect detection result.

The invention discloses an end-to-end fault diagnosis and identification method based on multi-modal fusion, and the method comprises the steps: 1), collecting a vibration signal and an acoustic signal, carrying out the preprocessing, and constructing a training sample set; 2) performing feature extraction to obtain a high-dimensional modal feature vector; 3) generating a sparse adjacency matrix through an end-to-end deep learning graph generation module, and establishing a graph generation structure relation; 4) constructing a multi-receptive field Chebyshev graph convolutional network, and extracting node-level features in a graph generation structure; 5) inputting the structure sensing features into a full-connection layer for mapping, and completing prediction and discrimination of a fault category to which an input sample belongs; performing model supervision training, and optimizing model parameters in an end-to-end mode; and 6) carrying out prediction output on the fault identification model on the test set, and carrying out quantitative evaluation on the fault identification result to obtain the fault identification device.The method belongs to the technical field of equipment operation state monitoring and fault diagnosis, and realizes accurate fault diagnosis of the rotating equipment.

In various examples, various types of sensor data from multiple ego-machines are used to infer lanes and / or generate lane graphs for use in autonomous systems and applications. In some embodiments, one or more DNNs may be used to infer lane data indicating a representation of a lane shape using sensor data from various vehicles to represent a 3D environment. The inferred lane data may include cross-section indicators that indicate cross-sections of a lane and / or connection indicators that indicate a lane channel connecting two locations (e.g., two lane portions). The inferred lane data may be used to generate a lane graph that represents lanes on a road and, in some cases, lane dividers (e.g., polyline represented as a solid line, a dashed line, a double line, etc.). A lane graph may be used, for example, to model the environment around a vehicle, facilitate localization, provide guidance for autonomous driving, etc.

The invention discloses a lightweight incremental question answering system based on graph structure index and double-layer retrieval. Comprises: a graph-based incremental index module for segmenting an input document into text blocks, extracting entities and relationships through a large language model (LLM), and constructing a dynamic knowledge graph; the double-level retrieval module is used for mapping user query into a graph structure, adopting a double-strategy retrieval mechanism associated with local keyword matching and global theme and combining graph vectors for collaborative query; and the path constraint sub-graph generation module is used for generating an identifier sub-graph based on an n-hop reasoning path on the basis of the retrieved content, and performing deep semantic extension and answer generation. During work, through the modular and progressive collaborative design, the system realizes full-process optimization from data acquisition to knowledge representation and from query response to deep reasoning.

The invention relates to the technical field of graph analysis, in particular to an intelligent counter semantic interaction system based on a knowledge graph, which comprises a subgraph generation and matrix analysis module, a semantic analysis module and a node mapping module, and the dynamic adjustment factor dynamically adjusts the activation priority of the service node based on the historical jump path frequency, and generates a service flow time sequence priority. According to the method, the dynamic regulation factors and the real-time session features are coupled, so that triple constraint collaborative optimization of association degree scoring, time sequence compliance and resource scheduling is innovatively realized; dynamically cutting the knowledge graph based on the session state vector, and eliminating redundant node interference; the service flow time sequence matrix is combined with historical path attenuation statistics and real-time semantic features, and it is ensured that the node activation priority accurately adapts to service scene changes.

The invention relates to the technical field of deep learning, in particular to a stress distribution monitoring and crack risk prediction method for a silicon carbide part, which realizes comprehensive sensing of the stress state of the silicon carbide part, accurate positioning of a risk area and advanced early warning of a crack fault. The method comprises the following steps: synchronously acquiring multi-modal data through multiple types of sensors, and realizing cross-modal time sequence synchronization through feature alignment; designing a crack risk multi-branch feature extraction module, and respectively extracting general depth features and risk features oriented to thermal stress mismatch, microcrack evolution and structural instability through a shared backbone network and a special branch network; constructing a stress nephogram generation and risk area positioning module based on a graph neural network, and realizing visual reasoning and risk area marking from discrete features to full-field stress distribution; and designing a crack risk comprehensive prediction module based on multi-dimensional risk feature fusion, fusing an instantaneous state and an evolution trend, outputting a multi-risk confidence vector and triggering graded early warning.

The invention belongs to the technical field of network security, and particularly relates to an intelligent association and global situation risk monitoring method based on multi-source data, which comprises the following steps: acquiring a multi-source heterogeneous data set; the multi-source heterogeneous data set is preprocessed, and preprocessed multi-source data is obtained; obtaining an attack behavior association graph according to the multi-source data, and performing anomaly detection on the association graph by using a graph neural network to obtain an explicit attack link and a potential attack link; obtaining a network security situation dynamic graph based on the explicit attack link and the potential attack link; and generating a risk assessment report according to the network security situation dynamic graph, triggering a corresponding security policy, and performing risk monitoring according to the security policy. According to the method, the accuracy and response speed of threat detection are remarkably improved, the global network security situation awareness capability is enhanced, and an intelligent solution is provided for security protection in a complex network environment.

The invention provides a clinical thinking examination question generation and step-by-step analysis construction method and system based on a large language model, and relates to the technical field of large language models. The method comprises the steps of analyzing a teaching outline and real questions over the years, constructing a multi-level clinical knowledge graph, intercepting a knowledge sub-graph according to target difficulty, and generating case question stems, candidate options and a preliminary reasoning path covering the sub-graph; the reasoning path consistency is verified through a logic engine, a simulation answer sample is constructed, item reaction modeling is executed, and the question difficulty is estimated; if the difficulty does not accord with the target difficulty, parameters are automatically fine-tuned and re-generated, and iteration is carried out until the standard is reached; performing error selection rate driven optimization on the interference options to form a final question version; and collecting student answering data feedback atlas weight and difficulty, and writing the content into the versioned question bank. According to the invention, the accuracy of examination question generation, the teaching suitability and the continuous updating ability are improved, and the intelligent evaluation of clinical thinking ability is realized.

The invention discloses a structured data self-learning method based on a graph neural network, and the method comprises the following steps: S1, analyzing structured data, extracting entity fields and relation fields, and constructing a structure candidate graph; s2, generating a node embedding feature matrix, and initializing and recording the adjacency relation of candidate edges; s3, constructing a graph neural network model, inputting node features and an adjacent matrix, and defining a task loss function; s4, evaluating the gradient contribution degree of edge connection by adopting a gradient sensitive sparse adjacency self-learning algorithm, and updating the graph structure representation; s5, introducing an embedded interpretability gradient backtracking mechanism, correcting an edge connection relation and enhancing interpretability; s6, training the graph neural network by using the corrected structure, and updating the node embedding and graph structure; and S7, outputting a final graph structure and an interpretability index, and generating a graph modeling visualization result. According to the method, efficient modeling and explanatory analysis of structured data are realized through a dynamic graph structure learning and gradient backtracking mechanism.

The invention relates to an LLM-based industrial control programming ladder diagram generation method, which is based on a ladder diagram in a text graph form generated by processing a natural language target programming demand through a large language model, generates a large language model by applying a pre-training base, generates a corresponding target XML file, converts the file into a corresponding target JSON file, and finally analyzes the file by applying a GOJS graphic component. Generating a graph componentized target programming ladder diagram; compared with the prior art, the generation rate and accuracy of the graphic componentization programming ladder diagram are greatly improved, the natural language serves as input, the technical threshold and difficulty of ladder diagram programming are reduced, the readability of the generated programming ladder diagram is high, and the corresponding XML file and JSON file can be directly imported into an industrial control programming platform to be compiled and debugged.

The invention discloses an intelligent data management system, and relates to the field of intelligent data management. The system comprises a data acquisition module for acquiring and preprocessing multi-source data and extracting field information; the decoupling analysis module is used for calculating a mutual information index and an information cementation degree according to the field information; the map generation module is used for constructing an attribute fusion map according to the field information and generating risk mapping skewness and a risk control map; the behavior analysis module is used for collecting action execution logs, extracting field behavior sequences and calculating field behavior interference factors; and the quality evaluation module is used for extracting a key field missing proportion and information entropy based on the field information to calculate an information completeness index, and fusing multiple indexes to generate a comprehensive quality score. Through graph structure construction, behavior sequence analysis and quality scoring fusion, the multi-source data field fusion accuracy and risk identification precision are improved, and intelligent assessment and compression abstract generation of high-quality data are realized.

The invention discloses a marketing strategy optimization management system based on six elements of order transaction, and belongs to the field of communication management systems, in terms of data acquisition, multi-source comprehensive information collection enables enterprises to perceive consumer demands in all directions and no longer blindly grope, a dynamic feature modeling unit fuses various kinds of data into six-dimensional feature vectors, and the six-dimensional feature vectors are integrated into a database; multiple factors of commodities, users and festivals are balanced and considered, a solid foundation is laid for a marketing strategy, a commodity-festival-user ternary association graph constructed by a festival graph generation engine enables commodities to be pushed in a targeted manner, a personalized strategy is generated by an intelligent decision module, the matching degree of the commodities and consumers is improved, and the marketing efficiency is improved. The overall continuous feedback optimization mechanism of the system enables the marketing strategy to be like a continuously evolved life entity, can better adapt to the market change, and improves the sales probability of commodities.

The invention relates to the technical field of data processing, and discloses a data processing method and application based on multi-source data fusion. The method comprises the following steps: acquiring multi-source time sequence data (including sensor, geographic space, user behavior data and the like) and space vector base map data of a target area; de-noising and normalizing the multi-source time series data, extracting an associated feature map through a heterogeneous model, generating a regional data object set based on base map adaptive segmentation, and extracting multi-dimensional features; optimizing a data object set through spatial topology verification, feature matching screening and weighted fusion; and generating statistics, space-time coupling analysis and a dynamic prediction result by utilizing optimized data, or generating a visual map supporting interaction by mapping a base map. The method improves the multi-source data fusion precision and spatial analysis capability, is suitable for multi-field cross-modal decision support, and can be used for scenes of smart cities, environment monitoring and the like.

The invention relates to a power equipment topological graph generation and query method based on a graph neural network, belongs to the technical field of data processing, and solves the problems that an existing power equipment topological graph is low in generation efficiency and long in response time. Comprising the following steps: updating a feature matrix and an adjacent matrix of equipment according to received new ledger data of the power equipment, and updating an equipment topological relation by using a graph convolutional neural network; according to a received query condition, obtaining the to-be-displayed device and the topological relation thereof from the updated device topological relation, identifying whether the query condition has a topological record, and if the query condition does not have the topological record or the difference between the to-be-displayed device and the topological relation thereof and the topological record exceeds a threshold value, displaying the to-be-displayed device and the topological relation thereof. If yes, generating coordinates of the to-be-displayed equipment by using the coordinate optimization model, and rendering to generate a topological graph; otherwise, generating a topological graph through rendering according to the topological record; the coordinate optimization model is constructed based on a graph attention network and introduces a multi-objective constrained loss function. And the query efficiency of the topological graph is improved.

A method for extracting data from a database for use in a well construction process includes receiving a question from a user. The question is in a well construction language. The method also includes determining context based upon the question. Determining the context includes retrieving key performance indicators (KPIs) based upon the question, and retrieving a plurality of tables from the database. The tables are retrieved based upon the question. The method also includes generating a prompt based upon the question and the context. The method also includes generating a structured query language (SQL) query based upon the prompt using a large language model (LLM). The method also includes running the SQL query against the tables in the database in an attempt to produce a new table. The method also includes performing a wellsite action in response to the new table.

The invention relates to the technical field of semantic label generation management, in particular to an enterprise-level customer data dynamic label management system which comprises an enterprise customer data storage unit, a knowledge fusion unit, a semantic analysis unit and a label generation application unit. The enterprise customer data storage unit obtains enterprise customer data and product related data and broadens data sources, the knowledge fusion unit carries out data standardization processing, entities and relationships are extracted by utilizing an advanced technology and are stored in a knowledge graph, and the semantic analysis unit selects samples by adopting stratified sampling and active learning, so that the data source is expanded. A plurality of technical means are combined to judge and process semantic drift, a deep learning model is optimized, a label generation application unit generates dynamic labels according to an optimization model, the dynamic labels are updated to an enterprise customer relationship management center in real time after clustering, screening and matching operation, real intentions of customers can be accurately obtained, accurate semantic labels can be generated, and the real intentions of the customers can be accurately analyzed. Enterprises are assisted to accurately serve customers, and the market competitiveness is improved.

The invention discloses an AI semantic understanding-based zero-code multi-terminal application automatic construction method, which comprises the following steps of: receiving a UI design draft image and layer metadata uploaded by a user, respectively extracting visual features and structural features through a double-branch feature extractor based on an AI semantic understanding technology, and establishing an AI semantic understanding model; a cross-modal attention module and a cavity space pyramid pooling module are combined to generate an enhanced feature graph, and a high-precision UI component mask and an interaction dependency graph are generated; according to target end equipment parameters, dynamic weights are generated through a multi-layer perceptron, layout constraints of the interaction dependency graph are adjusted, a layout target function is optimized, and component overlapping and visual unbalance are minimized; based on the predefined control library and the mapping rule, the UI component is mapped into the atomic control of the target platform, the interaction logic is converted into the event-action chain, and the target platform code is generated, and the automation degree and the cross-platform consistency of zero code development are improved through dynamic adaptation of the target end equipment and automatic code generation.

The invention provides a controllable text generation method, device and equipment, and the method comprises the steps: obtaining an input text, segmenting the input text, and obtaining a plurality of input words; inputting an input word into the trained vertical classifier, and outputting a category probability corresponding to the input word; constructing a word-level attribute graph according to all the input words and the category probability corresponding to each input word; processing the word-level attribute graph by utilizing a webpage ranking algorithm to obtain a target keyword; generating a semantic-level attribute graph according to the attribute graph generation prompt; determining a target combination prompt according to the input text, the original prompt, the target keyword, the context information and the semantic-level attribute graph; and inputting the target combination prompt into the large language model, processing the target combination prompt by the large language model, and outputting a controllable text. The word-level attribute graph and the semantic-level attribute graph are combined to guide the large model to better understand the input text, so that the high-quality response is generated, and the controllable generation of the vertical drive is realized.

The invention provides a vocational ability assessment system based on a skill map, and relates to the field of human resource technology and knowledge map application. The evaluation main system comprises a skill star map generation module, an occupational trajectory simulation module, a skill challenge module, a skill authentication chain module, a future skill navigation module, a skill collaborative ecological module, a skill evolutionary tree module, an occupational story generation module, a skill energy pool module, an occupational time capsule module and a control flow management module; according to the invention, the dynamic skill star map is constructed and the 5G network is combined to transmit data in real time, so that the occupational experience, learning records and test data of the user can be collected in time, and the node weight is dynamically adjusted through the semantic analysis algorithm and the weighted scoring algorithm; the problems that traditional manual evaluation and online testing are high in subjectivity and low in efficiency, and the capability cannot be accurately quantified are effectively solved.

The invention relates to the technical field of intelligent operation and maintenance, in particular to an artificial intelligence operation and maintenance decision support method and system for multi-source information fusion. The method comprises the following steps: acquiring multi-source operation and maintenance data to perform multi-dimensional feature extraction to obtain multi-dimensional operation and maintenance feature data; performing continuous spatial cross-modal embedding according to the multi-dimensional operation and maintenance feature data to obtain cross-modal embedded data; performing heterogeneous feature coupling graph generation on the cross-modal embedded data to obtain coupling graph data; performing heterogeneous space fusion coding according to the coupling graph data to obtain fusion coding data; performing expert knowledge driving graph embedding on the fusion coding data to obtain operation and maintenance fusion graph data; performing root cause positioning reasoning according to the operation and maintenance fusion graph data to obtain root cause positioning data; and performing operation and maintenance decision generation according to the root cause positioning data to obtain operation and maintenance decision data. Through multi-source information fusion and intelligent reasoning, the root cause positioning accuracy and intelligent operation and maintenance decision efficiency of the system can be effectively improved.

In various examples, various types of sensor data from multiple ego-machines are used to infer lanes and / or generate lane graphs for use in autonomous systems and applications. In some embodiments, one or more DNNs may be used to infer lane data indicating a representation of a lane shape using sensor data from various vehicles to represent a 3D environment. The inferred lane data may include cross-section indicators that indicate cross-sections of a lane and / or connection indicators that indicate a lane channel connecting two locations (e.g., two lane portions). The inferred lane data may be used to generate a lane graph that represents lanes on a road and, in some cases, lane dividers (e.g., polyline represented as a solid line, a dashed line, a double line, etc.). A lane graph may be used, for example, to model the environment around a vehicle, facilitate localization, provide guidance for autonomous driving, etc.

The invention relates to an emotion accompanying robot system based on multi-dimensional perception and an interaction method, and belongs to the technical field of intelligent robots. The system comprises a sensing module which captures micro-expression, voice, body temperature, touch and other multi-source data through a biological radar array, multispectral imaging and a touch sensing fabric; the decision-making module quantifies the emotion intensity by using an emotion state calculation engine, constructs a user personalized emotion file in combination with the dynamic knowledge graph, and generates a dynamic emotion graph; and the execution module realizes anthropomorphic emotion expression through bionic face driving, joint compliance control and thermal feedback. The interaction method comprises the steps of data capture, emotion file construction, dynamic graph generation, interactive execution and feedback adjustment, definition of an emotion intensity quantification formula, a graph edge weight model and the like. According to the method, physiological signals, environment situations and bionic expression are fused, multi-modal precise emotion perception and safe interaction are achieved, man-machine naturalness is improved, the method is suitable for elderly accompanying, psychological counseling and other scenes, and the emotion accompanying effect is enhanced.

The invention relates to the technical field of intelligent traffic, and discloses a semantic information-based vehicle spatial-temporal trajectory similarity calculation method and device, and the method comprises the steps: obtaining spatial-temporal trajectory data, dividing the spatial-temporal trajectory data with one hour as an interval, and constructing a Voronoi graph to generate a Thiessen polygon; performing multi-dimensional feature extraction on the Thiessen polygon region containing the trajectory to obtain spatial features, time features and POI semantic features of the region; constructing a space-time relation graph, and learning high-quality representation of nodes by adopting a space-time integrated random walk strategy with track perception and a GAT algorithm in sequence; constructing a model T-LSTM for enhancing the time expression ability to learn the final expression of the trajectory sequence; measuring similarities between nodes and between trajectories, and designing a graph-based double-level contrast loss function training model; according to the method, the capturing capability of the track representation on the functional semantic information can be improved.

The invention discloses a traffic flow prediction method based on a dynamic graph convolution circulation network, and belongs to the technical field of traffic information. The method comprises the following steps: acquiring historical traffic flow data, preprocessing the historical traffic flow data, and constructing a data set; an AGCRN model is improved, firstly, a dynamic filter of a self-attention mechanism is added to a graph generation part of an AGCRN to capture dynamic spatial features, and then a dynamic graph generation module is added to a gating recursive unit part to capture periodic time dependence. And finally, adding a residual error correction module into the AGCRN to carry out error feature extraction. Training the improved AGCRN model until a predetermined performance index is reached, and obtaining a traffic flow prediction model; and predicting the traffic flow of the traffic network by using the traffic flow prediction model. According to the method, the graph convolutional network is improved, so that the prediction accuracy of the traffic network flow is improved.