5093 results about "Information extraction" patented technology

The present invention discloses an interactive behavior understanding method for posture reconstruction based on skeleton and image features. The steps are as follows: constructing and preprocessing the data set, extracting skeleton and image features, fusing and reconstructing these features, and conducting experimental evaluation and validation. This method retains the purity of skeleton features for human behavior information extraction and uses image features to retain effective environmental information, complementing the model feature information. Skeleton features are extracted using a graph convolution network, enhancing the relevance of input skeleton point information for accurate feature extraction. Effective image features are quickly and accurately extracted through the Vision Transformer network combined with a multi-head attention mechanism.

An intelligent pulmonary nodule grading method and system based on multi-modality feature fusion. The method comprises: using clinical multi-modality data of a physical examination crowd to acquire an ROI and a VOI of a pulmonary nodule on the basis of a chest CT examination image and an examination report, designing a multi-task feature extraction network based on an attention mechanism, and acquiring radiomics features and deep-layer image features from the ROI and the VOI; designing a cross-modality feature fusion method based on graph representation learning, designing a multi-modality information extraction method, acquiring a specific feature representation and graph structure of each modality, and then performing feature representation and graph structure fusion; and providing a pulmonary nodule grading GCN model optimization and clinical verification method based on self-supervised learning, so as to implement fine grading of pulmonary nodule malignancy having only slight differences. A new way is provided for the design of a fine-grained classification algorithm; and in the aspect of solving a clinical problem, provided is a classification algorithm optimization method and clinical verification scheme in which generative self-supervised learning and comparative self-supervised learning are organically combined.

The invention discloses a multi-source remote sensing image semantic segmentation method based on Transform, Mama and a diffusion model, belongs to the technical field of electric digital data processing, graphic data reading and image data processing, and is used for extracting image information of a remote sensing image. Comprising the following steps: acquiring remote sensing images, constructing an image information extraction network based on the remote sensing images, and respectively inputting the remote sensing images in a training set, a verification set and a test set into the constructed remote sensing image information extraction network based on Transform, Mama and diffusion models. According to the method, the RGB features and the depth features are subjected to Transform feature extraction, and a cross Mama method is provided to perform cross enhancement on the RGB features and the depth features; according to the method, a row and column pooling Transform module is used for replacing a traditional Transform module, and detail texture information and large-scale structure information of a feature map in the vertical direction and the horizontal direction are extracted at the same time; according to the invention, a cross Mama attention feature fusion module is used to replace a traditional Mama module, and the modeling capability for complex scenes and diversified targets is enhanced.

The present invention provides a method and system for extracting and analyzing long text information and association based on a large model, which relates to the technical field of natural language processing, including determining sliding window parameters based on semantic density values ​​to divide long text into blocks, extracting feature vectors of text blocks and calculating semantic relevance, and constructing a hierarchical clustering feature matrix to merge text blocks to obtain semantic fragments. Afterwards, the semantic fragments are syntactically analyzed to construct a semantic dependency tree, and the semantic network is extracted and the entity relationship strength value is calculated to construct a semantic association graph. Finally, the node features are extracted using a pre-trained large model, the semantic representation is fused through a hierarchical attention mechanism, the features are enhanced by propagating an adaptive graph network, and candidate paths are generated through a hierarchical decision network, the optimal path is selected, and the associated path and probability distribution are output. The present invention can effectively extract key information from long texts and analyze the association relationship between entities, thereby improving the accuracy and efficiency of information extraction.

The invention provides a multi-agent collaborative question and answer enhancement method and system based on heterogeneous data knowledge, and belongs to the technical field of information management. The planning optimization intelligent agent carries out structured processing on the input complex natural language problem; a multi-modal retrieval mechanism on the heterogeneous knowledge source is constructed based on the problem disassembly and entity recognition result, and a multi-path recall agent passes through a semantic matching model of a double-tower structure; the abstract extraction agent performs semantic fusion and information extraction on the text segments and the knowledge graph sub-graphs output by the multi-path recall module; logic verification and quality evaluation are carried out on the answers generated by the abstract extraction module by the reflection iteration agent; and by setting a threshold mechanism and combining importance weights of the sub-questions, performing scoring and reflection optimization on the generated answers by utilizing a large language model LLM (Language Language Model). The method can effectively cope with cross-domain and multi-level complex question and answer tasks, and has good expandability and intelligent level.

The invention relates to a cross-modal knowledge reasoning method based on a multi-modal large model. In a cross-modal knowledge reasoning process, an existing model is usually limited by single-modal information extraction and shallow feature fusion, so that deep semantic association among data such as texts, images and videos is difficult to fully capture. In order to solve the problem, the invention provides a model for fusing multi-modal information such as texts, images, videos, documents and the like, and processing of multi-modal data is converted into unified feature extraction, interaction and deep reasoning tasks by fully utilizing a supervision fine tuning strategy, a self-adaptive attention mechanism and a cross-language processing technology. The model adopts a modular design, integrates multi-source data complementary analysis, spatial-temporal feature modeling and emotional semantic analysis, and realizes multi-modal collaborative interaction, dynamic scene understanding, long video key event analysis and man-machine co-emotional response. Through sufficient training, the multi-modal large model shows excellent logical reasoning ability and emotion understanding ability in a complex cognitive task, and a brand new solution is provided for efficient extraction, deep semantic analysis and intelligent response of cross-modal information.

The invention relates to the technical field of image processing, and particularly discloses a remote sensing image super-resolution reconstruction method and system based on a prior diffusion model, and the method comprises the steps: carrying out the shallow feature information extraction of a low-resolution image, and obtaining a shallow feature map; the method comprises the following steps: performing multi-scale down-sampling and convolution fusion processing on a shallow feature map and a high-resolution image, and performing high-frequency information extraction and adaptive high-frequency enhancement processing on the processed feature map to obtain condition information; according to the method, deep mining of high-frequency information is carried out on a low-resolution image input as condition information, so that the high-frequency information in the low-resolution image gradually approaches a high-resolution image, and the reconstruction precision is improved; the noise prediction network model guides step-by-step noise removal in the back diffusion process, and the detail reduction capability and precision of natural disaster remote sensing image super-resolution reconstruction are improved.

The invention relates to the technical field of intelligent auditing, in particular to a multi-modal document information processing method, device and equipment based on a large model agent and a storage medium. An input document and an auditing instruction are reasoned through a business information auditing model so as to determine a part to be extracted in document content and an auditing rule; and then calling a corresponding extraction tool to perform information extraction on the document, performing item-by-item check on the extracted content based on an auditing rule, and summarizing check results to obtain a final check result. According to the scheme, through the reasoning ability and the multi-modal data analysis ability of a large model, the verification instruction under the natural language can be understood and converted into the verification logic of the document, and meanwhile key information in the input document can be intelligently recognized and extracted based on the verification logic; according to the scheme, various document types including texts, tables, images and the like can be processed in a large-batch and multi-angle mode, the flexibility of auditing work is enhanced, the checking period is effectively shortened, and the auditing work efficiency is improved.

The invention discloses a point cloud semantic segmentation method and system based on local neighborhood attention, and belongs to the technical field of three-dimensional point cloud processing and deep learning, and the method comprises the following steps: S101, dividing three-dimensional point cloud data into training, verification and test sets, and carrying out the preprocessing; s102, constructing a deep learning network based on local neighborhood attention, training by using the training set, and obtaining an optimal model by using the verification set; and S103, inputting the test set into the optimal model to obtain a semantic segmentation result. According to the method, the semantic segmentation model is constructed by fusing local neighborhood attention of global position coding and local position coding, so that the problem of poor space geometric structure information extraction due to the fact that a conventional model only pays attention to the relative position of points in a neighborhood is effectively solved, and the feature extraction capability of an attention mechanism is enhanced; and the robustness, generalization and segmentation accuracy of the three-dimensional point cloud semantic segmentation model are improved.

The invention provides a knowledge graph construction and intelligent question and answer method and device based on deep learning. The method includes; obtaining target text data; information extraction is conducted on the target text data, entity relation data is generated, and the entity relation data comprises key data, entity data and relation data used for indicating the relation between entities; according to the entity relationship data, importing the target text data into a graph database to form a knowledge graph; obtaining an input question statement; the question sentence classification model is used for classifying the question sentences, question sentence features in the question sentences are determined, and the question sentence features comprise question sentence types and question sentence entity data; and retrieving text data matched with the question features in the knowledge graph as answer data of the question statements. According to the method, the public text information retrieval efficiency is improved, and the accuracy of intelligent question answering is improved.

The invention relates to the technical field of automatic report generation, and provides a test analysis and report generation method and system based on large model retrieval enhancement, and the method comprises the steps: firstly carrying out the semantic task decomposition of user query, and meanwhile, achieving the multi-dimensional information extraction through the butt joint of a knowledge vector library and a structured image-text knowledge system established by a private knowledge base module. And three strategies of RAG, Self-RAG and Graph-RAG are combined to enhance the generation capability of the large model so as to accurately obtain background knowledge. And constructing a standard SQL statement according to a structured query requirement, and performing data filling according to template prompt by an output result fusion module in combination with a query background and an SQL execution result to form a test report. By optimizing a retrieval enhancement generation method, professional knowledge supplement related to query is realized, and the reasoning ability of a large model in a professional scene is effectively enhanced, so that the accuracy and the intelligent level of data analysis are improved.

The invention belongs to the technical field of information, and relates to a multi-level dynamic intelligent automatic data processing and information evaluation method and system. The method comprises the steps that a database environment operation data set is constructed, targeted training and fine adjustment are conducted on a large language model in combination with reinforcement learning, and full-automatic data analysis, query and information extraction of multiple tables in a database are achieved; a multi-level agent interaction and cooperation framework is established, a scheduling agent is responsible for global decision making and strategy planning, an operation agent is responsible for execution of specific tasks, and the scheduling agent and the operation agent share information and coordinate the tasks to form a highly cooperative whole; and establishing a cyclic supervision and error correction mechanism, scheduling the intelligent agent to learn a normative workflow example of the target working environment and track the state of the operation intelligent agent in real time, identifying the behavior of the operation intelligent agent deviating from a predetermined path, and taking corrective measures. According to the method, high efficiency, intelligence and automation of data processing are realized, and the method has remarkable practical value and innovative significance.

The present invention provides a screenshot content extraction method based on a multimodal large model, which is used to extract OTA hotel structured information, including the following steps: obtaining a screenshot image with hotel information; performing text recognition on the image through an optical character recognition model to obtain a text information box set; performing text area positioning on the image through an image positioning model, and extracting the text content of the text area to obtain a text content set; inputting the text content set into a text large model, extracting structured information through preset prompt words, and obtaining first structured information; inputting the image into a visual language model, extracting structured information through prompt words, and obtaining second structured information; comparing and fusing the first structured information and the second structured information to obtain hotel structured information. The present invention can understand the image content more comprehensively and accurately, significantly improve the accuracy of hotel structured information extraction, and reduce the error rate of information extraction.

The invention relates to the technical field of intelligent vending, solves the problem that in the prior art, commodity identification cannot be accurately carried out in a multi-target scene, and provides a multi-target commodity identification method, device and system based on multi-modal data processing. The method comprises the following steps: acquiring multiple frames of real-time images in a commodity transaction scene; performing preprocessing and label information extraction on the real-time image, and determining character information corresponding to the target image and the commodity label; performing instance segmentation on the target image, and determining commodity position information; performing feature extraction on the target image, and determining commodity image feature information; according to pre-collected multi-source privatized data in an intelligent vending scene, performing fine adjustment and optimization processing on the open-source multi-modal visual language model to obtain a multi-modal large model; and inputting the commodity image feature information and the text information into the multi-modal large model for information fusion, and determining a commodity target identification result. According to the invention, commodity identification can be accurately carried out in a multi-target scene.

The invention provides a model pre-training method and device for a multi-language task, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a multi-modal training data set which comprises a plurality of pieces of sample text data and a plurality of pieces of sample image-text pair data with multi-language contents, and covers a universal field and a preset business field; based on a plurality of sample image-text pair data, performing comparative learning training of alignment of visual features and text features on the initial model, freezing model parameters of a decoding module and adjusting model parameters of a visual encoder and a projection module in the training process until a first ending condition is met; based on the multiple pieces of sample image-text pair data and the multiple pieces of sample text data, constraint training of content understanding is carried out on the initial model meeting a first ending condition, and model parameters of a visual encoder, a projection module and a decoding module are adjusted in the training process until a second ending condition is met to obtain a target model; the information extraction capability of the model in the specific field can be remarkably improved.

The invention provides a knowledge graph open domain construction and RAG question and answer method and device and a storage medium. The method comprises the steps of obtaining text content in a text file; performing text division on the text content to obtain text blocks; performing open domain information extraction on each text block by utilizing a large language model to obtain an initial triple; performing post-processing on the initial triad to obtain a post-processing triad for constructing the knowledge graph; constructing a knowledge graph according to the post-processing triple; the method for realizing RAG question answering based on the knowledge graph comprises the following steps: performing knowledge graph retrieval on a question to obtain a context related to the question, then constructing a question answering cue word by using the context and the question, and sorting out an answer to the question by a large language model based on the question answering cue word. According to the method, the problem that a complete knowledge graph cannot be obtained during defined domain knowledge graph construction in the face of diversified entities and relation types is solved, and the problems that RAG retrieval information is weak in correlation with problems and the multi-hop reasoning capability is limited are solved.

The invention discloses a thermal imaging temperature rise trend early warning system based on time-space sequence prediction, and particularly relates to the technical field of thermal imaging data prediction and early warning. The thermal imaging temperature rise trend early warning system comprises an image conversion module, a fluctuation feature extraction module, an edge prediction module, an anomaly characterization module and a prediction decision module; a temperature dynamic change rate and gradient intensity are calculated, edge model prediction is carried out based on a fluctuation index combination, when the fluctuation index combination does not exceed a stable interval, a lightweight deep network model deployed at a thermal imaging acquisition end is called, and when the fluctuation index combination exceeds the stable interval, a prediction decision module determines whether to switch to a high-order multi-modal model; the space-time information extraction capability is improved by constructing the temperature evolution data body, the prediction path is dynamically controlled based on the fluctuation index combination, and the prediction stability and efficiency are improved; and the abnormal activation index and the prediction offset index are combined to realize adaptive switching of model calling, so that the accuracy and adaptability of the early warning system are enhanced.

The invention discloses an enterprise audit information extraction method based on artificial intelligence, and relates to the technical field of enterprise audit information processing, and the method comprises the steps: obtaining an audit original document from an enterprise, carrying out the multi-mode conversion processing, and generating a standard data set; performing deep semantic analysis on the standard data set, and constructing an audit knowledge graph; performing cross-modal alignment on the auditing knowledge graph, the image features, the table features and the time sequence features, and constructing a joint auditing analysis model; performing dynamic verification on an audit analysis result and a historical case library to form a dynamic evolution knowledge base; and generating a final audit document based on the updated knowledge base. Through cross-modal feature vector alignment and attention-driven clause applicability scoring, dynamic association analysis of financial anomaly and business pipeline fluctuation is realized, and the cross-modal clue discovery efficiency is improved.

The invention relates to the technical field of archive information management, and provides an archive information extraction management method and system based on multi-modal learning, and the method comprises the steps: collecting multi-modal archive data, carrying out the preprocessing of the multi-modal archive data, and obtaining the processed multi-modal archive data; performing feature extraction on the multi-modal archive data, and obtaining an archive comprehensive feature vector through weight distribution and feature alignment; based on a preset archive knowledge graph, performing multi-modal entity recognition on the archive comprehensive feature vector, constructing an archive relation graph, and performing iterative calculation through a graph neural network to obtain an archive association rule; updating a preset archive knowledge graph according to the archive association rule to obtain an archive enhanced knowledge graph, and performing iterative optimization on the archive association rule to form and store an archive management strategy network; and constructing an archive information index system based on the archive enhanced knowledge graph and the archive management strategy network. According to the invention, efficient management and accurate retrieval of archive information are realized.

The invention relates to the field of computer vision, and discloses a photovoltaic panel defect category detection algorithm based on feature pyramid and cascade group attention, comprising the following steps: S1, data acquisition, labeling and preprocessing; s2, carrying out model training by using an improved YOLOv11 algorithm; and S3, predicting the input infrared picture, accurately positioning the defect and marking the category. According to the photovoltaic panel defect category detection algorithm based on the feature pyramid and the cascading group attention, the feature pyramid and the cascading group attention can be detected according to the internal circuit structure of the photovoltaic panel and an infrared imaging picture; accurately positioning, marking and classifying the defects on the photovoltaic panel; a newest YOLOv11 target detection algorithm is adopted and improved, and part of C3k2 modules are replaced by SCC3k2 modules, so that information capture on space and channels is more accurate, and adverse effects caused by redundant information extraction are reduced; and meanwhile, different input segmentation is provided for each attention head in the attention module, and the diversity of the characteristics input to the attention heads is enhanced by cross-head cascade output characteristics.

The invention provides a semi-structured file processing method based on an LLMs (Language Language Model), which is used for efficiently analyzing a semi-structured file and converting the semi-structured file, and belongs to the technical field of intelligent document processing and multi-modal learning. The method is characterized by comprising the following steps: multi-modal fusion analysis: adopting a hierarchical cross-modal attention mechanism HCMA, and combining a multi-head attention mechanism with a graph neural network GNN to realize multi-level feature fusion of a text and an image; adaptive structured conversion: adopting a dynamic structure analysis network DPN, and combining reinforcement learning and a generation model to realize adaptive analysis of a document structure; and model generation and multi-modal reasoning: adopting a multi-modal generation reasoning network MGRN, and combining a diffusion model and neural symbol reasoning to realize joint generation and reasoning of the text and the image. According to the method, a remarkable technical breakthrough is brought to the aspects of document analysis, information extraction and intelligent reasoning, and powerful support is provided for data-driven innovation of various industries.

The invention discloses a combinatorial optimization automatic modeling-oriented training data generation method and system, and the method comprises the steps: receiving a problem description text, extracting key information through semantic analysis and knowledge reasoning, and constructing a structured optimization problem representation; reading the structured representation, extracting a target function component from the binary label matrix, extracting a constraint function component from the constraint relation adjacent matrix, extracting a variable definition component from the variable dependency matrix, and integrating the components to form a model component set; a value interval parameter is generated for the variable, a coefficient parameter is generated for the constraint matrix, a weight parameter is generated for the target function, and a parameterized model instance is obtained; and applying corresponding templates to scene elements, variables and constraint descriptions to generate a problem description text. Through the systematic information extraction and parameter generation method, automatic generation of high-quality training data is realized, and the diversity and practicability of the data are improved.

The invention discloses a multi-modal rehabilitation data intelligent evaluation method and system based on a large model, and relates to the field related to data processing, and the method comprises the steps: carrying out the data collection based on a plurality of data sources, obtaining and integrating multi-modal data collection results, obtaining a multi-modal data set, synchronizing the multi-modal data set to an information extraction model, and carrying out the retrieval, a rehabilitation training scheme is executed to track the rehabilitation progress of a target user in real time; rehabilitation effect data is obtained and synchronized to the rehabilitation evaluation large model; a multi-dimensional evaluation result is obtained to perform feedback optimization on the rehabilitation training scheme; and a rehabilitation optimization training scheme is determined. The technical problem that traditional rehabilitation evaluation depends on professionals and lacks enough professional knowledge and experience, so that the accuracy and efficiency of evaluation are affected is solved, and the technical effects of performing intelligent personalized evaluation on rehabilitation training and improving the precision and efficiency of rehabilitation evaluation are achieved.

The invention discloses an engineering drawing label identification method and system based on multi-modal information extraction, and the method comprises the steps: obtaining an engineering drawing image, carrying out the preprocessing of the image, and setting a data structure Schema for controlling an extraction target and a relation mode; detecting a drawing label area in the drawing through the trained drawing label detection model to obtain a bounding box coordinate of each drawing label, mapping the detected bounding box coordinates back to a coordinate system of the original image, and cutting out a corresponding drawing label image from the original image; performing text recognition on the cut-out signature image, and extracting text content and corresponding textbox coordinate information; and inputting the clipped signature image and the text recognition result into a trained multi-modal information extraction model, performing information extraction according to a set Schema, and outputting a structured extraction result. According to the method, the extraction target can be flexibly defined, unified extraction of the two-tuples and the three-tuples is supported, and the extraction accuracy is high.

The invention belongs to the field of artificial intelligence, and particularly relates to a bill information extraction method and system and a storage medium, and the method comprises the following steps: obtaining an electronic file containing a bill, and preprocessing the electronic file into an image file facilitating information extraction; extracting bill information in the image file by using a multi-modal large model and an OCR model, and outputting text information in a structured format; and analyzing, correcting and comparing text information respectively output by the multi-modal large model and the OCR model through the large language model, and performing fine tuning or cue word engineering guidance on the large language model for different tasks, so that an output result conforms to expectation. Compared with the prior art, the problems that in the prior art, OCR recognition accuracy is not high, and extracted information is incomplete are solved, the bill information processing speed is increased, the precision of the automatic processing process is greatly improved, and in financial, auditing and other related business scenes, the working efficiency and the data reliability are remarkably improved.

The invention relates to the technical field of bearing service life prediction, in particular to a bearing residual service life prediction method based on an improved Attention-GRU model, and the method comprises the steps: monitoring full-life vibration data of a bearing when the bearing runs until the bearing is damaged, carrying out the preprocessing of the data, and dividing the data into a training set and a test set; time-frequency domain features are extracted from the vibration data and feature fusion is carried out to obtain a fused feature set; and constructing a bearing residual life prediction model. And inputting the fused feature set into a life prediction model for training and prediction to obtain the remaining service life of the bearing. According to the residual life prediction model in the scheme, time-frequency domain information extraction is carried out on bearing data through discrete wavelet down-sampling and Fourier frequency domain feature extraction, and then the receptive field of the bearing data in the channel and space is improved through channel and space attention feature enhancement; and finally, predicting and outputting the health index of the predicted time step through Attention-GRU regression, and combining the health index with the full-life data to obtain the predicted residual life.

The invention discloses an aviation converter fault detection method based on an information fusion algorithm, and the method comprises the steps: collecting the operation data of an aviation converter, and constructing a data set; performing time information extraction and space information extraction on the data in the data set, and performing feature fusion on time features and space features based on a variational auto-encoder to obtain spatio-temporal data; constructing a Bayesian deep learning model, and training and optimizing the Bayesian deep learning model; inputting the spatio-temporal data into the trained and optimized Bayesian deep learning model for processing, outputting a numerical value representing a fault occurrence probability, and taking a variance based on the probability numerical value as uncertainty estimation; presetting a probability threshold, and when the output probability is greater than the threshold, determining that the converter has a fault; otherwise, judging that the operation state is normal; a more accurate and reliable diagnosis result is provided for fault detection, and the method has high adaptability and robustness in practical application.

The invention provides a power equipment fault prediction method and system. The method comprises the steps of performing multi-modal fusion on acquired historical data of power equipment to obtain feature vectors of all modals; fusing the feature vectors by adopting principal component analysis to obtain fusion vectors; constructing the fusion vector into a graph to obtain a space-time diagram; learning spatial features of the spatial information extraction data by using a space-time diagram in a graph convolution mode to obtain a spatial feature representation vector; fusing the spatial feature representation vector in combination with the attention weight to generate an attention fusion vector; learning time sequence information by using a gating loop unit to obtain a time characteristic representation vector; performing weighted feature fusion on the spatial feature representation vector and the time feature representation vector to obtain a space-time fusion representation vector; and processing the space-time fusion representation vector by using a multi-layer perceptron, and outputting power equipment fault prediction. Based on the method, the invention further provides a power equipment fault prediction system. According to the invention, the accuracy and robustness of power equipment fault prediction are improved.

The invention discloses an intelligent bidding and tendering information identification method, which relates to the technical field of engineering management, and comprises the following steps: standardizing and preprocessing a bidding and tendering information document format; constructing a natural language processing model special for the bidding and tendering field; training a multi-level information extraction model based on deep learning; carrying out cross-document information integration and correlation analysis; performing anomaly detection and verification based on rules; semantic understanding and sentiment analysis; automatic subject matter description and technical specification extraction; a real-time dynamic monitoring and information updating mechanism; the invention discloses a graph-based decision support and recommendation system. According to the intelligent bidding and tendering information identification method, efficient and intelligent identification of bidding and tendering information is realized by constructing a natural language processing model specially aiming at the field of bidding and tendering. The bidirectional coding capability of BERT and the sequence modeling capability of LTM are combined, and a hierarchical attention mechanism is introduced, so that the model can accurately identify various key information in bidding and tendering documents.

The invention relates to a government affair file information extraction and question answering method and device and a medium, and the method comprises the steps: carrying out the entity extraction of a government affair file through employing a BERT-CRF joint model, and obtaining a structured entity set; performing relation extraction on the structured entity set to generate a semantic relation set between the entities; constructing a knowledge graph according to the structured entity set and the semantic relationship set, storing entity nodes into a graph database, and storing an embedded vector of an entity text into a vector database; when a query request of a user is received, relation query of the graph database and semantic retrieval of the vector database are carried out, sub-graph structures and semantic matching vectors related to query are extracted, and a mixed retrieval result is obtained; and inputting the mixed retrieval result into a large language model, and generating a question and answer response text conforming to a preset format by applying a dynamic prompt template. According to the method, the document processing efficiency and accuracy are effectively improved, and a solid technical support is provided for intelligent management of government affair documents.