53 results about "Historical document" patented technology

The invention relates to a historical image-text-based three-dimensional ancient city model construction method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining historical image-text data of a target city, the historical image-text data comprises related historical literatures of the target city in a specified period, an electronic ancient map corresponding to an ancient map drawn in the specified period, an electronic old map corresponding to an old map surveyed and mapped in the latest period from the specified period, and a shot satellite image; based on the historical image-text data, generating a city plane restored map of the target city in a specified period; extracting semantic information of the target city in a specified period from at least one kind of data in the historical image-text data, wherein the semantic information comprises city information of the target city in the specified period and attribute information of constituent elements; and generating a three-dimensional ancient city model of the target city in a specified period based on the semantic information city plane restoration map. Therefore, a high-precision and high-integrity three-dimensional ancient city model of the target city in the specified period can be constructed.

The invention discloses a historical document restoration method and system based on implicit interpolation network enhancement. The method comprises the following steps: acquiring a damaged historical document image, a position mask image for identifying a to-be-restored area and a content guide image for guiding the semantic meaning of restored content; constructing a historical document repair model based on a de-noising diffusion probability model; training the historical document repair model by using a training data set containing damaged image and intact image pairs; and inputting the damaged historical document image to be repaired, the position mask image and the content guide image into the historical document repair model, and outputting the repaired historical document image. According to the method, the accuracy of the character content in the restoration area and the naturalness of the visual boundary are remarkably improved, and high-fidelity digital restoration of complex degraded literatures such as ancient books and inscription rubbings is realized.

The invention discloses a knowledge pushing method and system based on a knowledge graph and intention prediction, a terminal and a storage medium. The method comprises the steps that historical document data and new document data of multiple systems are crawled regularly and preprocessed; a triple in the preprocessed data is recognized and stored in a knowledge graph database, updating is suspended and manual auditing is triggered when new and old data conflicts are found, and updating is executed and changes are recorded after auditing is passed; collecting user behavior data, generating a behavior sequence, combining post information of the user, and utilizing the intention prediction model to output a demand probability; extracting associated knowledge according to the demand probability, and sending a knowledge packet link and a brief description to the user through an agent pop-up window; and receiving and storing feedback information of the user on the pushed knowledge packet, and performing system optimization based on the feedback information. According to the method, multi-source heterogeneous knowledge can be automatically integrated, knowledge updating can be monitored and synchronized in real time, knowledge is actively perceived and pushed based on context, and the utilization efficiency of enterprise knowledge is improved.

The invention relates to the field of artificial intelligence, in particular to a long text generation method and system. The method comprises the steps of generating a structured template based on user input or historical documents, generating a structured outline conforming to logical coherence and structural constraints based on a user demand and an adaptation result of the structured template, retrieving multi-source knowledge based on the structured outline and fusing the multi-source knowledge, generating text content conforming to the constraints of the structured outline, and displaying the text content according to the structured outline. And performing evaluation based on the theme coherence, the semantic coherence and the logic coherence of the text content, adjusting a generation strategy through reinforcement learning, and outputting a final long text. And the content continuity is kept, and meanwhile, the accurate control on the generation process is realized.

The invention relates to the technical field of document automatic processing, in particular to a localized document automatic processing method and device, equipment and a storage medium. The localized document automatic processing method comprises the following steps: acquiring a current document and a historical document associated with the current document; extracting text features of the current document and the historical document; constructing a multi-document context sensing model based on the text features of the current document and the historical document, and obtaining a document context vector containing complete context information; a document summary is generated based on the document context vector, and / or a structured task is extracted based on the document context vector. According to the method, the current working context of the user can be fully combined, the accuracy of automatically processing the document is improved, and therefore the document abstract can be automatically and accurately formed and the structured task can be accurately extracted.

A question generation model training method, includes: obtaining a historical complex question, an answer corresponding to the historical complex question, and a historical document-based knowledge base corresponding to the historical complex question, wherein the historical document-based knowledge base includes at least one historical document related to the historical complex question; extracting an entity term related to the historical complex question from the historical document, constructing a simple question based on the entity term, and determining a plurality of historical knowledge points based on the simple question and the historical complex question; and training a question generation model based on the historical document and the plurality of historical knowledge points by using a preset loss function, to obtain a trained question generation model, wherein the question generation model is used to generate a historical complex question corresponding to the historical document.

Systems, methods, and computer-readable media are provided for using generative AI enriched with metadata about historical document characteristics to transform documents of various formats, including images, to the fields and values they represent. A prompt template may be selected in association with a type of document. The prompt template indicates field definition(s) of field(s) to be detected in the document and location(s) in which the field(s) have been detected in prior documents. A large language model is prompted with a prompt generated using the prompt template to generate a result that assigns value(s) to the field(s). Output from the language model is used for identifying the field to value mapping for the document, such that data detected from the document may be stored in appropriate database structures of a database. Metadata stored in association with the prompt template is updated based on location(s) in the document in which the field(s) were detected, and the value(s) of the field(s) are stored in a database. Outbound documents may be similarly translated to detect values of corresponding fields requested by third parties, even if those values are not stored in the database. In this scenario, values for fields may be detected in outbound documents using the prompt templates enriched with metadata as processed by the large language model before such information is prepared to be sent to a third party.

The invention relates to the field of document processing, and particularly provides a Word document display method and system based on structured semantic parse, a terminal and a medium. Firstly, a dynamic rule priority queue arranged in a descending order according to the number of failures is constructed by analyzing historical document samples; guiding the analysis engine to cooperate with the rule and the machine learning model to perform high-precision structured analysis on the document; constructing a document structure tree rich in semantic association, and extracting entity relationships to generate knowledge graph sub-graphs; and finally, the serialized data increment is transmitted to a front end, and catalog jump, reference tracking, term prompting and graph visualization are realized through componentization rendering. According to the method, intelligent interaction of the Word document is realized through a self-adaptive analysis strategy and deep semantic modeling, and the information acquisition efficiency is improved.

The invention provides a document knowledge retrieval method and system based on a two-way quantity library hybrid architecture, relates to the technical field of crossing of artificial intelligence and knowledge engineering, and solves the technical problems that a traditional RAG knowledge retrieval system is delayed in response to high-frequency repeated problems, high in cost and poor in answer consistency due to indifference processing. The method comprises the steps of obtaining a user query request and historical document data; constructing an FAQ vector library and a document vector library based on the historical document data; calculating the matching similarity between the user query request and the FAQ vector library through a similarity algorithm to obtain an FAQ similarity score; comparing the FAQ similarity score with a preset similarity threshold value; when the FAQ similarity score is greater than or equal to a similarity threshold value, marking a standard answer corresponding to the FAQ similarity score as a query result; when the FAQ similarity score is smaller than the similarity threshold value, the document vector library is retrieved, and a big language model is called for TOP-M related fragments to generate answers to serve as query results. The method and device are used in the knowledge retrieval process.

The application discloses a historical document version knowledge ontology dynamic collaborative construction method and system, relates to the technical field of humanistic knowledge graph, and comprises the following steps: inputting a version feature set into a multi-node cloud collaborative processing architecture, calculating local similarity scores between versions, and outputting a version similarity matrix; utilizing a distributed conflict detection mechanism to evaluate the consistency of the version similarity matrix, marking entity mapping pairs with a confidence level lower than a confidence threshold, generating a conflict transaction log; inputting the conflict transaction log into a five-level contradiction hierarchical processing protocol, outputting arbitration results, converting the arbitration results into corresponding OWL ontology update statements, adding version trace annotations, and outputting historical document update instructions; and according to the historical document update instructions, dynamically expanding an entity relationship network of a historical document knowledge ontology, and outputting a historical document version knowledge ontology through a three-layer intelligent analysis framework. The application significantly enhances the automation and reliability of multi-version document knowledge management.

The embodiment of the invention discloses a document analysis method and device based on a knowledge high-dimensional vector, equipment and a medium, relates to the technical field of information retrieval, and is used for solving the problems of low retrieval precision and low efficiency in the prior art. The method comprises the following steps: processing a multi-format document based on a preset analysis engine and a preset dynamic slicing strategy to obtain multi-format document fragments; performing vector coding on the multi-format document fragments to obtain current semantic vectors and current position vectors of the multi-format document fragments; screening a preset vector database according to the basic structure features corresponding to the multi-format document, and determining an analysis range corresponding to the multi-format document; performing similarity judgment on the current semantic vector and the current position vector with the slice semantic vector and the slice position vector of each historical document in the analysis range to obtain an originality analysis result of the multi-format document; and processing the originality analysis result based on a preset analysis model to obtain a structured analysis result of the multi-format document.

The invention relates to the technical field of artificial intelligence, and discloses a national region traditional settlement and building knowledge base construction method based on a large language model and a knowledge graph technology and a question answering system. The method comprises the following steps: converting building surveying and mapping data into semantic ontology data, splitting the structured data into a plurality of entity tables and relation tables, and importing the entity tables and the relation tables into a graph database; unstructured data such as interview records and historical literatures are sorted into an entity and relation structure, and the entity and relation structure is imported into a graph database; extracting an entity type, and creating an AI prompt word; constructing a query module, and converting a question proposed by a user by using a natural language into a graph database query statement as a first output item through a large language model; the query result serves as a second output item; and converting the second output item into a natural language by using a large language model, and returning the natural language to the user as a third output item in an agreed format. According to the system, the settlement and building research data query efficiency is improved, multi-level data is associated, and the knowledge reasoning ability is provided.

The invention relates to the technical field of building heritage protection, and provides a railway building heritage protection method based on multi-source data fusion in order to solve the problems of data splitting, significant subjective deviation and decision staticization in the building heritage protection process. The method comprises the following steps: constructing a multi-dimensional digital file by comprehensively collecting a BIM model, an Internet of Things sensor network, a historical literature database and a geographic information system; establishing a hierarchical evaluation index system, setting an objective reference value, and determining the weight of a corresponding index by adopting a collaborative process integrating an analytic hierarchy process and a Delphi method; carrying out quantitative scoring through an expert team, and carrying out weighted calculation on a total score of a comprehensive value to determine a protection grade; and performing section division according to the spatial position, identifying regional dominant value features through clustering analysis, and generating a hierarchical protection strategy in combination with a protection rule base. According to the method, normal form conversion from subjective qualitative to objective quantitative in the evaluation process is realized, subjective deviation is remarkably reduced, and the accuracy and operability of protection decision making are improved.

The invention provides a traction converter control software design method fused with a large language model, and the method comprises the steps: analyzing a historical document through a large language model subjected to the fine adjustment of traction converter field data, generating a structured demand, and building MATLAB demand tracing; an initial model is built in Simulink according to requirements, and multi-target iterative optimization is completed through model recommendation parameters; normal, extreme and fault working condition scenes are generated through the large language model, MATLAB batch simulation is driven, and a semantic analysis result is given; codes are automatically generated through an Embedded Coder, readability and real-time optimization is implemented through a large language model, and a final code is output after PIL verification; codes are downloaded to a real control board card, a closed loop is formed in combination with a virtual traction system built by a semi-physical simulation platform, a test case is automatically generated by a large language model, software and hardware collaboration data are analyzed, and iteration is carried out until the performance reaches the standard. According to the method, full-process AI driving including requirements, modeling, simulation, codes and testing is achieved, the development period is remarkably shortened, and the software reliability is improved.

A machine-learning based (ML-based) system and method for automatically extracting one or more data fields from one or more documents, are disclosed. The ML-based system includes a document obtaining subsystem to obtain documents, a document pre-processing subsystem to generate pre-processed data, a field identifying subsystem to identify data fields using a trained ML model, and a field extracting subsystem to extract financial information. The ML-based system also comprises an output subsystem to deliver the extracted data to end users via user interfaces. The ML model is trained using historical documents, labelled data fields, and features such as distance-based features, direction-based features, dimension-based features, positional features, and value-based features. The M-based system employs hyperparameter optimization, noise removal, and accuracy assessment mechanisms to enhance performance. This ML-based system provides a scalable, accurate, and automated solution for financial information extraction, ensuring efficiency, adaptability, and seamless integration with enterprise systems.

The invention relates to the technical field of power grid infrastructure projects. The invention discloses a structured design method and system for a feasibility research report of a power grid infrastructure project. The method comprises the steps of obtaining file data specified by related management of power grid infrastructure in a historical file library, and obtaining a structured text data set; performing structured information extraction on the text data set to obtain a structural framework set of the text; key review element extraction of the feasibility research content review rule is carried out on the text data set, and a key review element set is obtained; establishing logic association for each key review element, each node and each text in the text data set to obtain a mapping relation M of the key review element ID, the node ID and the text ID; according to the mapping relation M, a feasibility research report structure chart is formed, and a power grid infrastructure project feasibility research report template is created. The produced standardized template provides a solid and uniform data basis for treatment, integration and deep analysis of power grid data, thereby directly supporting scientificity and accuracy of investment decision.

The application relates to a low-cost entity labeling method and system based on user behavior analysis, which comprises the following steps: S1, data collection: using a state machine to provide a document layout service, collecting user historical documents, entity recognition results and user revision records; S2, data labeling: generating a labeling data set according to the entity recognition results, finding out suspicious incorrect labeling by using the user revision records and reconfirming to optimize the labeling data set; S3, model updating: training an NER model by using the labeling data set, and replacing the state machine in the step S1 with the NER model when the accuracy of the NER model exceeds that of the state machine. The method and system can obtain a higher labeling accuracy under the premise of less labeling workload.

The embodiment of the invention provides a full-process intelligent software detection method and system based on a large language model, and belongs to the technical field of software testing. Comprising the following steps: based on a large language model learning industry standard and a historical document, carrying out comprehensive scanning and consistency checking on tested data; automatically extracting demand features and constructing a knowledge graph in the process of learning industry standards and historical documents by the large language model; analyzing the demand document through the natural language processing capability of the large language model; performing security testing on the software application API and the host according to the test case; identifying software code vulnerabilities according to a traditional scanning tool and a large model reasoning capability; and generating a standard test report based on the repair suggestion and the detection report. According to the full-process intelligent software detection method, a linked and synergistic detection system is built through a large language model, all links of software detection are comprehensively enabled, and full-process detection of software is achieved.

The embodiment of the invention provides an intelligent retrieval method and system based on electricity transaction, and belongs to the technical field of electricity transaction retrieval. The intelligent retrieval method comprises the steps that historical documents of electricity transactions are acquired, and an electricity transaction knowledge base is constructed according to the historical documents; selecting a basic large model; finely adjusting the basic large model by adopting the electricity transaction knowledge base; obtaining a retrieval instruction of a current user; and obtaining a retrieval result according to the retrieval instruction and the fine-tuned basic large model. And by adopting a mode of finely adjusting the basic large model, the retrieval precision and efficiency of the power transaction can be effectively improved.

The invention relates to a document processing method and device, equipment and a storage medium. The method comprises the steps of obtaining a reference document, and performing paragraph segmentation on the reference document according to a document structure of the reference document to obtain first metadata of each paragraph in the reference document; positioning a paragraph to be verified in the reference document according to the first metadata and the existence condition of a historical version document of the reference document; and according to the target paragraph content of the paragraph to be checked, searching a to-be-processed document from the historical document, and according to the target paragraph content, revising the to-be-processed document. By adopting the method, the document processing efficiency and the accuracy of a document result can be improved.

The invention discloses an AR and AI fused immersive generation system for an ideological and political course historical scene. The system comprises a space-time narrative engine, a self-adaptive space anchor point system, a context awareness AI role generator, a multi-sensory fusion rendering engine and a cognitive load self-adaptive system. A space-time narrative engine constructs a knowledge graph from historical literatures through a named entity recognition model of a BERT architecture and a three-layer graph convolutional network, and an interaction script is generated through an LLaMA large language model. And the adaptive space anchor point system analyzes the physical space by adopting RANSAC and YOLO algorithms, and intelligently distributes the positions of the virtual elements after calculating the comprehensive quality score. A context aware (AI) role generator creates a historical character dialogue agent based on a Transformer decoder and a LoRA fine tuning. The multi-sensory fusion rendering engine realizes spatial audio based on HRTF, and uses MusicGen to generate age music. And the cognitive load self-adaptive system calculates a cognitive load value through four indexes such as eye movement tracking. According to the invention, intelligent generation of historical scenes, spatial adaptive mapping and personalized learning experience are realized.

This invention provides a method and system for assisting in the writing of audit text based on a large language model, belonging to the field of intelligent office technology. By constructing a structured knowledge base, it utilizes natural language processing technology to segment, identify entities, and cluster topics in historical documents, enabling efficient reuse of experience. It dynamically generates adaptable templates, automatically fills in fields with business data, and supports conditional logic adjustments to content. It calls the large language model to generate standardized text, and combines Pandas and visualization tools to generate tables and charts, achieving multi-element mixed layout and automated format verification. Through an interactive interface, it guides users to edit and save to the knowledge base, forming a closed-loop evolution mechanism for audit knowledge from reuse and generation to feedback, improving the standardization and efficiency of text writing.

A document management system can include an artificial intelligence-based document manager that can perform one or more predictive operations based on characteristics of a user, a document, a user account, or historical document activity. For instance, the document management system can apply a machine-learning model to determine how long an expiring agreement document is likely to take to renegotiate and can prompt a user to begin the renegotiation process in advance. The document management system can detect a change to language in a particular clause type and can prompt a user to update other documents that include the clause type to include the change. The document management system can determine a type of a document being worked on and can identify one or more actions that a corresponding user may want to take using a machine-learning model trained on similar documents and similar users.

The application discloses an optimization processing method and device of document search and electronic equipment, which can be applied to the field of big data. The method comprises the following steps: obtaining a document search request containing a search keyword; obtaining a document search result in a document library according to the search keyword; the document search result contains a plurality of target documents; outputting document information corresponding to the target documents through a search page, and a target page in the search page is in a visible state; obtaining a target search path matched with the search keyword in a path library; the path library contains a plurality of document search paths, and the document search path at least represents a document position of document information corresponding to a corresponding historical document in a historical search result; according to the document position represented by the target search path, at least the search page where the document information corresponding to the target document is located is adjusted, so that the document information corresponding to a first document corresponding to the target search path in the target document is displayed in the target page.

The invention discloses a policy graph dynamic updating method and device, and the method comprises the steps: obtaining a policy document, obtaining a historical related document according to the policy document, and judging whether the policy document forms an updated document of the historical related document or not based on the matching of a preset updating keyword rule base and / or based on the similarity calculation of a semantic vector; if it is determined that the updated documents are formed, extracting second knowledge of the updated documents through the first semantic model, collecting the updated documents, and accumulating vocabularies or semantic variations based on the updated documents; when the accumulated vocabulary or semantic variation meets a preset model training trigger condition, performing incremental training on the first semantic model based on the gathered update document and historical document to obtain a second semantic model, and replacing the first semantic model; and based on the second knowledge, updating the historical policy atlas to generate a target policy atlas, and replacing the historical policy atlas with the target policy atlas. According to the invention, automatic and intelligent dynamic updating of the policy knowledge graph is realized.

Systems, methods, and computer-readable media are provided for using generative Al enriched with metadata about historical document characteristics to transform documents of various formats, including images, to the fields and values they represent. A prompt template may be selected in association with a type of document. The prompt template indicates field definition(s) of field(s) to be detected in the document and location(s) in which the field(s) have been detected in prior documents. A large language model is prompted with a prompt generated using the prompt template to generate a result that assigns value(s) to the field(s). Output from the language model is used for identifying the field to value mapping for the document, such that data detected from the document may be stored in appropriate database structures of a database. Metadata stored in association with the prompt template is updated based on location(s) in the document in which the field(s) were detected, and the value(s) of the field(s) are stored in a database. Outbound documents may be similarly translated to detect values of corresponding fields requested by third parties, even if those values are not stored in the database. In this scenario, values for fields may be detected in outbound documents using the prompt templates enriched with metadata as processed by the large language model before such information is prepared to be sent to a third party.

The invention discloses a non-residual digital intelligent village construction method based on digital twinning and generative artificial intelligence, and the construction method comprises the following steps: constructing a rural non-residual digital twinning base, and collecting the static asset parameters and dynamic adjustment process of a rural non-residual item through a multi-modal sensing device; constructing a rural non-abandoned cultural knowledge graph, and extracting rural non-abandoned entities and semantic relationships thereof from historical literatures, oral histories and folk customs to form a structured knowledge base; training a rural non-abandoned generation type artificial intelligence large model, taking a knowledge graph as a constraint condition, and performing LoRA fine tuning by using a multi-modal data set; immersive interaction and content generation are realized through interaction between natural language or gestures and non-perpetual numeral intellectual twinborn villages; a cloud-side-end collaborative operation mechanism is established, user interaction data and offline workshop sales and reservation data are returned to a construction system for knowledge graph updating and artificial intelligence model iteration, and a self-evolution closed-loop mechanism is formed.

Method includes obtaining historical document images including text that correspond to different document classes; and generating a dictionary using text of the historical document images. The dictionary includes base words occurring with a greatest frequency in each document class. The base words are extracted from the text of the historical document images and arranged in datasets by a document class, where each dataset includes the base words of a same document class that occur with the greatest frequency within that document class. Trie structure is generated using the base words of the datasets that occur with a greatest frequency in each dataset. The trie structure includes internal nodes including root node and leaf nodes in which keys corresponding to the base words occurring with the greatest frequency in each dataset are respectively stored in predefined order. The trie structure is searchable in the predefined order starting with the root node.

The embodiment of the invention provides a document management method and device, equipment and a storage medium, and relates to the field of artificial intelligence and the application field of a large model in the field of financial science and technology. The method comprises the following steps: acquiring a newly added document; processing the newly added document through the large model, and determining a first function node corresponding to the demand in the newly added document; the first function node comprises execution information and management information of the first function; analyzing the first function node and a plurality of second function nodes stored in a database through a large model, and determining a demand type; the second function node information is a function node corresponding to a demand in the historical document and comprises execution information and management information of a second function; and according to the demand type, updating a part corresponding to the demand type in the current latest document version number to obtain a new version number, and updating the database. In this way, the document management efficiency can be improved.

This application provides a method, apparatus, device, storage medium, and program product for generating text data. It relates to the fintech field or other related fields. The method includes: responding to a user's input operation at a user interface, acquiring a user input command and a user identifier; acquiring historical documents corresponding to the user identifier from a pre-built historical document library; acquiring a user style feature vector based on the historical documents corresponding to the user identifier; acquiring a scene style feature vector of the scene to which the user input command belongs; acquiring a final style feature vector based on the user style feature vector and the scene style feature vector; inputting the user input command and the final style feature vector into a text generation model, causing the text generation model to generate text data; and displaying the text data at the user interface. This method improves the accuracy of generated text data and enhances the user experience.