24 results about "Document tree" patented technology

The invention relates to a knowledge base construction method and device, computer equipment and a storage medium. The method comprises the following steps: performing directory structure extraction on a target document to obtain a directory structure tree; based on the directory structure tree, splitting the target document to obtain a plurality of knowledge blocks; when the label data does not exist in the target document, generating the label data corresponding to the target document based on a preset large language model and the abstract text of the target document; based on the label data of the at least one target document and the plurality of knowledge blocks, constructing a target knowledge base; the target knowledge base comprises a document tree composed of label data of the target document. By adopting the method, the retrieval efficiency and accuracy of the knowledge base can be improved.

The invention discloses a vehicle knowledge question-answering method based on a GraphRAG method and computer equipment, and the method comprises the steps: building a document tree structure based on the title hierarchy of a vehicle knowledge document, distributing a text identifier, and adding metadata information; converting the stored text units into text unit vectors by adopting a vector embedding model, and establishing a text vector index; calculating a similarity score between the query vector and the text unit vector, and determining a traditional RAG path text unit based on the similarity score; identifying entities and entity relationships of the text units according to the entity types and the relationship types by adopting a large language model so as to generate a vehicle knowledge graph; performing multi-level community division on the vehicle knowledge graph by adopting a Leiton community detection algorithm, and generating a structured community abstract according to each level of community; determining a GraphRAG path community abstract by adopting a large language model; and fusing the community abstract of the GraphRAG path with the text unit of the traditional RAG path and generating an answer.

The invention provides a retrieval method based on collaborative document tree reconstruction, which comprises the following steps: taking vector representation of each document block as a first-layer node of a multi-dimensional system document tree, executing GMM clustering on the nodes, and taking a structured abstract of a clustering center obtained by combining LLM as a next-layer node of the document tree; according to the vector representation of each document block and the hierarchical relationship and / or path distance among other nodes in the tree, based on a quinary training sample, enhancing an Embedding model and rearranging a Reranker model; representing a first cosine similarity between nodes except a root node in the tree based on a to-be-retrieved vector of the query text to obtain a global candidate text set; pruning and traversing nodes of the document tree layer by layer, and obtaining the similarity between the traversed nodes and the to-be-retrieved vector to obtain a fine candidate text set; and rearranging a union set of the global candidate text set and the fine candidate text set by adopting the enhanced Reranker model to obtain a retrieval result.

The invention discloses a question answering method based on structured document retrieval enhancement, and belongs to the technical field of artificial intelligence. The method comprises the following steps: performing structured information analysis and blocking processing on a structured document to construct a tree structure reflecting a hierarchical relationship of the document; based on a cross-document association relationship, combining the tree structures corresponding to the plurality of structured documents to form a multi-document tree structure network; and when a question request is received, generating corresponding retrieval enhancement information through the tree structure network, and inputting the retrieval enhancement information into the question and answer model to obtain answer information matched with the question request. According to the method, the multi-document tree structure network which takes the tree hierarchical relationship of the structured document as a core and fuses cross-document topic clustering and reference mapping is constructed, so that efficient retrieval enhanced questions and answers oriented to the structured document are realized, and the accuracy, context consistency and traceability of answers are remarkably improved.

This application discloses a method and an apparatus for training a model for text retrieval, a computer device, and a storage medium. Specifically, the method includes: parsing a text file and establishing a document tree; constructing a training sample according to the document tree, the training sample including a text pair and a relevance label related to the text pair; and training a model by using text and the relevance label in the training sample. In above way, valid positive and negative training samples can be established. The model can be configured for text retrieval after being trained by the text and the relevance label in the training sample, and a retrieval result is accurate and effective accordingly.

Embodiments of the present application disclose a webpage generation method and device, electronic equipment and storage medium, and relate to the technical field of computers, which can realize modification of a main color tone of a webpage. The webpage generation method comprises: in response to a webpage template loading instruction, loading and displaying a webpage template; in response to a main color tone configuration parameter input on the webpage template, configuring the main color tone configuration parameter in a cascading style sheet; in response to a webpage generation instruction, parsing document content of the webpage template to construct a document tree; parsing the cascading style sheet of the webpage template to construct a cascading style sheet object model tree; creating a rendering tree based on the document tree and the cascading style sheet object model tree; and generating a target webpage based on the rendering tree. The present application is suitable for a webpage design scenario.

The invention relates to the technical field of electromagnetic compatibility testing, in particular to a test document automatic generation method based on intelligent plan making. The method comprises the following steps: firstly, locking an effective standard based on environment-standard capability mapping, and triggering cascade constraints to determine a to-be-tested sample and a working mode; then analyzing a script to construct a parameter vector, and executing multi-dimensional Cartesian product operation; in the test data generation process, frequency parameters are extracted to serve as dynamic index keys, real-time retrieval is conducted, standard limit value constraints are injected, and a structured object list with embedded criteria is generated; and finally, positioning to a project XML document tree, serializing and injecting the data, driving a document object model to perform row-level style cloning and data filling by utilizing structured data, and associating resources to generate a persistent file. According to the method, full-coverage generation of the test data and automatic closed-loop of the criterion are realized, the problem of self-adaptive rendering of the variable-length data in a strict-format template is solved, and the completeness and normalization of document generation are improved.

The present disclosure provides a model training method and device, a knowledge extraction method and device, equipment and a medium, relating to the field of artificial intelligence, specifically to the field of knowledge graph technology. The scheme is: analyzing a sample document, determining the level to which each element in the sample document belongs in the sample document, and generating a document tree according to the level to which each element belongs, wherein the nodes of each level in the document tree are used to indicate the elements of the corresponding level in the sample document; for each node in the document tree, querying a target knowledge extraction model matching the type of the element indicated by the node, and using the target knowledge extraction model to extract knowledge from the element indicated by the node to obtain predicted knowledge; and training the target knowledge extraction model according to the difference between the predicted knowledge and the annotated knowledge corresponding to the element indicated by the node on the sample document. Thus, based on deep learning technology, each knowledge extraction model is trained, which can improve the prediction effect of each knowledge extraction model.

The invention provides a DNA storage medium-oriented scalable vector graphics (SVG) image coding method and system, and the method comprises the steps: reading an SVG file, analyzing the SVG image to obtain a document tree containing a plurality of nodes, and distributing structure information for representing the structure position of the nodes in the document tree for the nodes; coding the nodes to generate node fragments, wherein the node fragments comprise label codes generated according to labels of the nodes and attribute codes generated according to attributes of the nodes; aggregating the plurality of node fragments according to the label codes thereof to form at least one aggregation block, the aggregation block comprising header information and the plurality of node fragments, the header information being used for indexing the node fragments contained therein; the at least one aggregate block is converted into at least one DNA base sequence, and error correction information for error verification or repair is added to the at least one DNA base sequence. According to the method, the encoding length is reduced while the semantic integrity of the SVG is maintained, the error-resistant capability is improved, and the progressive decoding capability is provided.

The invention discloses a tree document retrieval method and device, equipment and a storage medium. Comprising the steps that a pre-stored document tree is obtained, leaf nodes of the document tree correspond to document content, and non-leaf nodes correspond to directory information; a task tree representing a retrieval path is initialized, the initial state of the task tree comprises a root node, and tetrad information of the root node comprises an original problem, a document tree root node path, document tree root node content and an empty historical record list; inputting information on the current task tree into the large language model, planning a retrieval path by the large language model based on the task tree, and outputting text information returned by a root node of the task tree; and obtaining a retrieval result based on the text information returned by the task tree root node. According to the document retrieval method based on the task tree and the document tree, reasoning and multi-step thinking ability of a large language model are utilized, the model can further explore and reasone according to existing node information, and more accurate and more comprehensive answers are generated.

The application discloses a tree document retrieval method, device, equipment and storage medium. The method comprises the following steps: obtaining a pre-stored document tree, the leaf nodes of the document tree correspond to document content, and the non-leaf nodes correspond to directory information; initializing a task tree representing a retrieval path, the initial state of the task tree comprises a root node, and the four-tuple information of the root node comprises an original question, a document tree root node path, document tree root node content and an empty history record list; inputting the information on the current task tree into a large language model, the large language model plans a retrieval path based on the task tree, and outputs text information returned by the root node of the task tree; and obtaining a retrieval result based on the text information returned by the root node of the task tree. The application proposes a document retrieval method based on a task tree and a document tree, uses the reasoning and multi-step thinking ability of the large language model, so that the model can further explore and reason according to the existing node information, and generate more accurate and comprehensive answers.

The application discloses a research report generation method and electronic equipment, relates to the computer technical field, carries out multi-modal content structured extraction to the uploaded document, forms a structured document tree and stores into a vector database; the document key information is retrieved from the database, the document key information is filled into the corresponding task decomposition template, and the research outline is generated after adaptive adjustment; knowledge gap is identified, the optimal execution scheme is determined in combination with the task priority and information demand characteristics; the task is decomposed and iterative retrieval is executed, information is retrieved from the database and external resources to fill the knowledge gap, and the iterative convergence condition is judged through information entropy change and multi-round gain threshold; after judging the iterative convergence, the analysis results generated by the multi-round iteration are integrated and optimized in stages. In this way, the document analysis and report writing efficiency can be improved, the research report with deep thinking is output, the knowledge gap is actively filled, the human limitations are avoided, and the analysis results are comprehensive and accurate.

The invention relates to a UML (Unified Modeling Language) model-oriented event-driven behavior code automatic generation method and system. The method comprises the following steps: exporting an XMI file containing class, activity and state machines from a UML (Unified Modeling Language) modeling tool, constructing a document tree through DOM (Document Object Model) analysis, and extracting meta-model elements and relationships thereof; and generating attributes and methods of the main class and the associated class and Java source codes of the main method according to an analysis result, taking the main method of the main class as a simulation / operation entry, and calling the main method of the associated class during operation. Aiming at the behavior type of the classifier: when the behavior type is Activity, traversing an identification sequence, branching / merging, parallelizing / synchronizing and circulating from an InitialNode, constructing a behavior control flow diagram, decomposing the behavior into SimpleTask and LoopTask, and generating a Java code containing an Execute method and a scheduling statement; and when the state conversion diagram is the StateMach, reconstructing the state conversion diagram and generating a state machine instance and a StateMach Runner operation management type instance code, thereby realizing migration execution driven by a signal event and a time event and mixed triggering of the signal event and the time event and the activity. The method is suitable for the field with high requirements for model consistency and rapid verification.

Embodiments include a content collaboration system that can be configured to display a hierarchical document tree that includes graphical objects corresponding to content items hosted by the content collaboration system. The collaboration system can receive a selection of a graphical object corresponding to a content item for archiving, and in response, generate a first updated hierarchical relationship that includes the archived content item and generate a second updated hierarchical relationship that excludes the archived content item. The collaboration system can construct a first hierarchical document tree instance based on the first updated hierarchical relationship for displaying the graphical objects for the first user account and construct a second hierarchical document tree instance based on the second updated hierarchical relationship for displaying the graphical objects for a second user account.

Embodiments include a content collaboration system that can be configured to display a hierarchical document tree that includes graphical objects corresponding to content items hosted by the content collaboration system. The collaboration system can receive a selection of a graphical object corresponding to a content item for archiving, and in response, generate a first updated hierarchical relationship that includes the archived content item and generate a second updated hierarchical relationship that excludes the archived content item. The collaboration system can construct a first hierarchical document tree instance based on the first updated hierarchical relationship for displaying the graphical objects for the first user account and construct a second hierarchical document tree instance based on the second updated hierarchical relationship for displaying the graphical objects for a second user account.

The invention provides a task template generation method and device based on reverse engineering, equipment and a medium. The method comprises the steps of obtaining an example document, analyzing the example document, extracting a plurality of text blocks, and associating visual layout metadata for each text block; constructing a document tree based on content semantics and visual layout metadata of the text blocks, and performing hierarchical clustering on text contents in the document tree to generate a logic step sequence; identifying named entities in the full text of the example document, replacing the named entities with parameterized variables, and generating a parameter mapping table based on the corresponding relationship between the parameterized variables and the named entities; according to the method, the logic step sequence is associated with the parameter mapping table, the task template is generated, the expert analysis normal form can be extracted from diversified documents in a robust mode, the highly reusable and parameterized task template is generated, the efficiency, the normalization and the intelligent level of research work are remarkably improved, and the method has wide application prospects in the field of knowledge automation.

The application relates to a knowledge base construction method and device, computer equipment and a storage medium. The method comprises the following steps: extracting a directory structure of a target document to obtain a directory structure tree; splitting the target document based on the directory structure tree to obtain a plurality of knowledge blocks; when the target document does not have label data, generating label data corresponding to the target document based on a preset large language model and an abstract text of the target document; and constructing a target knowledge base based on the label data of at least one target document and the plurality of knowledge blocks. The target knowledge base comprises a document tree composed of the label data of the target document. The method can improve the retrieval efficiency and accuracy of the knowledge base.

The application provides a graduation thesis review method and system based on task decomposition and collaborative decision-making, and belongs to the technical field of artificial intelligence. That is, a thesis analysis intelligent agent analyzes the received graduation thesis to generate a structured document tree; a thesis review intelligent agent reviews the structured document tree according to a graduation thesis review prompt framework to generate a graduation thesis review result; a thesis review result optimization intelligent agent optimizes the graduation thesis review result to generate an optimized graduation thesis review result; and a thesis review result finalization intelligent agent self-corrects the optimized graduation thesis review result to output a final version of a graduation thesis review report. The application realizes the automatic decomposition and collaborative decision-making evaluation of long text tasks such as graduation theses by systemically integrating prompt words, tool calls and context protocols, thereby not only enhancing the explainability and scalability of the review process, but also providing efficient and stable technical support for actual review work scenarios.

In various examples, a technique for querying a hierarchical representation of a document involves receiving a hierarchical document tree containing one or more parent and child nodes, each associated with a part of a source document, and generating, using a machine learning model and based on at least one query prompt and the contents of each of the one or more parent and child nodes, a similarity numerical value for each of the one or more parent and child nodes.The technique also includes calculating combined similarity number values ​​assigned to each of one or more child nodes, based at least on the similarity number value assigned to the child node and one or more similarity number values ​​assigned to one or more nodes with a parent relationship to the child node, and generating a query result based at least on the combined similarity number values ​​assigned to the one or more child nodes.

This invention proposes a document tree-based intelligent information matching and retrieval method. It combines document structure with feature information to generate a document tree carrying document feature information, enabling hierarchical matching and rapidly converging the number of target documents. This reduces computational load, increases retrieval speed, and improves matching accuracy. Applying this technical solution enables fast, efficient, timely, and accurate matching and searching of massive amounts of document data, improving knowledge production efficiency. This invention effectively addresses the cost and accuracy issues in knowledge production and information retrieval. The document tree designed in this invention intelligently and objectively captures document keyword information and carries the document's theme, structure, and content information, eliminating the need for manual annotation and extraction. This effectively reduces labor costs and solves the problems of inaccurate and incomplete keyword annotation, significantly improving the accuracy of matching and retrieval.

In various examples, a technique for querying a hierarchical representation of a document includes receiving a hierarchical document tree including one or more parent and child nodes each associated with a portion of a source document and generating, using a machine learning model and based at least on a query prompt and contents of each of the one or more parent and child nodes, a similarity score for each of the one or more parent and child nodes. The technique also includes calculating combined similarity scores associated with each of one or more child nodes based at least on the similarity score associated with the child node and one or more similarity scores associated with one or more nodes having a parent relationship to the child node and generating a query result based at least on the combined similarity scores associated with the one or more child nodes.

This invention proposes a method and system for constructing a knowledge base based on a tree structure. The method includes: document parsing and hierarchical tree initialization; semantic segmentation and leaf node generation, traversing nodes to divide long text content into a set of semantically independent text fragments, creating new leaf nodes for each text fragment, and using the corresponding nodes in the original document tree as their parent nodes to form a basic knowledge tree; bottom-up multi-granularity summary generation, starting from the leaf nodes of the basic knowledge tree, generating summaries for each internal node; cross-node relationship enhancement, forming an enhanced knowledge graph; structured indexing and persistence, constructing a searchable index for each node in the knowledge base. This invention maintains the logic and coherence of the input obtained by the large language model, thereby significantly improving the accuracy, professionalism, and logical consistency of the generated answers, effectively avoiding misinterpretation or logical contradictions caused by contextual fragmentation. It also improves the accuracy and efficiency of retrieval.

Systems, methods, and software are disclosed herein for automating the editing of documents based on ink gestures via an LLM integration in various implementations. In an implementation, a computing device receives an ink gesture made by a user with respect to a document displayed in the user interface of an application. The ink gesture includes ink strokes made on a portion of the document. The computing device generates a prompt for an LLM to obtain suggested revisions to the document, wherein the prompt includes an image of the ink strokes made on the portion of the document. The prompt also includes a document tree indicative of a structure of the document. The computing device surfaces a suggested revision generated by the LLM in response to the prompt in the user interface.

The application provides a retrieval method based on collaborative document tree reconstruction, taking the vector representation of each document block as the first layer node of a multi-dimensional system document tree, performing GMM clustering on the node, and combining the structured summary of the clustering center obtained by LLM as the next layer node of the document tree; based on the vector representation of each document block and the hierarchical relationship and / or path distance between other nodes in the tree, the Embedding model and the Reranker model are enhanced based on a five-element training sample; based on the first cosine similarity between the vector representation of the query text to be retrieved and the nodes in the tree except the root node, a global candidate text set is obtained; the nodes of the document tree are pruned and traversed layer by layer, the similarity between the traversed node and the vector to be retrieved is obtained, and a fine candidate text set is obtained; the union of the global candidate text set and the fine candidate text set is rearranged by using the enhanced Reranker model, and the retrieval result is obtained.