148 results about "PageRank" patented technology

The invention relates to a document processing method and system based on text content extraction. The method comprises the steps that an original document containing text, image and format information is received, the encoding format of the document is automatically detected, character set conversion is executed, and hierarchical indexes including page numbers, paragraphs and tables are established for an unstructured document; the method comprises the following steps: synchronously processing text content and visual layout through a pre-trained visual-language model, extracting word-level and sentence-level semantic features by a text stream embedding layer, analyzing spatial distribution features of document elements by a visual encoder, and fusing text and visual features through a cross-modal attention mechanism; and loading the domain knowledge graph matched with the document type, and executing entity linking to associate the text mentions to the knowledge nodes. According to the document processing method and system based on text content extraction, through the synergistic effect of vision-text joint coding and knowledge enhancement, the accuracy of financial contract key clause recognition tasks is improved, the error rate is lower than that of industry benchmark products, and the semantic understanding precision is remarkably improved.

The invention discloses a document understanding and question-answering method and system based on multiple modes and multiple agents, and the method comprises the steps: firstly carrying out the preprocessing of a document, generating a text paragraph and a page image, and then employing a retrieval enhancement generation frame to retrieve and query related contexts from two channels of a text and the page image, obtaining a text candidate set and an image candidate set; performing preliminary fusion by using five types of functional agents and forming a candidate conclusion, forming cross-modal key information to limit a subsequent analysis range, performing fine-grained examination and analysis in text and image modalities, and outputting intra-modal conclusions with position indications such as page numbers, paragraphs and regions; and finally, performing consistency analysis and weighted fusion on the multi-source result to generate a final answer. According to the method, the text and the visual clues can be integrated at the same time, the complex multi-modal reasoning capability is improved, and the accuracy in tasks such as document question and answer is improved.

The invention belongs to the technical field of data analysis, and provides a questionnaire-based sentiment analysis method, system and equipment in order to solve the problem of inaccurate user sentiment analysis in the existing questionnaire. Using a pre-trained language model to extract a semantic vector of the topic text, and combining with the co-occurrence frequency of the multiple topic options to generate node-level local features; carrying out dynamic reasoning by adopting an improved graph neural network model, calculating a dynamic attention coefficient based on semantic similarity calculated by a topic text semantic vector and a jump probability between topics, and weighting and aggregating neighbor features, so as to obtain context-associated topic node features; the global emotion is calculated by using the PageRank thought, a final emotion analysis result is obtained, dynamic context association generated due to questionnaire jump logic is effectively captured, and the reliability of grasping the overall emotion venation of the user is improved.

The invention relates to the technical field of computer network data acquisition and processing, in particular to an intelligent news crawling system and method based on SpiderFlow, and the system comprises a dynamic time control module, an intelligent duplicate removal module, a paging processing module, a fault-tolerant data extraction module and a dynamic parameter configuration module. The dynamic time control module generates a time variable through a date calculation engine to realize accurate limitation of a crawling time range; the intelligent deduplication module adopts a three-level deduplication mechanism to ensure data uniqueness; the paging processing module supports static parameter and dynamic AJAX loading dual-mode paging, and realizes full-amount acquisition in combination with a self-increasing page number iterator and an intelligent termination strategy; the fault-tolerant data extraction module is integrated with CSS / XPath double parsers, and data integrity is guaranteed through field missing detection and messy code transcoding. Through multi-module cooperation and an intelligent algorithm, crawling efficiency, data quality and system stability are remarkably improved, and challenges such as webpage structure change and anti-crawling limitation are effectively handled.

The invention discloses a PDF contract file identification method and system, a medium and a program product, and relates to the technical field of information identification, and the method comprises the steps: carrying out the image analysis of an obtained PDF contract file through employing a deep learning model, generating an editable text, and synchronously identifying the page layout of the document, outputting structured data including page numbers, text paragraphs and coordinate information; performing multi-dimensional feature matching on the text paragraphs according to an adaptive semantic analysis algorithm, a preset contract template library and a dynamic keyword library, and positioning contract core element information; performing entity relationship verification on the content of the contract core element information, correcting an extraction error through dependency syntactic analysis and semantic role labeling, and establishing a contract element data set containing a confidence coefficient weight; and performing serialized packaging on the verified contract element data set to generate a contract element list conforming to electronic signature authentication. According to the invention, the processing efficiency and recognition precision of the PDF contract file are improved.

The invention relates to an intelligent procurement collaboration method and system, and the method comprises the following steps: S1, building a supplier knowledge graph based on multi-source heterogeneous data fusion through employing a graph embedding algorithm, achieving the associated storage of industrial and commercial information, performance records and quality reports through a Neo4j graph database, calculating the weight of a supplier node through employing a PageRank improved algorithm, and carrying out the calculation of the weight of the supplier node; s2, based on the three-dimensional supplier portrait matrix, using an improved collaborative filtering algorithm to carry out demand matching, analyzing a purchase demand document through an ElasticSearch semantic analysis engine, combining TF-IDF weighted cosine similarity calculation to realize intelligent recommendation, and generating a purchase demand scheme with a weight score. The supplier knowledge graph is constructed through the multi-source heterogeneous data fusion and graph embedding algorithm, the scattered industrial and commercial information, performance records and quality reports are stored in an associated mode, traditional data island limitation is broken through, and effective integration of supplier multi-dimensional features is achieved.

The invention relates to a multi-modal interaction-oriented automatic operation backtracking and digital person continuous talking control method and system, and the system comprises an operation snapshot module which records operation nodes through a semantic anchor point marking technology, such as an interface XPath page number and a PPT page number; the interruption detection module is used for capturing a user interruption instruction in real time; the progress prediction engine dynamically adjusts a follow-up demonstration path based on a program; and the continuous talk controller is used for coordinating RPA operation execution, digital human rendering and voice synthesis through a three-level priority thread pool. The problems of inaccurate operation flow backtracking and multi-modal thread conflict after demonstration interruption are solved, the breakpoint continuous talk error is obviously lower than that of a traditional scheme, and the method is suitable for PPT demonstration, cockpit explanation, business system training, course explanation and 3D modeling guidance scenes.

The invention belongs to the field of contract document translation, and provides a contract document translation method and system based on a large model, and the method comprises the steps: obtaining a target contract document, recognizing a picture in the target contract document, converting the picture into a text, carrying out the recognition of the text in the target contract document through taking the recognized line feed as a mark of a block, and taking a page number as a maximum unit, the method comprises the following steps of: performing preliminary blocking to form a plurality of text blocks, identifying character features in each block during preliminary blocking, correcting the text blocks according to the identified character features, merging the text blocks belonging to the same natural segment, and splitting the text blocks of incoherent texts; and translating the corrected text blocks by using a pre-trained large model, and writing the translated text blocks into a new file in a form conforming to the character features corresponding to the text blocks to obtain a translated contract document. According to the method, the professionality, correctness and readability of translated documents are improved.

The invention discloses a question answering method for low-altitude multi-modal fusion retrieval, which comprises the following steps of: extracting visual information of a document, generating structured description by utilizing a fine-tuning multi-modal visual language model, and performing semantic alignment to form a context enhancement fragment; inputting the fragments into an embedded coding model to construct a high-dimensional semantic vectorization library, and establishing a keyword retriever based on BM25; when a user inquires, vector and keyword retrieval results are weighted through a weight adaptive fusion device, reordering is carried out through a pre-trained expert model, and Top-k high-correlation evidence fragments are screened; top-k fragments are injected into a question and answer template, a large language model is called to generate a final answer, a document name and a page number are attached, evidence traceability is achieved, and meanwhile dynamic retrieval and parameter adjustment in a multi-session and multi-task environment are supported.

The invention discloses an automatic code defect repairing method based on a code attribute graph and a large model, which comprises the following steps: collecting a code sample, analyzing a source code, generating the code attribute graph fusing an abstract syntax tree, a control flow graph and a data flow graph, and then carrying out dynamic pruning according to node importance calculated based on node degrees, PageRank and data propagation dependency, so as to obtain a large model; key nodes are reserved to improve the graph learning efficiency; extracting features of the optimized graph through GAT, introducing multi-scale coding, and generating a graph embedding vector; in combination with node importance, high-value Tokens are screened, semantic embedding of structure perception is carried out, and more accurate text representation is obtained; generating a comprehensive feature by fusing the image and the text vector, and then inserting a mask at a position with weak dependence or high risk; performing fine tuning optimization on the mask position; and inputting the comprehensive features and the mask source codes into a pre-training language model to generate patches, performing multi-task evaluation and reordering, and selecting an optimal patch to complete repair.

The invention provides a PDF analysis method and system based on image-text semantic alignment, and belongs to the technical field of document processing. The method comprises the following steps: reading PDF or picture file byte data, and converting pictures into PDF byte data in a unified format; creating an image and a Markdown directory according to the output directory, the PDF name and an analysis method; extracting byte data of a specified page of the PDF according to starting and ending page numbers; selecting a back end to analyze PDF byte data to obtain a reasoning result, an image list and the like; generating an intermediate JSON containing PDF detailed information according to an analysis result; analyzing the positions of the text and the picture, and associating information to realize semantic alignment of the picture and the text; and generating multiple types of machine readable files according to the intermediate JSON and storing the files in a specified directory. According to the method, the PDF multi-mode content can be accurately extracted and subjected to semantic alignment, the PDF multi-mode content is efficiently converted into a machine readable format, the analysis accuracy and usability are improved, and the method is suitable for PDF analysis of multiple image-text tables such as scientific and technical literatures.

A domain adaptation procedure, such as fine-tuning training, is required to utilize a large-capacity PLM for a specific domain. Attempts in existing research have been made to improve performance of a PLM through domain adaptor technology based on an N-gram in order to reduce errors on the basis of the results of domain text error analysis of the PLM. Proposed is a method of selecting a semantic chunk through a domain semantic chunk graph and PageRank based on the existing domain adaptor research, with an N-gram as the semantic chunk. Proposed is also a method of domain-adapting a large-capacity PLM using semantic chunk dynamic weight masking, which reflects an output value of a PLM rather than simply integrating embedding values of semantic chunks, in a semantic chunk domain adaptor technology.

The invention relates to the technical field of parameterization modeling, in particular to a construction drawing intelligent design optimization method and system based on parameterization, and the method comprises the following steps: extracting boundary and annotation information, judging the orientation and affiliation relation of components, analyzing the coordinates of end points, adjusting the order of side lines, calculating the occupation offset, comparing the positions of intersection points, recognizing preorder nodes, and sorting. Extracting junction point distribution to judge a contact range and a connection relation, and determining a drawing page number by combining a component starting point and a path direction. According to the method, the spatial affiliation relation of the components is established through geometric boundaries and labeling information, the trend is adjusted according to the included angle difference value of side lines, the connection strength is recognized by combining the joint angle and the contact range, the drawing sequence is deduced according to the serial number sequence and the intersection point sequence, and the component combinations are continuously divided through attachment sections. The drawing page distribution is positioned based on the starting point of the component and the path direction, dynamic cooperation of component arrangement, sequence organization and drawing splicing is achieved, and the accuracy of component structure expression and the integrity of construction drawing generation are improved.

The invention provides a PageRank-based identification method and system for important nodes in a fused directed weighted network, and relates to the technical field of complex network analysis, the method comprises the following steps: constructing a fused directed weighted network, and fusing two single-layer directed weighted complex networks a and b by multiplexing partial nodes to form a fused network c; the output intensity of the node c of the fusion network is calculated according to the calculation formula that # imgabs0 # and # imgabs1 # are the output intensity of the node vi, and # imgabs2 # and # imgabs3 # are the number of the nodes, belonging to the single-layer directed weighted complex network a and the node number of the single-layer directed weighted complex network b, of neighbor nodes of the node vi; the importance value of a node v to be evaluated is iteratively calculated according to the calculation formula that # imgabs4 # sigma (0 < sigma < 1) is a damping coefficient, n is the total number of nodes of the fusion network, INR (vi) is the importance value of a node source vi pointing to the node v, and wc (vi, v) represents the directed edge weight from the node vi to the node v in the fusion network c; and according to the calculation result of the c node importance value of the fusion network, node importance sorting is carried out, and key nodes are identified.

The invention provides an OFD archive file retrieval method and system, and the method comprises the steps: receiving file data, and carrying out the authentication verification of the file data; splitting the file into file units and storing the file units into a message queue for asynchronous processing in sequence; extracting text information, table information and picture information in the OFD file to generate comprehensive text information; preprocessing the comprehensive text information, and recording the occurrence frequency of each segmented word and the corresponding page number; creating an index in a search engine cluster, wherein an index field comprises a file ID (Identity), word segmentation content, occurrence frequency and a page number; searching according to an input keyword, and arranging results in a descending order according to the occurrence frequency of word segmentation; and marking the page number of the keyword for retrieval and positioning. The method has the beneficial effects that rapid analysis and preprocessing of large-scale OFD data are realized through asynchronous processing of the message queue and an OCR (Optical Character Recognition) technology; redundancy is reduced through coarse-grained word segmentation during indexing, and recall is enhanced through fine-grained word segmentation during retrieval; the keyword granularity input by the user is adapted, and the retrieval scene adaptability is improved.

The invention discloses an electronic archive intelligent data management method and system based on a large model. The method comprises the following steps: firstly, acquiring an electronic archive file, performing unified access, identifying a text through an OCR (Optical Character Recognition) technology, and reserving original layout coordinate information; analyzing the document page by page to generate a page-level structured object containing text and position mapping; then, on the basis of a preset domain prompt template, calling a large language model to extract file metadata from the page-level object, recognizing entities and relationships in combination with file knowledge ontology, and generating a semantic triple containing an evidence source page number; and finally, fusing the metadata and the triple to construct an electronic archive knowledge graph with a page-level traceability function. According to the invention, through a whole-process semantic governance mechanism from access, analysis to knowledge generation, the problems of difficult analysis of unstructured data, weak semantic understanding and lack of traceability in traditional archive management are effectively solved, and the intelligent level and data credibility of archive governance are significantly improved.

The accuracy of existing machine learning models, software technologies, and computers are improved by using one or more machine learning models to predict a type of data that one or more numerical characters and / or one or more natural language word characters of a document correspond to. For instance, a Question Answering systems can be used to predict that a particular number value corresponds to a date, a billing amount, a page number, or the like.

The invention provides a content awareness and intelligent routing document analysis method and device, equipment and a medium, and relates to the technical field of artificial intelligence and natural language processing. The method comprises the steps of performing content perception on a to-be-recognized multi-modal document, and extracting a multi-dimensional feature vector of each page of the to-be-recognized multi-modal document; a preset routing decision model is adopted to route the multi-dimensional feature vectors, and a document analysis tool of each page of document is determined; carrying out statistics on document analysis tools of multi-page documents of the to-be-identified multi-modal document to obtain analysis page numbers of the document analysis tools; according to the analysis page number of each document analysis tool, calling each document analysis tool to analyze a document corresponding to the analysis page number to obtain a local analysis result; and integrating all the local analysis results to generate a target analysis result of the to-be-identified multi-modal document. According to the method, the selection and use threshold of a document analysis tool can be reduced, and the efficiency, quality and automation degree of document analysis are improved.

The invention discloses an automatic Markdown text conversion method and device, a medium and a program product, and the method comprises the steps: firstly analyzing a Markdown text to be converted by using a python-docx engine, and generating a basic Word document which comprises a cover, a text and a directory placeholder; then the basic Word document is stored as a temporary docx file; and then loading a temporary docx file by utilizing a win32com engine, identifying and releasing a space of a directory placeholder, inserting a real directory into the space of the directory placeholder, and refreshing an association relationship between a directory title and a page number in real time to obtain a converted document. According to the method, the advantages of Python-docx and win32com are integrated, so that automatic conversion from a Markdown text to a Word document with an automatic update directory and a standard page format is realized, and the document processing efficiency and quality are effectively improved.

The invention provides an intelligent file evaluation method and system based on cooperation of a multi-modal large model and a rule engine, and the method comprises the steps: carrying out the multi-modal content analysis of a file file, and generating an analysis result set containing a content segment, a page number identifier, a region position identifier, layout type information and a credibility score; combining the field description text with the anchor point description text to generate a structured evaluation fact set; inputting the set into a rule engine, executing normative rule judgment and generating a verification result with a judgment score; and finally, according to an evaluation dimension aggregation rule result, outputting a dimension score, an overall evaluation score and a question detail item. The system comprises a multi-modal content analysis module, an evaluation fact generation module, a rule engine module and an evaluation result output module. According to the method, high-precision automatic evaluation of complex file documents is realized, the evaluation efficiency and consistency are improved, and the method is suitable for smart office scenes.

The application discloses a code defect automatic repairing method based on code attribute graph and large model, comprising the following steps: collecting code samples, analyzing source code, generating code attribute graph fusing abstract syntax tree, control flow graph and data flow graph, then performing dynamic pruning according to node importance calculated based on node degree, PageRank and data propagation dependency degree, and retaining key nodes to improve graph learning efficiency; extracting features of the optimized graph through GAT, introducing multi-scale coding, and generating graph embedding vectors; combining node importance, screening high-value Token, performing structure-aware semantic embedding, and obtaining more accurate text representation; fusing graph and text vectors to generate comprehensive features, then inserting masks at weakly dependent or high-risk positions; fine-tuning and optimizing the mask positions; inputting the comprehensive features and the mask source code into a pre-trained language model to generate patches, performing multi-task evaluation and reordering, and selecting the optimal patch to complete the repairing.

The invention discloses a semantic structure feature-based ancient book content integrity detection method and system, and relates to the technical field of information processing, the system comprises a data acquisition unit, a chapter integrity analysis unit, a paragraph integrity analysis unit, an integrity comprehensive analysis unit, a result feedback unit and a display terminal; structural features of chapters and paragraphs are accurately extracted by utilizing a semantic structure recognition technology, and the content integrity of the ancient books is comprehensively detected from the aspects of chapters and paragraphs in combination with an integrity analysis model, so that the defect that a traditional means depends on manual proofreading or simple page number comparison and keyword extraction, and the efficiency is high is effectively overcome. According to the method, the problems of typesetting confusion, chapter page missing, quotation omission and fuzzy paraphrasing which are common in the ancient books are difficult to deal with, automatic and multi-level analysis of the structural content and semantic level content integrity of the ancient books is realized, and the detection and recognition efficiency of the content integrity of the ancient books is improved.

A system is provided.SOLUTION: A system comprising: means for inputting a page number read by a user; means for transmitting the input page number to a server; means for analyzing contents based on the page number and generating information of a synopsis, a line of sight, and characters by the server; means for receiving and displaying information from the server; means for inputting a user's question and transmitting the question to the server; means for generating and transmitting an answer based on the question by the server; and means for generating and transmitting an illustration using a AI model by the server.SELECTED DRAWING: Figure 1

The application discloses a document analysis and export method and device based on a multi-modal large model, equipment and a storage medium, and relates to the technical field of computers, which comprises the following steps: generating a task object based on the original file name of a to-be-processed document, determining a to-be-processed task; rendering the current page into a raster image, setting the text content of the current page as a page text clue, determining the to-be-processed analysis main text of the current page based on a prompt word template and in combination with the raster image and the page text clue, obtaining an update request for the analysis main text, calling a multi-modal large model based on the request, the to-be-processed analysis main text and the prompt word template to generate target analysis main text, marking the current page as confirmed based on a confirmation instruction, and storing the corresponding target analysis main text and raster image, generating a document file according to a merging and exporting instruction, and sequentially writing the original file name row, the page number row, the raster image and the target analysis main text into each confirmed page, so that the efficiency of document analysis and export is improved.

This invention discloses a multi-task operation ticket continuous printing system and method. The system includes: a status management module for centrally maintaining and managing the status of tasks, steps, and page numbers during the printing process; a dynamic layout calculation module for real-time measurement and pagination decisions on the operation ticket content, including a paginator unit to determine whether drawing is allowed based on the real-time remaining space on the page; and a structured drawing module for executing page drawing based on the layout calculation results, with its drawing cycle following a "measurement, decision, execution" process and synchronously updating the status. The method, based on the above system, forms a closed-loop control process through four steps: initializing the printing status, performing dynamic layout calculation, performing structured page drawing, and updating the printing status, which is executed cyclically until all tasks are completed. This invention achieves accurate dynamic content adaptation, intelligent pagination, globally continuous page numbers, and unified format output for multi-task, multi-step operation tickets, solving the problems of chaotic layout and difficult status management in existing technologies. The system architecture is clear and easy to maintain and expand.

The invention discloses a data query method and device and a storage medium, which can be applied to various scenes such as cloud technology, artificial intelligence, smart traffic, Internet of Vehicles and the like, and the method comprises the following steps: obtaining a starting line number and a target number of to-be-queried data in a database; determining the sum of the starting line number and the target number to obtain a target termination line number of the to-be-queried data; if the target termination line number is greater than or equal to a preset threshold value, obtaining a data query interval corresponding to the to-be-queried data; splitting the data query interval into at least two sub-intervals to obtain a data volume corresponding to each sub-interval; traversing the at least two sub-intervals, and determining a target sub-interval corresponding to the initial row number; and querying the to-be-queried data in the target sub-interval based on the target number and the sum of the number of the preorder sub-intervals. According to the method and the device, the target subinterval to which the to-be-queried data belongs is positioned during paging query, so that the data of the random page number can be quickly and accurately queried in the target subinterval.

This invention provides a drug interaction prediction method based on a bidirectional cross-view attention network, belonging to the field of drug interaction prediction technology. The method first constructs a Morgan fingerprint similarity view of the drug, an original DDI view, and a multi-scale diffusion view based on personalized PageRank. Then, a graph convolutional network with shared weights is used to co-encode the multiple views, generating a unified drug embedding representation. Finally, a bidirectional cross-view attention mechanism is used to achieve fine-grained interaction and alignment between the structural and attribute views, and interaction prediction is completed via a multilayer perceptron. This invention effectively alleviates the sparsity problem of the DDI network through multi-scale topology enhancement and achieves complementary enhancement between views using bidirectional cross-view attention, significantly improving the accuracy and generalization ability of drug interaction prediction. It can provide reliable technical support for drug development screening and clinical combined drug safety assessment.

This invention relates to the field of data processing technology, specifically to an intelligent proofreading system and method for electronic document formats based on multi-dimensional verification rules. It includes: a document structure parsing unit; a dynamic verification rule generation unit; an intelligent verification execution unit; and a verification result output unit. This invention eliminates differences in electronic document parsing. After extracting semantic blocks such as titles and body text by a paragraph-level semantic block recognition module, the intelligent verification execution unit calls a multi-dimensional rule set to verify core format elements. The verification result output unit accurately locates the page numbers, paragraphs, and character positions of deviations, generating correction guidelines and multi-format reports containing problem descriptions and compliance examples. This invention solves the problem of the lack of a formal, dedicated proofreading system for core document formats in existing technologies. It can comprehensively identify deviations and clarify the direction of correction, ensuring compliant document circulation while reducing labor costs and improving proofreading efficiency and convenience.

To provide a printed matter with page numbering by a reverse counting method. [Solution] A reverse counting method is used, with the total page number listed on the first page, then decreasing sequentially thereafter, with the final page being marked "1" or "0." This clearly shows the number of remaining pages, making it easier to feel a sense of accomplishment upon finishing reading. With the usual ascending numbering method, it is difficult to intuitively grasp the number of remaining pages, which can be a factor in reducing motivation to study. In addition, by using both reverse counting and regular ascending numbering, it is possible to accommodate conventional reference methods while facilitating the creation of study plans and maintaining motivation to study. Furthermore, it can be differentiated from the regular method, enhancing the originality of the teaching materials.

The application provides a method, system, device, medium and electronic equipment for completing a list after deleting a paging list. The application receives a user's request for deleting a target data item in a current paging list, sends a deletion instruction for the target data item to a server, removes the target data item from locally cached list data, acquires the number of locally cached list data after deletion, judges whether the current page number meets a completion trigger condition, wherein the completion trigger condition includes that the number of locally cached list data after deletion plus 1 is less than the number of total list data of the server, calculates a target offset position in response to meeting the completion trigger condition, sends a paging query request to the server according to the target offset position to determine completion data, receives the completion data returned by the server, and appends the completion data to the end of the local list to complete the list. Thus, the application provides a list completion method with low delay, high consistency and no inductance.