CN-122019768-A - Visual display method and system for duplicate checking result of standard document

CN122019768ACN 122019768 ACN122019768 ACN 122019768ACN-122019768-A

Abstract

The invention discloses a visual display method and a visual display system for a duplicate checking result of a standard document, which relate to the technical field of document information processing and comprise the steps of obtaining a target standard document and a comparison standard document set, and processing the target standard document and the comparison standard document set to generate a structured document object set; the method comprises the steps of carrying out multi-level semantic association degree fusion based on a structured document object set, constructing a multi-dimensional association knowledge network, rendering and generating an initial visual view, monitoring a user interaction load index and a view complexity index of the initial visual view in the interaction process of a user and the initial visual view, evaluating real-time cognitive load of the user based on the user interaction load index and the view complexity index, dynamically scheduling and optimizing the presentation form of the initial visual view according to the real-time cognitive load, and generating an optimized view stream. According to the invention, through self-adaptive view optimization and interactive strategy deduction, the analysis efficiency and decision support capability in a complex information environment are effectively improved.

Inventors

SHI BOLI
WANG YALI
NIE ZHAOYING
CAO RUIJIE
ZHANG GUOSHU
CAO LILI
LIU HUI
XU QING
WU LINFENG
WANG LIJUN
XU JINGTING
CHEN JING

Assignees

中国标准化研究院

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. The visual display method of the standard document duplicate checking result is characterized by comprising the steps of obtaining a target standard document and a comparison standard document set, and processing to generate a structured document object set; carrying out multi-level semantic association degree fusion based on the structured document object set, constructing a multi-dimensional association knowledge network, and rendering to generate an initial visual view; monitoring a user interaction load index and a view complexity index of an initial visual view in the interaction process of a user and the initial visual view; Based on the user interaction load index and the view complexity index, the real-time cognitive load of the user is estimated, an optimized view stream is generated according to the real-time cognitive load dynamic scheduling and optimizing the presentation form of the initial visual view, a sand table deduction branch is created by receiving a virtual modification instruction of the user to a target standard document, and the structural evolution of the multidimensional associated knowledge network caused by each sand table deduction branch is deduced and compared; Extracting a weighing mode and a potential path from the structural evolution, and performing strategy path visual guidance and dominant strategy marking in the optimized view flow; And recording the scheduling log of the optimized view flow and the guiding information of the sand table deduction branch, and generating a comprehensive report by combining the multidimensional associated knowledge network state corresponding to the sand table deduction branch of the user.
2. The visual display method of the standard document duplicate checking result in the technical protection scope 1 is characterized by comprising the following steps of: Receiving a target standard document, retrieving a comparison standard document set from a comparison document database, analyzing document formats of the target standard document and the comparison standard document set, and extracting pure text; analyzing the chapter structure of the pure text by utilizing rules and a natural language processing model, and dividing a technical field chapter block, a background technical chapter block and a technical protection range chapter block; Generating semantic vectors for the whole text, the technical field chapter block, the background technical chapter block and the technical protection scope chapter block of each document in the target standard document and the comparison standard document set respectively by using a pre-trained language model; And packaging the pure text, the technical field chapter block, the background technical chapter block, the technical protection scope chapter block and semantic vectors of all parts of each document in the target standard document and the comparison standard document set into a structured document object set.
3. The visual display method for the duplicate checking result of the standard document in the technical protection range 2 is characterized by comprising the following steps of: performing multi-level semantic association degree fusion based on a structured document object set, constructing a multi-dimensional association knowledge network, and rendering to generate an initial visual view, wherein the method comprises the following steps of: obtaining a document-level association degree based on comparing the whole text semantic vector of the target standard document in the structured document object set with the whole text semantic vector of each comparison standard document; Obtaining a technical protection scope level association degree based on comparing the technical protection scope chapter block semantic vector of the target standard document in the structured document object set with the technical protection scope chapter block semantic vector of each comparison standard document; obtaining a background level association degree based on comparing the background section block semantic vector of the target standard document in the structured document object set with the background section block semantic vector of each comparison standard document; applying a preset weighted fusion rule, and fusing the document-level association degree, the technical protection range-level association degree and the background-level association degree of each comparison standard document with the target standard document to generate a multi-level semantic association degree fusion score; constructing a multidimensional associated knowledge network by taking all documents in the structured document object set as nodes and taking the fusion score of the multi-level semantic association degree and the reference relationship among the documents as edges; And (3) encoding the nodes and the edges in the multidimensional associated knowledge network according to a preset visual mapping rule, and rendering on an interactive canvas by using a graphic layout algorithm to generate an initial visual view.
4. The visual display method of the standard document duplicate checking result in the technical protection range 3 is characterized by monitoring a user interaction load index and a view complexity index of an initial visual view in the interaction process of the user and the initial visual view, and comprising the following steps: capturing each click operation event, each dragging operation event and each zooming operation event in the interaction process of the user and the initial visual view to form an interaction operation event; based on the interaction operation event, counting the click frequency, the average speed of the dragging operation and the frequency of the zooming operation in unit time, and taking the click frequency, the average speed of the dragging operation and the frequency of the zooming operation as user interaction load indexes; in the interaction process of a user and the initial visual view, counting the total number of nodes rendered in the current visual area by the initial visual view, and taking the total number of nodes as a complex index of the total number of nodes; In the interaction process of a user and the initial visual view, counting the total density of the edges rendered by the initial visual view in the current visual area, and taking the total density as a total density complex index of the edges; In the interaction process of a user and the initial visual view, counting the types and the numbers of different visual coding attributes used by the initial visual view in the current visual area, and taking the types and the numbers as complex indexes of the visual coding types; the user interaction load index, the node total number complex index, the total density complex index of the edges and the visual coding type complex index are collectively called as a user interaction load index and a view complex index of an initial visual view.
5. The visual display method of the standard document duplicate checking result in the technical protection range 4 is characterized by evaluating the real-time cognitive load of a user based on the user interaction load index and the view complexity index, and comprises the following steps: comparing the click frequency, the average speed of the dragging operation and the frequency of the zooming operation in the user interaction load index with preset cognitive load reference parameters to obtain a first load evaluation value; comparing the total number of nodes in the view complexity index, the total density complexity index of the edges and the visual coding type complexity index with corresponding visual complexity thresholds respectively, and obtaining a second load evaluation value by weighting and summing; And synthesizing the first load evaluation value and the second load evaluation value according to a predefined cognitive load synthesis rule, and outputting a real-time cognitive load value of the user.
6. The visual display method of the standard document duplicate checking result in the technical protection range 5 is characterized in that an optimized view stream is generated according to the real-time cognitive load dynamic scheduling and optimizing the presentation form of an initial visual view, and the method comprises the following steps: Comparing the real-time cognitive load value with a first load threshold value and a second load threshold value, and judging the cognitive state of the current user; When the real-time cognitive load value exceeds a first load threshold value, triggering a view simplification rule, and executing operations of desalting secondary associated edges, folding non-focal area nodes and hiding partial metadata labels on the initial visual view; When the real-time cognitive load value is lower than a second load threshold value and the interactive operation event is concentrated in a specific area, triggering a view focusing rule, and executing the operation of expanding details on the specific area in the initial visual view; The adjusted initial visual view, resulting from the view reduction rule or view focusing rule, is serialized into an optimized view stream.
7. The visual display method of the standard document duplicate checking result in the technical protection scope 6 is characterized by receiving a virtual modification instruction of a user to a target standard document to create a sand table deduction branch, deducing and comparing the structural evolution of a multidimensional associated knowledge network caused by each sand table deduction branch, and comprising the following steps: in the optimized view flow, receiving a virtual modification instruction proposed for the content of a target standard document, and updating a corresponding semantic vector of the target standard document; Using the updated target standard document semantic vector to re-derive multi-level semantic association degree fusion scores of all document nodes in the multi-dimensional association knowledge network, and reconstructing the multi-dimensional association knowledge network with each sand table deduction branch corresponding to the update; And (3) comparing the updated multidimensional associated knowledge network corresponding to each sand table deduction branch with the original multidimensional associated knowledge network in parallel, and identifying the enhancement, the weakening and the disappearance of the associated edges between the nodes as the structural evolution of the multidimensional associated knowledge network.
8. The method for visualizing and displaying the standard document duplicate checking result in the technical protection scope 7 is characterized in that a trade-off mode and a potential path are extracted from structural evolution, and policy path visualization guiding and dominant policy marking are performed in an optimized view stream, and the method comprises the following steps of: based on the structural evolution of the multidimensional association knowledge network, comparing the association edge change combination caused by different sand table deduction branches, and summarizing an association countermeasure mode of increasing and decreasing association strength and an association cooperative mode of weakening multilateral association in the same direction to jointly form a balance mode; Based on the weighing mode, the structure evolution with logic continuity is connected in series, and an evolution sequence leading to a low-association-density area from the current network state is planned to be used as a potential path; Generating a semitransparent guide track which is covered on the current view and extends dynamically in the optimized view flow according to the evolution direction of the potential path and the key nodes, and determining a strategy path visual guide; and (3) quantitatively scoring the performance of the sand table deduction branch on the modification cost and the risk reduction effect by applying a preset multi-objective evaluation rule, judging the branch with the score higher than a preset dominance threshold value as a dominance strategy, marking a specific visual symbol different from a common branch in the optimized view flow, and determining a dominance strategy mark.
9. The visual display method of the standard document duplicate checking result in the technical protection scope 8 is characterized by recording the scheduling log of the optimized view flow and the guiding information of the sand table deduction branch, and generating a comprehensive report by combining the multidimensional associated knowledge network state corresponding to the sand table deduction branch of the user, and comprising the following steps: capturing and storing a scheduling log of an optimized view stream, wherein the scheduling log comprises triggering time, reasons and specific operation records of view simplification rules and view focusing rules; capturing and storing guiding information of a sand table deduction branch, wherein the guiding information comprises track data of strategy path visual guiding and scoring basis of dominant strategy marks; when a user selects one sand table deduction branch as a final strategy, locking the updated multidimensional associated knowledge network state corresponding to the sand table deduction branch; Based on the scheduling log, the guiding information and the updated multidimensional associated knowledge network state corresponding to the final strategy, formatting and integrating and image-text mixed arranging are carried out according to a preset report template, and a comprehensive report is output.
10. The visual display system for the standard document duplicate checking result is characterized by comprising a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring a target standard document and a comparison standard document set, and processing the target standard document and the comparison standard document set to generate a structured document object set; The rendering module is used for carrying out multi-level semantic association degree fusion based on the structured document object set, constructing a multi-dimensional association knowledge network and rendering to generate an initial visual view; The monitoring module is used for monitoring a user interaction load index and a view complexity index of the initial visual view in the interaction process of the user and the initial visual view; The deduction module is used for evaluating the real-time cognitive load of the user based on the user interaction load index and the view complexity index, generating an optimized view stream according to the real-time cognitive load dynamic scheduling and optimizing the presentation form of the initial visual view, receiving the virtual modification instruction of the user to the target standard document, creating a sand table deduction branch, deducting and contrasting the structural evolution of the multidimensional associated knowledge network caused by each sand table deduction branch; The strategy generation module is used for extracting a weighing mode and a potential path from the structural evolution and carrying out strategy path visual guidance and dominant strategy marking in the optimized view flow; And the comprehensive report generation module records the scheduling log of the optimized view flow and the guide information of the sand table deduction branch, and generates a comprehensive report by combining the multidimensional associated knowledge network state corresponding to the sand table deduction branch of the user.

Description

Visual display method and system for duplicate checking result of standard document Technical Field The invention relates to the technical field of document information processing, in particular to a visual display method and a visual display system for a standard document duplicate checking result. Background With the application of artificial intelligence technology, standard document searching has been developed from keyword matching to semantic analysis based on deep learning, the existing mainstream scheme generally analyzes documents and generates semantic vectors by using a large language model, document association is quantified by calculating vector similarity, and finally results are presented in a list or static knowledge graph form. When facing to complex associated networks formed by massive documents, the existing visualization technology has limitations in that static state presentation of the state of the repeated results is lacking, the complex network is easy to cause overload of visual information, users are forced to consume a great deal of effort to conduct manual view adjustment, so that deep research and judgment on core laws and technical problems are dispersed, the existing scheme is read-only, users cannot directly conduct virtual modification on the technical scheme in a visualization environment, and dynamic influences of different modification strategies on associated patterns are observed timely and contrastively, so that key strategy deduction and decision making are difficult to effectively support in the analysis process. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides a visual display method for a standard document duplicate checking result, which solves the problems that the cognitive load of a user is too high and dynamic policy deduction and decision support cannot be directly carried out in a visual environment due to a static presentation and read-only interaction mode in the prior art. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the present invention provides a method for visually displaying a duplication result of a standard document, which includes obtaining a target standard document and a comparison standard document set, and processing the target standard document and the comparison standard document set to generate a structured document object set; carrying out multi-level semantic association degree fusion based on the structured document object set, constructing a multi-dimensional association knowledge network, and rendering to generate an initial visual view; monitoring a user interaction load index and a view complexity index of an initial visual view in the interaction process of a user and the initial visual view; Based on the user interaction load index and the view complexity index, the real-time cognitive load of the user is estimated, an optimized view stream is generated according to the real-time cognitive load dynamic scheduling and optimizing the presentation form of the initial visual view, a sand table deduction branch is created by receiving a virtual modification instruction of the user to a target standard document, and the structural evolution of the multidimensional associated knowledge network caused by each sand table deduction branch is deduced and compared; Extracting a weighing mode and a potential path from the structural evolution, and performing strategy path visual guidance and dominant strategy marking in the optimized view flow; And recording the scheduling log of the optimized view flow and the guiding information of the sand table deduction branch, and generating a comprehensive report by combining the multidimensional associated knowledge network state corresponding to the sand table deduction branch of the user. The invention relates to a method for visualizing and displaying a repeated result of a standard document, which comprises the following steps of: Receiving a target standard document, retrieving a comparison standard document set from a comparison document database, analyzing document formats of the target standard document and the comparison standard document set, and extracting pure text; analyzing the chapter structure of the pure text by utilizing rules and a natural language processing model, and dividing a technical field chapter block, a background technical chapter block and a technical protection range chapter block; Generating semantic vectors for the whole text, the technical field chapter block, the background technical chapter block and the technical protection scope chapter block of each document in the target standard document and the comparison standard document set respectively by using a pre-trained language model; And packaging the pure text, the technical field chapter block, the background technical chapter block,