Enterprise registration file sharing and shareholder informed right guarantee system based on whole-chain data traceability and audit

Abstract

The application discloses an enterprise registration file sharing and shareholder right-to-know guarantee system based on full-chain data tracing and auditing, and relates to the technical field of file management. The system comprises an identity authentication module, a file storage management module, a file sharing module, a shareholder right-to-know guarantee module, a comprehensive statistics module and a system security module. The system uses blockchain technology to realize full-chain tracing of file data, completely records all change operations, and significantly improves the transparency of file management. The safety and authenticity of shared files are ensured through electronic signatures, watermarks and two-dimensional code anti-counterfeiting technology. Shareholders can track the enterprise equity status and change records in real time, and timely understand the enterprise change information through a dynamic notification mechanism, thereby comprehensively guaranteeing the right-to-know of shareholders. Through dynamic encryption, permission management and abnormal behavior detection, the safety of file storage and access is ensured, and data leakage and tampering are effectively prevented.

Description

Technical Field

[0001] The present invention relates to the technical field of archive management, and in particular to a system for sharing enterprise registration archives and protecting shareholders' right to know based on full-chain data tracing and auditing. Background Art

[0002] With the rapid development of informatization and digitization, the digitization and informatization of enterprise registration files has become a growing trend. As crucial foundational data for enterprises, enterprise registration files contain key information such as company establishment, changes, and equity structure. However, existing enterprise registration file management and sharing models still face numerous challenges in practical application, particularly in areas such as query efficiency, management transparency, security, and user rights protection, which require urgent improvement and enhancement.

[0003] Currently, most business registration file inquiries still rely on traditional offline methods, requiring users to submit paper applications at the industrial and commercial enterprise inquiry window. This cumbersome and time-consuming process, especially for remote users, can be costly to obtain. Existing systems struggle to support rapid online inquiries, preventing users from instantly accessing the latest business registration information, limiting the timeliness and availability of this information.

[0004] Enterprise registration files are typically managed in a centralized manner, with relevant institutions or departments storing and maintaining them. Due to the large volume of data involved and the frequent changes, centralized management struggles to synchronize information in a timely manner, potentially resulting in users accessing outdated data. Under this traditional model, the storage and access records of archival data are untraceable, making it difficult for users to verify whether archival information has been modified or leaked, reducing transparency in archival management.

[0005] With the advancement of enterprise information management, the demand for sharing corporate registration records has increased significantly. Different user groups (such as shareholders and the general public) have different access requirements for records, but existing systems struggle to flexibly adjust access rights based on user identity, potentially leading to insufficient or excessive data sharing. During record sharing, data anti-counterfeiting and tampering protection capabilities are insufficient, potentially leading to malicious tampering, forgery, or leakage of records.

[0006] Corporate registration records play a crucial role in safeguarding shareholder rights. However, current record systems lack effective verification methods for verifying the authenticity of inquiring users, preventing legitimate shareholders from accessing the information they deserve. This makes it difficult for shareholders to monitor changes in a company's equity in real time and track significant changes, limiting their ability to oversee the company. Summary of the Invention

[0007] In response to problems such as insufficient efficiency in enterprise registration file management and insufficient data credibility, the present invention proposes an enterprise registration file sharing and shareholder right to know protection system based on full-chain data traceability and auditing, which can improve query efficiency, enhance information transparency, ensure data security and improve user rights protection mechanism to better adapt to the needs of enterprise informatization development.

[0008] In order to achieve the above object, the present invention designs the following technical solutions:

[0009] A system for sharing enterprise registration records and protecting shareholders' right to know based on full-chain data traceability and auditing, including:

[0010] The identity authentication module is used to implement user identity authentication for public users, government agency users, and contracted users through a distributed key management system combined with a CA authentication platform and an electronic business license platform, and to determine whether the inquirer is a shareholder through an interface with a natural person identity authentication platform;

[0011] The archive storage management module is used to implement the storage, management, traceability and auditing of enterprise registration archives based on blockchain technology and distributed storage technology;

[0012] The archive sharing module is used for multi-level directory index tree-like information display and quick query of enterprise registration archives, automatic retrieval of archive information, interactive multi-level index query, and quick location of archive images based on the case file directory and the file directory within the volume. Under authorized conditions, archive images are processed with electronic signatures, watermarks, and QR codes for anti-counterfeiting, and enterprise registration archive information is shared;

[0013] The shareholder right to know protection module is used to provide share ownership tracking through blockchain technology, check the status of corporate shares and the update of corporate registration information, and ensure shareholders' right to know;

[0014] Comprehensive statistics module, used to collect statistics on archive query and sharing data based on the query person, enterprise directory, query content and query time, and generate various forms of reports;

[0015] The system security module is used to provide file confidentiality control, user authority management and full-process security log recording functions to ensure the security of file query and sharing.

[0016] As a preferred solution of the present invention, the identity authentication module includes:

[0017] The identity credential verification unit is used to receive identity credentials submitted by users, including CA certificates, electronic business licenses, or natural person identity information, verify the legitimacy of the CA certificate by connecting to the CA authentication platform, verify the validity of the business license by connecting to the electronic business license platform, and call the natural person identity authentication platform to verify whether the user is a legitimate corporate shareholder;

[0018] The permission allocation unit is used to dynamically generate the user's access permission information based on the identity credential verification result, including the user's identity category, permission level and the data range that the user can access;

[0019] A distributed key generation unit, which is used to generate dynamic keys for user authentication and permission information based on a distributed key management system to support subsequent data encryption and decryption operations;

[0020] The verification result output unit is used to output the user identity authentication result, including user identity authentication status, user role type, user authority level, authentication timestamp, verification source, and whether it matches the enterprise information.

[0021] As a preferred solution of the present invention, the archive storage management module includes:

[0022] The data storage unit is used for the storage and management of enterprise registration files. It uses distributed storage technology to store the archival image data, associated information, and multi-level index data of enterprise registration files. The metadata of enterprise registration files is stored on the blockchain. The stored data is dynamically encrypted using a distributed key management system.

[0023] The data tracing unit is used to trace the change records, user operation logs, and data access records of enterprise registration files through the blockchain timestamp function. It supports querying the traceability of file information at any time point, and generates a change chain to display all historical versions of the file data. It supports visual display of the tracing results, including the time of the file data change, the operating entity, and the specific content of the change;

[0024] The data audit unit is used to record and audit all archival data access operations in the system in real time through blockchain smart contracts, verify the permissions of archival operations, mark and record unauthorized operations or abnormal behaviors, and generate audit logs.

[0025] As a preferred solution of the present invention, the data tracing unit performs pattern matching and change chain analysis of archive change records based on an improved multiple sequence alignment algorithm, specifically including:

[0026] When formatting archival data, we use knowledge graph technology to construct semantic associations within archival data. We model enterprise registration information, shareholder structure, and change records as directed graphs of entities and relationships. Nodes represent fields, and edges represent semantic relationships between fields. The fields of each archival version are structured into a graph model.

[0027] The multi-version archive comparison process is expanded to a comparison based on graph embedding and graph neural networks. The formatted archive field graph is input into the graph embedding model, and the graph nodes are mapped into feature vectors of fixed dimensions. The feature similarity of the multi-version archive fields is learned through GNN, and the change similarity of each field is calculated. The comparison result matrix is ​​generated and the matching, change, or insertion status of the fields is automatically annotated.

[0028] Among them, the graph embedding model formula is: h i =f(W·A·h i-1 +b);

[0029] Where h i represents the embedding vector of node i, h i-1 represents the embedding vector of node i-1 in the previous layer; A is the adjacency matrix of the archive field graph; W is the weight matrix; b is the bias term; f is the activation function;

[0030] The field change similarity formula is:

[0031] Where h j represents the embedding vector of node j; Sim(i,j) represents the change similarity between node i and node j;

[0032] When Sim(i,j) is greater than the preset threshold, it is judged as a matching field, otherwise it is changed;

[0033] Use density clustering algorithms to cluster field states in change chains and automatically identify common change patterns;

[0034] Use dynamic network diagrams to represent the change relationship between fields. Changes to multiple fields by the same subject are represented by dynamic edges.

[0035] Use heatmaps to show how often fields are changed.

[0036] As a preferred solution of the present invention, the file sharing module includes:

[0037] A directory index generation unit is used to generate a multi-level directory index based on the structured data of the enterprise registration file and organize and display it in a tree-like information form;

[0038] The archival image positioning unit is used to quickly locate and retrieve archival images based on the file directory and the file directory within the volume, combining OCR technology and super-resolution generative adversarial networks;

[0039] The archive anti-counterfeiting processing unit is used to perform comprehensive anti-counterfeiting processing on shared archive images using electronic signatures, watermarks, and dynamic QR codes to ensure the authenticity and tamper-proofness of shared archives;

[0040] The quick query and authorization unit is used for interactive multi-level index query of enterprise registration archives, dynamically authorizes access to shared archives based on user permissions, and provides archive sharing function when the authorization conditions are met.

[0041] As a preferred solution of the present invention, the shareholder right to know protection module includes:

[0042] The share tracking unit is used to track the status and change records of corporate shares based on graph neural networks and blockchain technology, ensuring the real-time update and transparency of all share information;

[0043] The equity status display unit is used to generate shareholder shareholding ratios and share change history charts through dynamic data visualization technology to intuitively display the equity status to shareholders;

[0044] The real-time update notification unit is used to trigger the notification mechanism through the blockchain when there is a change in the company's registration information or the status of shareholders' shares, and send real-time update reminders to shareholders.

[0045] As a preferred solution of the present invention, the steps for implementing the share tracking unit are as follows:

[0046] Define dynamic graph: Use dynamic graph G t =(V t ,E t ,W t ) represents the time series model of equity relationship, where V t is the node set at time t, E t is the edge set at time t, W t is the edge weight at time t;

[0047] Add a time attribute to each node and edge: in is the initial eigenvector of node v; Init(v) represents the function that initializes the initial characteristics of node v and generates the eigenvector related to the node; Represents the feature concatenation operation; TimeEmbed(t) is the time embedding vector;

[0048] Incremental update of equity change: When the blockchain records equity changes, it dynamically modifies G t, add new nodes or modify edge weights;

[0049] Aggregation formula optimization: When aggregating neighbor features, introduce time weight α and edge weight W e The joint weighted: where m v is the aggregation result of neighbor features of node v; N(v) is the set of neighbor nodes of node v; α t is the time weight; is the feature representation of neighbor node u in the lth layer of the graph neural network;

[0050] Node feature update: Use the optimized neighbor feature aggregation formula: in, is the feature vector of node v in the l+1 layer; is the feature vector of node v in layer l; Q is the weight matrix; k is the bias vector; σ is the activation function;

[0051] Tiered storage: Dynamic graphs are stored in layers by time, with recently changed nodes and edges stored in high-priority layers and historical data stored in low-priority layers. Distributed storage technology is used to manage data tiers.

[0052] Index optimization: Create a time index for each node and edge to support quick search of equity relationships at a specified time point:

[0053] Index(v,t)={edges(v)|t start ≤t≤t end};

[0054] In the formula, Index(v,t) is the edge index set of node v at time t; edges(v) represents the edge set connected to node v; t start is the starting time of the time range, t end The end time of the time range.

[0055] As a preferred solution of the present invention, the comprehensive statistics module includes:

[0056] Multidimensional data analysis unit, used to perform statistical analysis on archival data based on the query person, company directory, query content and query time;

[0057] Report generation unit, used to generate various forms of reports based on statistical results, including charts and tables, and supports exporting to PDF or Excel files;

[0058] The dynamic statistics display unit is used to display statistical data in real time through a visual dashboard and supports dynamic adjustment of statistical conditions.

[0059] As a preferred solution of the present invention, the system security module includes:

[0060] The confidentiality control unit is used to dynamically control the user's query permissions according to the confidentiality level of the archive, ensuring that users of different levels can only access the archive data within their authority range;

[0061] The security log management unit is used to record user login, query, sharing and other operation logs, and ensure the logs are tamper-proof based on blockchain technology;

[0062] The abnormal behavior detection unit is used to analyze user behavior through machine learning algorithms, automatically detect and mark abnormal operations, and issue security alerts to administrators.

[0063] As a preferred solution of the present invention, the distributed key management system includes:

[0064] Key generation unit, used to generate dynamic keys based on distributed key management technology, supporting user authentication and encrypted storage of archival data;

[0065] Key distribution unit, used to securely distribute dynamic keys to different modules to support file access, sharing and encrypted transmission;

[0066] The key update and recovery unit is used to dynamically update keys when user permissions change, and to securely recover keys that are no longer in use to avoid key leakage.

[0067] The beneficial effects of the present invention are: using blockchain technology to achieve full-chain traceability of archival data, completely recording all change operations, significantly improving the transparency of archival management, and ensuring the authenticity and credibility of data; multi-level directory indexing and interactive query functions make archival queries fast and convenient, while ensuring the security and authenticity of shared archives through electronic signatures, watermarks and QR code anti-counterfeiting technologies; shareholders can track the company's equity status and change records in real time, and promptly understand corporate change information through a dynamic notification mechanism, fully protecting shareholders' right to know; through dynamic encryption, authority management and abnormal behavior detection, the security of archival storage and access is ensured, and data leakage and tampering are effectively prevented; support for multi-dimensional data analysis and report generation provides strong data support for corporate management and decision-making; dynamic allocation of user permissions through an efficient identity authentication module ensures that different users only access data within their authorized scope. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0069] in:

[0070] Figure 1 It is a modular structure diagram of the system of the present invention;

[0071] Figure 2 This is a structural diagram of a comprehensive statistics module in an embodiment of the present invention;

[0072] Figure 3 2 is a structural diagram of the system security module in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0074] like Figure 1-Figure 3 FIG. 1 is an embodiment of the present invention, which provides a system for sharing enterprise registration files and protecting shareholders' right to know based on full-chain data tracing and auditing, including:

[0075] (1) Identity authentication module

[0076] It is used to realize user identity authentication for public users, government agency users and contracted users through the distributed key management system combined with the CA authentication platform and the electronic business license platform, and to determine whether the inquirer is a shareholder through the interface of the natural person identity authentication platform.

[0077] The distributed key management system includes:

[0078] Key generation unit, used to generate dynamic keys based on distributed key management technology, supporting user authentication and encrypted storage of archival data;

[0079] Key distribution unit, used to securely distribute dynamic keys to different modules to support file access, sharing and encrypted transmission;

[0080] The key update and recovery unit is used to dynamically update keys when user permissions change, and to securely recover keys that are no longer in use to avoid key leakage.

[0081] In one embodiment, the identity authentication module includes:

[0082] The identity credential verification unit is used to receive identity credentials submitted by users, including CA certificates, electronic business licenses, or natural person identity information, verify the legitimacy of the CA certificate by connecting to the CA authentication platform, verify the validity of the business license by connecting to the electronic business license platform, and call the natural person identity authentication platform to verify whether the user is a legitimate corporate shareholder;

[0083] The permission allocation unit is used to dynamically generate the user's access permission information based on the identity credential verification result, including the user's identity category, permission level and the data range that the user can access;

[0084] A distributed key generation unit, which is used to generate dynamic keys for user authentication and permission information based on a distributed key management system to support subsequent data encryption and decryption operations;

[0085] The verification result output unit is used to output the user identity authentication result, including user identity authentication status, user role type, user authority level, authentication timestamp, verification source, and whether it matches the enterprise information.

[0086] User authentication status: Whether the authentication has passed, with the output value being "successful" or "failed." For example, the CA certificate or electronic business license verifies whether it matches the legitimate user information in the system database. If the authentication result is a failure, the user is denied access to subsequent modules, and an error message is returned to the user terminal. The failed authentication log is recorded for tracking by the audit module.

[0087] User role type: This verifies the user's identity category, such as corporate shareholder, corporate legal representative, government agency user (such as the public security, procuratorial and judicial organs), and public user. The role type determines the scope of data that the user can access.

[0088] User permission level: Verifies the user's permission level in the system, such as: ordinary user permission (can only access public corporate information), shareholder user permission (can view share information and corporate registration files), high-level permission user (such as corporate legal person, management personnel, can operate sensitive data). The output value is the permission identifier (such as level 1, level 2, level 3, etc., or a specific permission list);

[0089] Authentication timestamp: records the authentication time for logging and subsequent tracking. Example: 2024-12-28 10:15:00;

[0090] Verification source: Indicate the method used for verification (such as CA certificate, electronic business license, liveness authentication, etc.), as well as the platform source of the verification (such as the National Unified Identity Authentication Platform);

[0091] Whether the enterprise information matches (for shareholder or legal representative verification): Check whether the user identity matches the enterprise information stored in the system, whether the user belongs to the target enterprise (such as shareholder, legal person, etc.), and whether the enterprise registration information in the integrated data platform is linked.

[0092] The authentication result is output as structured data, which is convenient for other modules to use. The following is an example output:

[0093] {

[0094] "status":"Success",

[0095] "user_role":"Enterprise Shareholder",

[0096] "user_permissions":["View share information","Query company registration files"],

[0097] "timestamp":"2024-12-28 10:15:00",

[0098] "verification_source":"CA certification platform",

[0099] "company_match": "Match successful",

[0100] "company_name":"ABC Technology Co., Ltd."

[0101] }

[0102] (2) Archive storage management module

[0103] Used to realize the storage, management, traceability and auditing of enterprise registration files based on blockchain technology and distributed storage technology.

[0104] Specifically, the archive storage management module includes:

[0105] The data storage unit is used for the storage and management of enterprise registration archives. It uses distributed storage technology to store the archival image data (such as scanned documents and PDF files), associated information, and multi-level index data of the enterprise registration archives. The metadata of the enterprise registration archives (such as index information, access logs, and change records) is stored on the blockchain to ensure the integrity and immutability of the archive data. The metadata storage content includes: timestamp, operation subject, operation type, data hash value, etc. The stored data is dynamically encrypted using a distributed key management system (DMS) to ensure that only authorized users can access the archive data. The encryption strategy is dynamically adjusted based on user permissions to ensure the privacy and security of the stored data.

[0106] Association information can be understood as the association information between enterprise registration archive data and other relevant data. This information is usually used to support the system's retrieval, management and traceability functions, mainly including the logical association of archive content, the association between archives and external data, the association of archive management information, the upstream and downstream association of archive data, the extended association of archive indexes, etc.

[0107] The logical association of file contents refers to the association information between different data within the enterprise registration file, which is used to describe the logical structure and content relationship of the enterprise file:

[0108] Basic company information: such as company name, unified social credit code, registered address, date of establishment, etc.;

[0109] Shareholder information: shareholder name, shareholding ratio, shareholder change record, etc.;

[0110] Business information: business scope, business license information, permit information, etc.

[0111] Historical change information: change records and time of company name, registered address, legal representative, shareholder structure, and business scope;

[0112] Archival image index: The relationship between archival files and corresponding image data, such as the mapping of file directories to specific image files.

[0113] The association between archives and external data refers to the association information between enterprise registration archives and other systems or external data resources, which is used to achieve cross-system data sharing or expand functions:

[0114] Industrial and commercial data: the correspondence between enterprise registration files and data in the industrial and commercial registration database;

[0115] Tax data: the association of corporate tax records or tax registration information with archival data;

[0116] Judicial records: litigation information related to the enterprise, court judgment records, etc.;

[0117] Regulatory information: inspection reports and penalty records shared with market regulatory authorities.

[0118] The association of archive management information refers to metadata related to archive storage and operations, which is used to support the process and traceability of archive management:

[0119] Archive version record: describes different versions of archive files (such as revised or changed versions) and their timestamps;

[0120] Access control information: stores the association between access rights of archive files and user identities, such as archive level (public, restricted, confidential);

[0121] Operation records: modification records of archive data, operation subjects, modification time, etc.;

[0122] Cryptographic information: Key distribution and encryption logic associated with a distributed key management system.

[0123] The upstream and downstream association of archival data refers to the upstream and downstream data relationship of enterprise registration archives in the business process:

[0124] Business data association: enterprise files and their associated business process data, such as enterprise qualification approval, administrative licensing process, etc.

[0125] Shareholder relationship information: information on each shareholder's relationships with other companies or individuals in the company registration file (such as other companies held shares, control relationship diagrams, etc.);

[0126] Inter-enterprise relations: information about the relationship between an enterprise and its affiliated enterprises (such as parent companies, subsidiaries, branches, cooperative enterprises, etc.). Extended association of archive indexes refers to auxiliary information generated by expanding archive data in order to improve retrieval efficiency and sharing capabilities:

[0127] Full-text index information: the association between text content generated by OCR technology and the corresponding archival images;

[0128] Multi-level directory structure: organization and index mapping of archives at different levels (such as enterprises, shareholders, subsidiaries, etc.);

[0129] Search keyword association: The correspondence between keywords and specific file content established based on user search behavior.

[0130] The data tracing unit is used to record the change time, operating subject and specific change content of the archive based on the blockchain timestamp function, and conduct full-chain tracing of the change records, user operation logs and data access records of the enterprise registration archive. It supports the tracing of archive information at any time point and generates a change chain to display all historical versions of the archive data. It supports the display of tracing results in a visual form (such as a timeline, change chart or chain structure), including the time, operating subject and specific change content of the archive data change;

[0131] In one preferred embodiment, the data tracing unit performs pattern matching and change chain analysis of archive change records based on an improved multiple sequence alignment algorithm, specifically including:

[0132] Introducing knowledge graph-based archival data modeling: When formatting archival data, knowledge graph technology is used to construct semantic associations within the data. Business registration information, shareholder structure, and change records are modeled as directed graphs of entities and relationships, with nodes representing fields and edges representing semantic relationships between fields. The fields of each archival version are structured as a graph model rather than a linear sequence (combined with the MSA algorithm) to capture potential connections between fields. For example, whether a change to one field will trigger simultaneous changes in other fields. Knowledge graphs can enhance the ability to understand the semantic information of archival data, optimize in-depth analysis of pattern matching, and provide more intuitive data structure support for subsequent change chain analysis.

[0133] Combining graph embedding and GNN (graph neural network) technology: The multi-version archive comparison process is expanded from traditional dynamic programming to an efficient comparison method based on graph embedding (Graph Embedding) and Graph Neural Network (GNN). The formatted archive field graph is input into the graph embedding model, and the graph nodes (fields) are mapped to fixed-dimensional feature vectors. The GNN learns the feature similarity of archive fields across multiple versions and calculates the change similarity of each field. The resulting comparison matrix is ​​generated, automatically annotating fields as matched, changed, or inserted. GNN can leverage the structured relationships between fields to improve the efficiency and accuracy of multiple sequence alignments. The feature vectors of the graph embedding model can be used as input for change pattern analysis, providing a global perspective across versions.

[0134] Among them, the graph embedding model formula is: h i =f(W·A·h i-1 +b);

[0135] Where h i represents the embedding vector of node i, h i-1represents the embedding vector of node i-1 in the previous layer; A is the adjacency matrix of the archive field graph; W represents the learnable weight matrix; b is the bias term; f is the activation function;

[0136] The field change similarity formula is:

[0137] Where h j represents the embedding vector of node j; Sim(i,j) represents the change similarity between node i and node j;

[0138] When Sim(i,j) is greater than the preset threshold, it is judged as a matching field, otherwise it is changed.

[0139] Use clustering algorithms to identify change patterns: Use the density-based spatial clustering algorithm (DBSCAN, Density-Based Spatial Clustering of Applications with Noise) to cluster field states in the change chain and automatically identify common change patterns. For example, multiple changes to a field may be clustered into the same pattern, and the clustering results are used to identify abnormal patterns.

[0140] The specific steps are: represent the field status of the change chain with a vector:

[0141] x i =[change time, operation subject characteristics, change type];

[0142] The change status of each field is represented by a three-dimensional vector;

[0143] Calculate the Euclidean distance between changed states:

[0144] Cluster change chains and output common patterns.

[0145] Clustering algorithms can automatically discover high-frequency change patterns, providing a foundation for summarizing the regularity of file changes. Combined with anomaly annotation mechanisms, they can identify abnormal change behaviors and improve the intelligence of the system.

[0146] Combine dynamic network diagrams and heat maps to display change chains: Use dynamic network diagrams to represent the change relationships between fields. For example, changes to multiple fields by the same entity are represented by dynamic edges. Use heat maps to display the frequency of field changes, for example, light and dark colors represent the number of changes.

[0147] Tool support: Visualization tools such as D3.js or Cytoscape.js.

[0148] Dynamic network diagrams facilitate intuitive display of change relationships, and heat maps can quickly identify frequently changed fields.

[0149] The data audit unit is used to record and audit all archival data access operations in the system in real time (including user ID, operation time, operation type (such as query, change, export) and specific archival content accessed), perform permission verification on archival operation behaviors, and review whether the user's archival operations are within the scope of their authority. For example, shareholder users can only access information related to their own shares; and mark and record unauthorized operations or abnormal behaviors (such as unauthorized operations), generate audit logs, and provide complete user operation audit information (including access frequency, abnormal behavior statistics and operation log lists) to the system's risk management module for further security assessment and analysis.

[0150] (3) File sharing module

[0151] It is used for multi-level directory index tree information display and quick query of enterprise registration archives, automatic retrieval of archive information, interactive multi-level index query, and quick positioning of archive images based on file directories and file directories within the volume. It also performs anti-counterfeiting processing of archive images with electronic signatures, watermarks and QR codes under authorized conditions, and shares enterprise registration archive information.

[0152] In one embodiment, the file sharing module includes:

[0153] A directory index generating unit is used to generate a multi-level directory index based on the structured data of the enterprise registration file and organize and display it in a tree-like information form;

[0154] The archival image positioning unit is used to quickly locate and retrieve archival images based on the file directory and the file directory within the volume, combining OCR technology and super-resolution generative adversarial networks;

[0155] Optionally, the archival image locating unit may be implemented as follows:

[0156] The target archival images retrieved from the archival storage system are formatted and resized to ensure compatibility between the input images and the deep learning model. The standardized low-resolution images are used as input for subsequent processing.

[0157] By introducing the super-resolution generative adversarial network (SRGAN), the quality of low-resolution archival images is enhanced and details are reconstructed. The self-attention mechanism is combined with the generator network to improve the modeling ability of the global features of the image and enhance the restoration effect of complex textures and detailed areas; the discriminator network optimizes the perceptual quality of the generated image through adversarial training, making the enhanced image closer to the real high-resolution image; the multi-scale fusion of perceptual loss, adversarial loss and pixel loss further improves the image restoration and ensures the clarity and realism of the archival image.

[0158] Combined with Fourier transform, frequency domain loss is introduced to optimize high-frequency details of the image (such as character boundaries and image borders), improving the image's ability to restore texture and fine areas. It is particularly suitable for processing old, blurred or low-quality archival images.

[0159] By combining the super-resolution generation task with the OCR text recognition task through a multi-task learning architecture, the model not only reconstructs the image but also enhances the restoration accuracy of the text area, improving the accuracy of subsequent OCR analysis.

[0160] After image enhancement, OCR technology is used to extract textual information from archival images. Combined with the semantic understanding capabilities of the Transformer model, it intelligently matches fuzzy query conditions entered by the user to quickly locate the target archival image. Similarity calculations (such as cosine similarity or edit distance) are used to sort the extracted textual information and query conditions to ensure the accuracy of the positioning results.

[0161] A dynamic weight allocation mechanism is introduced in super-resolution model training to adaptively adjust the proportion of perceptual loss, pixel loss, and adversarial loss according to the characteristics of the input image, ensuring the performance stability and adaptability of the model in diverse scenarios (such as high-noise, low-resolution images);

[0162] Combining multi-task learning and semantic understanding technology, the unit can quickly locate the target archive of the user's query and generate high-resolution images for display; it supports a variety of interactive functions, such as image zooming in, zooming out, rotating and area labeling, to facilitate further processing by users.

[0163] The archive anti-counterfeiting processing unit is used to perform comprehensive anti-counterfeiting processing on shared archive images using electronic signatures, watermarks, and dynamic QR codes to ensure the authenticity and tamper-proofness of shared archives;

[0164] The quick query and authorization unit is used for interactive multi-level index query of enterprise registration archives, dynamically authorizes access to shared archives based on user permissions, and provides archive sharing function when the authorization conditions are met.

[0165] (4) Shareholders’ right to know protection module

[0166] It is used to provide share ownership tracking through blockchain technology, check the status of corporate shares and updates of corporate registration information, and ensure shareholders' right to know.

[0167] Specifically, the shareholder right to know protection module includes:

[0168] The share tracking unit is used to track the status and change records of corporate shares based on blockchain technology, ensuring real-time update and transparency of all share information;

[0169] The equity status display unit is used to generate shareholder shareholding ratios and share change history charts through dynamic data visualization technology to intuitively display the equity status to shareholders;

[0170] The real-time update notification unit is used to trigger the notification mechanism through the blockchain when there is a change in the company's registration information or the status of shareholders' shares, and send real-time update reminders to shareholders.

[0171] Furthermore, the steps for implementing the share tracking unit are as follows:

[0172] 1) Dynamic graph modeling

[0173] Define dynamic graph: Use dynamic graph G t =(V t ,E t ,W t ) represents the time series model of equity relationship, where V t is the node set at time t, E t is the edge set at time t, W t is the edge weight (shareholding ratio) at time t;

[0174] Add a time attribute to each node and edge: in is the initial feature vector of node v, which represents the feature representation of the node at the time of graph neural network initialization, combining the characteristics of the node itself and time embedding; Init(v) represents the function that initializes the initial characteristics of node v and generates a feature vector related to the node. For example, for a shareholder node, the initial characteristics can be the type of shareholder (natural person or legal person), total shareholding ratio, etc.; for an enterprise node, the characteristics can be the enterprise market value, number of shareholders, etc. It represents the feature concatenation operation, which combines the initial feature vector and time embedding vector of the node to form a new feature representation; TimeEmbed(t) is the time embedding vector used to represent time information.

[0175] Incremental update of equity change: When the blockchain records equity changes, it dynamically modifies G t , add new nodes (new shareholders or enterprises); modify edge weights (changes in shareholding ratio).

[0176] 2) Weighted aggregation strategy

[0177] Aggregation formula optimization: When aggregating neighbor features, introduce time weight α and edge weight W e The joint weighted: where m v is the neighbor feature aggregation result (message vector) of node v, which represents the information aggregated from all neighbor nodes of node v; N(v) is the set of neighbor nodes of node v, which represents other nodes connected to node v; αt is the time weight, which indicates the degree to which the importance of neighbor node features decays over time (the farther away from the current time, the lower the weight); is the feature representation of neighbor node u in the lth layer of the graph neural network.

[0178] Node feature update: Use the optimized neighbor feature aggregation formula:

[0179] Where, is the feature vector of node v at layer l+1, which combines its own features and the aggregation results of neighbor features; is the feature vector of node v in layer l; Q is a trainable weight matrix used for linear transformation to convert the concatenated feature vector into a new feature representation; k is the bias vector; σ is the activation function;

[0180] 3) Hierarchical storage and indexing mechanism

[0181] Tiered storage: Dynamic graphs are stored in layers by time, with recently changed nodes and edges stored in a high-priority layer (memory) and historical data stored in a low-priority layer (disk). Distributed storage technology (such as IPFS) is used to manage data tiers.

[0182] Index optimization: Create a time index for each node and edge to support quick search of equity relationships at a specified time point:

[0183] Index(v,t)={edges(v)|t start ≤t≤t end};

[0184] In the formula, Index(v,t) is the edge index set of node v at time t, which means that the edge is connected to node v and is in the time range [t start ,t end ]; edges(v) represents the set of edges connected to node v; t start is the starting time of the time range, t end The end time of the time range.

[0185] Dynamic graph modeling is used to more accurately capture the time series characteristics of equity changes; the weighted aggregation strategy effectively balances the impact of time and equity weight, and the hierarchical storage and indexing mechanism significantly improves query efficiency.

[0186] (5) Comprehensive statistics module

[0187] Used to collect statistics on archive query and sharing data based on the query person, enterprise directory, query content and query time, and generate various forms of reports;

[0188] The comprehensive statistics module includes:

[0189] Multidimensional data analysis unit, used to perform statistical analysis on archival data based on the query person, company directory, query content and query time;

[0190] Report generation unit, used to generate various forms of reports based on statistical results, including charts and tables, and supports exporting to PDF or Excel files;

[0191] The dynamic statistics display unit is used to display statistical data in real time through a visual dashboard and supports dynamic adjustment of statistical conditions.

[0192] (6) System security module

[0193] It is used to provide file classification control, user rights management and full-process security log recording functions to ensure the security of file query and sharing;

[0194] Specifically, the system security module includes:

[0195] The confidentiality control unit is used to dynamically control the user's query permissions according to the confidentiality level of the archive, ensuring that users of different levels can only access the archive data within their authority range;

[0196] The security log management unit is used to record user login, query, sharing and other operation logs, and ensure the logs are tamper-proof based on blockchain technology;

[0197] The abnormal behavior detection unit is used to analyze user behavior through machine learning algorithms, automatically detect and mark abnormal operations, and issue security alerts to administrators.

[0198] In summary, the present invention uses blockchain technology to achieve full-chain traceability of archival data, completely record all change operations, significantly improve the transparency of archival management, and ensure the authenticity and credibility of data; multi-level directory indexing and interactive query functions make archival queries fast and convenient, while electronic signatures, watermarks and QR code anti-counterfeiting technologies are used to ensure the security and authenticity of shared archives; shareholders can track the company's equity status and change records in real time, and promptly understand corporate change information through a dynamic notification mechanism, fully protecting shareholders' right to know; through dynamic encryption, authority management and abnormal behavior detection, the security of archive storage and access is ensured, and data leakage and tampering are effectively prevented; support for multi-dimensional data analysis and report generation provides strong data support for corporate management and decision-making; dynamic allocation of user permissions through an efficient identity authentication module ensures that different users only access data within their authorized scope.

[0199] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing is characterized by: The system comprises: The identity authentication module is used to implement user identity authentication for public users, government agency users, and contracted users through a distributed key management system combined with a CA authentication platform and an electronic business license platform, and to determine whether the inquirer is a shareholder through an interface with a natural person identity authentication platform; The archive storage management module is used to implement the storage, management, traceability and auditing of enterprise registration archives based on blockchain technology and distributed storage technology; The archive sharing module is used for multi-level directory index tree-like information display and quick query of enterprise registration archives, automatic retrieval of archive information, interactive multi-level index query, and quick location of archive images based on the case file directory and the file directory within the volume. Under authorized conditions, archive images are processed with electronic signatures, watermarks, and QR codes for anti-counterfeiting, and enterprise registration archive information is shared; The shareholder right to know protection module is used to provide share ownership tracking through blockchain technology, check the status of corporate shares and the update of corporate registration information, and ensure shareholders' right to know; Comprehensive statistics module, used to collect statistics on archive query and sharing data based on the query person, enterprise directory, query content and query time, and generate various forms of reports; System security module, which is used to provide file classification control, user rights management and full-process security log recording functions to ensure the security of file query and sharing; The archive storage management module includes a data tracing unit, which is used to trace the change records, user operation logs and data access records of enterprise registration archives based on the blockchain timestamp function. It supports the tracing of archive information at any time point, generates a change chain, and displays all historical versions of archive data. It supports the visualization of tracing results, including the time of archive data change, the operating entity and the specific content of the change; The data tracing unit performs pattern matching and change chain analysis of archive change records based on an improved multiple sequence alignment algorithm, specifically including: When formatting archival data, we use knowledge graph technology to construct semantic associations within archival data. We model enterprise registration information, shareholder structure, and change records as directed graphs of entities and relationships. Nodes represent fields, and edges represent semantic relationships between fields. The fields of each archival version are structured into a graph model. The multi-version archive comparison process is expanded to a comparison based on graph embedding and graph neural networks. The formatted archive field graph is input into the graph embedding model, and the graph nodes are mapped into feature vectors of fixed dimensions. The feature similarity of the multi-version archive fields is learned through GNN, and the change similarity of each field is calculated. The comparison result matrix is ​​generated and the matching, change, or insertion status of the fields is automatically annotated. Among them, the graph embedding model formula is: h i =f(W·A·h i-1 +b); Where h i represents the embedding vector of node i, h i-1 represents the embedding vector of node i-1 in the previous layer; A is the adjacency matrix of the archive field graph; W is the weight matrix; b is the bias term; f is the activation function; The field change similarity formula is: Where h j represents the embedding vector of node j; Sim(i,j) represents the change similarity between node i and node j; When Sim(i,j) is greater than the preset threshold, it is judged as a matching field, otherwise it is changed; Use density clustering algorithms to cluster field states in change chains and automatically identify common change patterns; Use dynamic network diagrams to represent the change relationships between fields, and use dynamic edges to represent changes to multiple fields by the same entity. Use heatmaps to show how often fields are changed.

2. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized by: The identity authentication module includes: The identity credential verification unit is used to receive identity credentials submitted by users, including CA certificates, electronic business licenses, or natural person identity information, verify the legitimacy of the CA certificate by connecting to the CA authentication platform, verify the validity of the business license by connecting to the electronic business license platform, and call the natural person identity authentication platform to verify whether the user is a legitimate corporate shareholder; The permission allocation unit is used to dynamically generate the user's access permission information based on the identity credential verification result, including the user's identity category, permission level and the data range that the user can access; A distributed key generation unit, which is used to generate dynamic keys for user authentication and permission information based on a distributed key management system to support subsequent data encryption and decryption operations; The verification result output unit is used to output the user identity authentication result, including user identity authentication status, user role type, user authority level, authentication timestamp, verification source, and whether it matches the enterprise information.

3. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized by: The archive storage management module also includes: The data storage unit is used for the storage and management of enterprise registration files. It uses distributed storage technology to store the archival image data, associated information, and multi-level index data of enterprise registration files. The metadata of enterprise registration files is stored on the blockchain. The stored data is dynamically encrypted using a distributed key management system. The data audit unit is used to record and audit all archival data access operations in the system in real time through blockchain smart contracts, verify the permissions of archival operations, mark and record unauthorized operations or abnormal behaviors, and generate audit logs.

4. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized in that: The file sharing module includes: A directory index generation unit is used to generate a multi-level directory index based on the structured data of the enterprise registration file and organize and display it in a tree-like information form; The archival image positioning unit is used to quickly locate and retrieve archival images based on the file directory and the file directory within the volume, combining OCR technology and super-resolution generative adversarial networks; The archive anti-counterfeiting processing unit is used to perform comprehensive anti-counterfeiting processing on shared archive images using electronic signatures, watermarks, and dynamic QR codes to ensure the authenticity and tamper-proofness of shared archives; The quick query and authorization unit is used for interactive multi-level index query of enterprise registration archives, dynamically authorizes access to shared archives based on user permissions, and provides archive sharing function when the authorization conditions are met.

5. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized in that: The shareholder right to know protection module includes: The share tracking unit is used to track the status and change records of corporate shares based on graph neural networks and blockchain technology, ensuring the real-time update and transparency of all share information; The equity status display unit is used to generate shareholder shareholding ratios and share change history charts through dynamic data visualization technology to intuitively display the equity status to shareholders; The real-time update notification unit is used to trigger the notification mechanism through the blockchain when there is a change in the company's registration information or the status of shareholders' shares, and send real-time update reminders to shareholders.

6. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 5 is characterized by: The steps for implementing the share tracking unit are as follows: Define dynamic graph: Use dynamic graph G t =(V t ,E t ,W t ) represents the time series model of equity relationship, where V t is the node set at time t, E t is the edge set at time t, W t is the edge weight at time t; Add a time attribute to each node and edge: in is the initial eigenvector of node v; Init(v) represents the function that initializes the initial characteristics of node v and generates the eigenvector related to the node; Represents feature concatenation operation; TimeEmbed(t) is the time embedding vector; Incremental update of equity change: When the blockchain records equity changes, it dynamically modifies G t , add new nodes or modify edge weights; aggregation formula optimization: when aggregating neighbor features, introduce time weight α and edge weight W e The joint weighted: where m v is the aggregation result of neighbor features of node v; N(v) is the set of neighbor nodes of node v; α t is the time weight; is the feature representation of neighbor node u in the lth layer of the graph neural network; Node feature update: Use the optimized neighbor feature aggregation formula: in, is the feature vector of node v in the l+1 layer; is the feature vector of node v in layer l; Q is the weight matrix; k is the bias vector; σ is the activation function; Tiered storage: Dynamic graphs are stored in layers by time, with recently changed nodes and edges stored in high-priority layers and historical data stored in low-priority layers. Distributed storage technology is used to manage data tiers. Index optimization: Create a time index for each node and edge to support quick search of equity relationships at a specified time point: Index(v,t)={edges(v)|t start ≤t≤t end }; In the formula, Index(v,t) is the edge index set of node v at time t; edges(v) represents the edge set connected to node v; t start is the starting time of the time range, t end The end time of the time range.

7. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized by: The comprehensive statistics module includes: Multidimensional data analysis unit, used to perform statistical analysis on archival data based on the query person, company directory, query content and query time; Report generation unit, used to generate various forms of reports based on statistical results, including charts and tables, and supports exporting to PDF or Excel files; The dynamic statistics display unit is used to display statistical data in real time through a visual dashboard and supports dynamic adjustment of statistical conditions.

8. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized by: The system security module includes: The confidentiality control unit is used to dynamically control the user's query permissions according to the confidentiality level of the archive, ensuring that users of different levels can only access the archive data within their authority range; The security log management unit is used to record the user's login, query, and sharing operation logs, and ensures the logs are tamper-proof based on blockchain technology; The abnormal behavior detection unit is used to analyze user behavior through machine learning algorithms, automatically detect and mark abnormal operations, and issue security alerts to administrators.

9. The enterprise registration file sharing and shareholder right to know protection system based on full-chain data tracing and auditing as claimed in claim 1 is characterized by: The distributed key management system includes: Key generation unit, used to generate dynamic keys based on distributed key management technology, supporting user authentication and encrypted storage of archival data; Key distribution unit, used to securely distribute dynamic keys to different modules to support file access, sharing and encrypted transmission; The key update and recovery unit is used to dynamically update keys when user permissions change, and to securely recover keys that are no longer in use to avoid key leakage.

Images