Enterprise data asset classification management method and system

Abstract

The application discloses a kind of enterprise data asset classification management method, system, belong to data governance and data asset management technical field.The present application is aimed at the problems such as fragmentation of classification system in prior art, insufficient intelligent coverage, label transfer is easy to drift, disjointed with asset accounting, static management is easy to fail, etc., constructs the unified classification rule system of multidimensional fusion of business, security, right ownership, accounting, realizes global data intelligent marking and confidence level grading review using rule engine and semantic recognition combination, realizes cross-system label inheritance and differential management and control through label blood relationship transfer mechanism, built-in accounting attribute mapping supports asset operation, and establishes dynamic audit closed loop to realize continuous iteration.The present application can effectively unify classification yardstick, improve management efficiency, guarantee control landing, support enterprise data asset compliance operation and value release.

Description

Technical Field

[0001] This invention belongs to the field of data governance and data asset management technology, and in particular relates to a method and system for classifying and managing enterprise data assets. Background Technology

[0002] With the deepening of the digital economy and enterprise digital transformation, data has become an important production factor and intangible asset for enterprises. Enterprise data asset classification and management is a core foundation of the data governance system. By classifying, organizing, and tagging massive amounts of enterprise data resources, it provides support for data security protection, compliance control, sharing and reuse, and asset-based operation.

[0003] However, existing enterprise data asset classification management methods and systems still have many shortcomings in practical applications:

[0004] (1) Fragmented classification system and conflicting multiple standards: The business, security, legal and financial departments of enterprises often establish their own classification standards, and the classification results of the same data asset are inconsistent in different systems; the classification rules of multiple subsidiaries of group enterprises are not unified, and it is impossible to achieve unified management and consolidated statistics of data assets across organizations.

[0005] (2) Low level of intelligence and limited coverage: Most existing classifications rely on manual sorting and regular rule matching, which is inefficient, costly and difficult to adapt to the massive growth of data; at the same time, the semantic recognition ability of unstructured documents, contracts, drawings, audio and video data is weak, and a large number of data assets are outside the classification management system, forming a management blind spot.

[0006] (3) The lack of a transfer mechanism for labels makes control easily fail: Classification labels only take effect in the original storage system. After data is synchronized across systems, called through interfaces, or exported and copied, labels are lost or arbitrarily modified, causing the classification level of downstream systems to drift, and security control strategies cannot be implemented end-to-end.

[0007] (4) Classification is disconnected from asset value and accounting: Existing classifications are mostly designed around business and security dimensions, without incorporating accounting asset attributes and value dimensions. The classification results cannot directly support data asset entry into the table, asset valuation, pledge financing and data trading, and are difficult to match the development needs of data assetization.

[0008] (5) Static management lacks a closed loop, and the catalog is prone to failure: Classification management is mostly a one-time project-based work, and there is no continuous audit verification and update mechanism after completion; with business iteration and new data access, the asset catalog and actual data gradually become disconnected, forming a "zombie catalog" that cannot continuously support business applications.

[0009] (6) One-size-fits-all management and control strategies, resulting in insufficient value release: either all are protected at the highest level, leading to low sharing efficiency, or excessive openness brings security risks; it is impossible to achieve differentiated management and control based on classification, such as "high value, high protection; low value, high sharing", and it is difficult to balance security compliance and business value release.

[0010] Therefore, there is an urgent need for a method and system for enterprise data asset classification management that can unify multi-dimensional classification standards, achieve intelligent full-domain coverage, ensure cross-system transfer of tags, adapt to asset accounting, and have dynamic closed-loop iteration capabilities, in order to solve the above-mentioned problems in existing technologies. Summary of the Invention

[0011] This invention provides a method and system for classifying and managing enterprise data assets. By constructing a unified classification rule system that integrates multiple dimensions, it achieves automatic tagging of data assets across the entire domain by combining intelligent identification and confidence verification. It realizes end-to-end implementation of control strategies through a tag lineage transfer mechanism, supports data assetization operations through accounting attribute mapping, and achieves continuous iterative optimization through a dynamic audit closed loop. This comprehensively improves the accuracy, efficiency, and value of enterprise data asset classification and management, thus solving the problems in the background technology.

[0012] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0013] The present invention provides a method for classifying and managing enterprise data assets, comprising the following steps:

[0014] S1. Multi-dimensional integrated classification rule construction: Construct an enterprise-level unified classification and grading rule system that integrates business theme dimension, security sensitivity dimension, ownership source dimension, and accounting asset dimension, and configure the judgment criteria, tag fields and control strategy templates for each type of data;

[0015] S2, Automatic Discovery and Intelligent Tagging of Full-Domain Data Assets: Automatically scans the enterprise's full-domain structured, semi-structured, and unstructured data resources, classifies and identifies them through a combination of rule engine and semantic recognition model, outputs classification labels and confidence scores, triggers manual review for data with confidence scores below the threshold, and forms an initial data asset catalog and ledger.

[0016] S3, Tag lineage transfer and differentiated management: Bind the category tags to the lineage of data assets. When data flows across systems and interfaces, it automatically carries and inherits the category tags. Based on the tags, it automatically matches the corresponding level of access permissions, security protection, sharing approval, and lifecycle disposal strategies.

[0017] S4. Asset Value and Accounting Attribute Mapping: Based on the classification results, the value assessment and accounting attribute determination of each type of data asset are carried out, and the corresponding intangible assets, inventory or off-balance sheet disclosed asset categories are mapped to generate an asset ledger subset that is suitable for financial accounting.

[0018] S5. Dynamic Audit and Closed-Loop Iteration: Regularly audit and verify the accuracy of data asset classification, the completeness of labels, and the effectiveness of control implementation; receive label objection feedback from the business side; dynamically update classification rules and asset catalogs; and form a continuously iterative closed-loop management mechanism.

[0019] Furthermore, in step S1, the business subject dimension is divided into three levels: business domain, subject domain, and data entity; the security sensitivity dimension is divided into four levels: public data, internal general data, sensitive data, and core confidential data; the ownership source dimension is divided into four categories: owned data, externally purchased data, cooperative exchange data, and publicly collected data; and the accounting asset dimension is divided into three categories: data that can be included in intangible assets, data that can be included in inventory, and off-balance sheet data resources, based on the accounting treatment regulations for data resources.

[0020] Furthermore, in step S2, the intelligent marking specifically includes:

[0021] The system automatically collects and identifies table names, field names, and comment information from structured data using metadata, and identifies sensitive fields through regular expression matching.

[0022] Extract the main content and sensitive information of unstructured documents, contracts, drawings, audio and video through natural language processing and content semantic recognition;

[0023] For each data asset, a classification label confidence score is generated. Labels with a confidence score higher than the preset high confidence threshold are automatically applied. Labels with a confidence score lower than the high confidence threshold but higher than the low confidence threshold are pushed to the corresponding business owner for review. Labels with a confidence score lower than the low confidence threshold are marked as pending confirmation.

[0024] Furthermore, in step S3, the transfer of tagged lineage specifically includes:

[0025] Generate a unique identifier and category tag metadata for each data asset. During data synchronization, interface calls, export downloads, and copy generation, the category tag is transmitted synchronously along the data lineage.

[0026] When receiving data, the downstream system automatically reads and inherits the classification label. It is forbidden to downgrade or modify the security level without authorization. Label changes must go through an approval process and the labels across the entire chain must be updated synchronously.

[0027] Furthermore, in step S3, the differentiated management strategy includes:

[0028] Access control: Core confidential data is subject to the principle of least privilege and multi-person approval, while public data is accessible to all employees;

[0029] Security protection and control: Highly sensitive data is encrypted and stored, de-identified, and audited; ordinary data is protected with basic security measures.

[0030] Shared workflow management: Automatically match shared approval processes based on classification level; the higher the level, the more approval levels there are.

[0031] Lifecycle management: Different retention periods, archiving rules and destruction mechanisms are configured for different categories of data, and personal information data complies with legal retention requirements.

[0032] Furthermore, in step S4, the accounting attribute determination criteria include: the controllability of data assets, the measurability of costs, and the expected inflow of economic benefits; data assets intended for external sale are mapped to inventory, non-monetary data assets intended for internal use and which are identifiable are mapped to intangible assets, and those that do not meet the recognition criteria are mapped to off-balance-sheet disclosed data resources.

[0033] This invention also provides an enterprise data asset classification management system, comprising:

[0034] Rule configuration module: used to configure multi-dimensional integrated classification and grading rules, tag system, control strategy templates and confidence thresholds;

[0035] Asset discovery and intelligent tagging module: used to automatically scan data resources across the entire domain, classify and identify them through a rule engine and semantic recognition model, generate classification tags with confidence, trigger the manual review process, and build a data asset catalog and ledger;

[0036] Tag Flow and Management Module: Used to bind data lineage and category tags, enabling automatic tag carrying and inheritance across systems, and implementing differentiated permission, security, sharing, and lifecycle management based on tags;

[0037] Asset value mapping module: used to assess the value of data assets and determine their accounting attributes, generate asset ledgers that are adapted to financial accounting, and support asset entry, valuation and transaction;

[0038] Dynamic Audit Iteration Module: Used to periodically perform classification quality audits, receive feedback from business side regarding label objections, and drive dynamic updates of classification rules and asset catalogs.

[0039] Furthermore, the asset discovery and intelligent tagging module includes a structured data recognition unit, an unstructured data recognition unit, a confidence calculation unit, and a manual review unit;

[0040] The structured data recognition unit is used to collect metadata from databases and business systems, and to match business topics with sensitivity levels through a rule engine;

[0041] The unstructured data recognition unit is used to perform content parsing and semantic recognition on documents, contracts, drawings, and audio / video materials, and to extract classification features.

[0042] The confidence calculation unit is used to calculate the confidence of the classification label by combining the rule matching degree and semantic similarity.

[0043] The manual review unit is used to push low-confidence labels to the corresponding data owners for review and confirmation.

[0044] Furthermore, the tag circulation and control module includes a lineage tracking unit, a tag transmission unit, and a strategy execution unit;

[0045] The lineage tracking unit is used to record the entire link flow path and copy relationship of data assets;

[0046] The tag transmission unit is used to synchronously carry classification tag metadata during data flow, which is automatically inherited by downstream systems;

[0047] The strategy execution unit is used to automatically invoke the corresponding access control, encryption and desensitization, approval process, and lifecycle management interfaces based on the category tags.

[0048] The present invention has the following advantages over the prior art:

[0049] (1) Construct a unified classification system that integrates multiple dimensions to eliminate conflicts in the scope of data: By integrating four dimensions—business theme, security sensitivity, ownership source, and accounting assets—a unified classification standard for enterprises is formed, which simultaneously meets the needs of multiple departments such as business operations, security compliance, legal risk control, and financial accounting, supports unified management and hierarchical expansion of the group, and solves the problem of multiple systems running in parallel and inconsistent data scope.

[0050] (2) Intelligent full-domain classification and labeling to improve coverage and efficiency: The hybrid mode of rule engine + semantic recognition + manual review is adopted, which supports the rapid recognition of structured data and has the semantic classification capability of unstructured documents, contracts, drawings and other data; through the confidence level mechanism, the manual cost is greatly reduced while ensuring accuracy, and the full-coverage classification management of full-domain data assets is achieved.

[0051] (3) Tag lineage transfer mechanism to achieve end-to-end control: By binding the classification tags with the data lineage link, the tags are automatically carried and inherited when the data is transferred across systems and interfaces, which fundamentally solves the tag drift problem; at the same time, the tag drives the automatic execution of differentiated control strategies, realizing the closed loop of the whole link of "classification-tag-transfer-control", avoiding the separation of classification and control.

[0052] (4) Built-in accounting and value dimensions to open up the assetization path: accounting asset attributes and value assessment are incorporated into the classification system. The classification results can be directly mapped to intangible assets, inventory and off-balance sheet resources, supporting the accounting of data assets, asset assessment, pledge financing and data trading, and realizing the link from data governance to asset value release.

[0053] (5) Dynamic closed-loop iterative management to ensure the continuous effectiveness of the catalog: Establish a dual-drive update mechanism of regular audit + business feedback, automatically discover new data assets, promptly check and correct classification deviations, dynamically iterate classification rules, avoid the "zombie catalog" problem caused by static one-time management, and ensure the accuracy and timeliness of the asset catalog;

[0054] (6) Differentiated and precise control, balancing security and value: Based on the classification and grading results, differentiated permissions, protection, sharing and lifecycle strategies are matched to achieve "high value, high protection; low value, high sharing", maximizing the sharing and reuse efficiency and business value of data assets while meeting compliance requirements.

[0055] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of the enterprise data asset classification management system architecture described in this invention;

[0058] Figure 2 This is a schematic diagram of the enterprise data asset classification management method described in this invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] In the description of this invention, it should be understood that the terms "relative", "one end", "inner", "lateral", "end", "both ends", "both sides", "front", "one end face", "the other end face", etc., which indicate orientation or positional relationship, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting this invention.

[0061] like Figure 1 As shown, this embodiment provides an enterprise data asset classification management system, including a rule configuration module, an asset discovery and intelligent tagging module, a tag circulation and control module, an asset value mapping module, and a dynamic audit iteration module. Figure 2 As shown, the corresponding enterprise data asset classification management method includes five steps, S1-S5.

[0062] This example is applied to a large manufacturing group enterprise with multiple business domains including R&D, production, supply chain, sales, finance, and human resources. Its data resources encompass structured ERP, CRM, and MES system data, unstructured R&D drawings, contract documents, production videos, and externally sourced industry and supply chain data. The enterprise needs to establish a unified data asset classification and management system that simultaneously meets the requirements for security compliance, data sharing, and asset registration.

[0063] like Figure 2 As shown, the specific implementation process of the present invention will be described in detail below with reference to this scenario:

[0064] Step S1: Construction of multi-dimensional fusion classification rules

[0065] A unified four-dimensional classification and grading rule system is built for the entire group through the rule configuration module:

[0066] 1. Business Theme Dimension: A three-level classification architecture is adopted. The first-level business domain includes R&D, production, supply chain, sales, finance, human resources, operations, etc.; the second-level theme domain is further subdivided under the business domain, such as customer data, order data, channel data, etc. under the sales domain; the third level is the specific data entity, such as customer profiles, sales order details, etc.

[0067] 2. Security Sensitivity Dimension: Divided into four levels: Level 1 is public data (such as publicly available corporate promotional materials), Level 2 is internal general data (such as internal notices and general statistical reports), Level 3 is sensitive data (such as customer data containing personal information and operational data), and Level 4 is core confidential data (such as core formulas, R&D drawings, and core financial data).

[0068] 3. Ownership Source Dimension: Divided into four categories: owned data, externally purchased data, cooperative exchange data, and publicly collected data. Each category is marked with the authorized scope of use and circulation restrictions.

[0069] 4. Accounting Asset Dimension: According to the Ministry of Finance's "Interim Provisions on Accounting Treatment Related to Enterprise Data Resources", it is divided into three categories: data resources that can be included in intangible assets, data resources that can be included in inventory, and off-balance sheet data resources.

[0070] Simultaneously, corresponding control strategy templates are configured for each type of combination, including permission rules, encryption rules, shared approval processes, retention periods, etc. The group maintains the basic rules uniformly, and each subsidiary can expand industry-specific classifications on this basis.

[0071] Step S2: Automatic Discovery and Intelligent Tagging of Full-Domain Data Assets

[0072] The asset discovery and intelligent tagging module connects to all enterprise data sources, including relational databases, unstructured file servers, business systems, and data lakes, performing full-domain scanning and intelligent classification.

[0073] 1. Structured data recognition: Automatically obtains data table names, field names, field comments, and data samples through the metadata collection interface, and matches business theme tags through the rule engine; identifies sensitive fields such as mobile phone numbers, ID card numbers, and bank card numbers through regular expression rules, and automatically matches the security sensitivity level.

[0074] 2. Unstructured data recognition: OCR is used to recognize the text content of drawings and scanned contracts, and natural language processing models are used to extract document themes and entity information to determine business themes and sensitivity levels; for example, when a purchase contract text is recognized, it is automatically classified as supply chain-purchase-contract data and marked as sensitive.

[0075] 3. Confidence Calculation and Manual Review: Classification confidence is calculated based on rule matching and semantic similarity, with a high confidence threshold of 0.85 and a low confidence threshold of 0.6. Tags with a confidence score ≥ 0.85 are automatically applied; tags with a confidence score between 0.6 and 0.85 are pushed to the corresponding business domain's data owner for manual review; tags with a confidence score < 0.6 are marked as assets awaiting confirmation.

[0076] 4. After completing the classification of all data, generate an enterprise data asset catalog and ledger. Each asset record includes fields such as asset ID, business category, security level, ownership source, accounting attribute, responsible department, and storage location.

[0077] Step S3: Labeling of lineage and differentiated management

[0078] The tag circulation and management module enables the end-to-end transmission of category tags and the automatic execution of strategies.

[0079] 1. Lineage Tracking: Generate a unique asset ID for each data asset and record the upstream and downstream flow relationships of data through data lineage technology, including the entire path such as cross-database synchronization, interface output, file export, and copy generation.

[0080] 2. Tag Transfer: When data flows between systems, category tag metadata is transmitted synchronously with the data. Downstream systems automatically parse and inherit the category tags and security levels after receiving the data, without the need for reclassification. For example, after customer data from a CRM system is synchronized to the data platform, it automatically inherits the category tag "Sales-Customer-Sensitive".

[0081] 3. Strategy Execution: Automatically implement differentiated control based on category tags:

[0082] (1) Access control: Core confidential R&D drawings are only authorized to core R&D personnel and require approval from two people; publicly available product information is accessible to all employees.

[0083] (2) Security protection: Sensitive customer data is encrypted and automatically desensitized when provided to external parties; internal ordinary data is only subject to basic access control.

[0084] (3) Shared approval: cross-departmental applications for core confidential data require approval from both the department head and the data governance department; applications for internal general data only require approval within the department.

[0085] (4) Lifecycle: Transaction log data will be automatically destroyed after 3 years; core R&D data will be permanently archived; personal information data will be stored in accordance with legal requirements.

[0086] Label changes require an application, and once approved, the entire process is updated synchronously to prevent unauthorized downgrades.

[0087] Step S4: Mapping Asset Value to Accounting Attributes

[0088] The asset value mapping module assesses the value of data assets based on the classification results, combined with the costs, benefits, and scarcity of the data assets, and completes the mapping of accounting attributes.

[0089] 1. Accounting Attribute Determination: Customer profiles and production operation data collected and processed by the enterprise itself for daily operation and management are determined to be intangible assets; data products and de-identified industry datasets provided to external parties are determined to be inventory; temporary data and public data that do not meet the requirements of controllability or reliable cost measurement are determined to be off-balance sheet disclosure.

[0090] 2. Generate a financial matching ledger: Filter data assets that meet the asset recognition criteria from the master ledger, group them by accounting category, calculate cost and value information, and output a list of data assets that the finance department can directly use for accounting entries.

[0091] 3. At the same time, it marks the scope of data assets that can be traded externally, supporting data product trading and pledge financing scenarios.

[0092] Step S5: Dynamic Auditing and Closed-Loop Iteration

[0093] The dynamic audit iteration module ensures the continuous effectiveness of classification management:

[0094] 1. Regular audit: Automatically perform a full-volume classification quality audit every quarter to check the accuracy, completeness, and compliance rate of labeling and control strategies, and generate an audit report; trigger rule optimization for categories with high classification error rates.

[0095] 2. Business Feedback: The data asset portal provides an entry point for submitting feedback on label objections. Business personnel can submit their findings of classification errors with one click, and the data governance department will verify and correct the labels and rules.

[0096] 3. Automatic discovery: Daily scanning of new data sources and new data tables automatically triggers the classification and tagging process to ensure that new assets are included in management in a timely manner.

[0097] 4. Each year, based on business changes and regulatory updates, review and revise the classification and grading rules to achieve continuous iterative optimization of the system.

[0098] Through the above embodiments, the manufacturing group has achieved unified classification management of data assets across the entire domain. The classification coverage has increased from 60% of the original structured data to more than 95% of all types of data, reducing the workload of manual classification by 70%. At the same time, it has opened up the path from data asset classification to financial table entry, effectively supporting the company's data asset operation.

[0099] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for classifying and managing enterprise data assets, characterized in that, Includes the following steps: S1. Multi-dimensional integrated classification rule construction: Construct an enterprise-level unified classification and grading rule system that integrates business theme dimension, security sensitivity dimension, ownership source dimension, and accounting asset dimension, and configure the judgment criteria, tag fields and control strategy templates for each type of data; S2, Automatic Discovery and Intelligent Tagging of Full-Domain Data Assets: Automatically scans the enterprise's full-domain structured, semi-structured, and unstructured data resources, classifies and identifies them through a combination of rule engine and semantic recognition model, outputs classification labels and confidence scores, triggers manual review for data with confidence scores below the threshold, and forms an initial data asset catalog and ledger. S3, Tag lineage transfer and differentiated management: Bind the category tags to the lineage of data assets. When data flows across systems and interfaces, it automatically carries and inherits the category tags. Based on the tags, it automatically matches the corresponding level of access permissions, security protection, sharing approval, and lifecycle disposal strategies. S4. Asset Value and Accounting Attribute Mapping: Based on the classification results, the value of each type of data asset is assessed and the accounting attributes are determined, and the corresponding intangible assets, inventory or off-balance sheet disclosed asset categories are mapped to generate an asset ledger subset that is suitable for financial accounting. S5. Dynamic Audit and Closed-Loop Iteration: Regularly audit and verify the accuracy of data asset classification, the completeness of labels, and the effectiveness of control implementation; receive label objection feedback from the business side; dynamically update classification rules and asset catalogs; and form a continuously iterative closed-loop management mechanism.

2. The enterprise data asset classification management method according to claim 1, characterized in that, In step S1, the business subject dimension is divided into three levels: business domain, subject domain, and data entity; the security sensitivity dimension is divided into four levels: public data, internal general data, sensitive data, and core confidential data; the ownership source dimension is divided into four categories: owned data, externally purchased data, cooperative exchange data, and publicly collected data; and the accounting asset dimension is divided into three categories: data that can be included in intangible assets, data that can be included in inventory, and off-balance sheet data resources, based on the accounting treatment regulations for data resources.

3. The enterprise data asset classification management method according to claim 1, characterized in that, In step S2, intelligent marking specifically includes: The system automatically collects and identifies table names, field names, and comment information from structured data using metadata, and identifies sensitive fields through regular expression matching. Extract the main content and sensitive information of unstructured documents, contracts, drawings, audio and video through natural language processing and content semantic recognition; For each data asset, a classification label confidence score is generated. Labels with a confidence score higher than the preset high confidence threshold are automatically applied. Labels with a confidence score lower than the high confidence threshold but higher than the low confidence threshold are pushed to the corresponding business owner for review. Labels with a confidence score lower than the low confidence threshold are marked as pending confirmation.

4. The enterprise data asset classification management method according to claim 1, characterized in that, In step S3, the transfer of tagged lineage specifically includes: Generate a unique identifier and category tag metadata for each data asset. During data synchronization, interface calls, export downloads, and copy generation, the category tag is transmitted synchronously along the data lineage. When receiving data, the downstream system automatically reads and inherits the classification label. It is forbidden to downgrade or modify the security level without authorization. Label changes must go through an approval process and the labels across the entire chain must be updated synchronously.

5. The enterprise data asset classification management method according to claim 1, characterized in that, In step S3, the differentiated management strategy includes: Access control: Core confidential data is subject to the principle of least privilege and multi-person approval, while public data is accessible to all employees; Security protection and control: Highly sensitive data is encrypted and stored, de-identified, and audited; ordinary data is protected with basic security measures. Shared workflow management: Automatically match shared approval processes based on classification level; the higher the level, the more approval levels there are. Lifecycle management: Different retention periods, archiving rules and destruction mechanisms are configured for different categories of data, and personal information data complies with legal retention requirements.

6. The enterprise data asset classification management method according to claim 1, characterized in that, In step S4, the accounting attribute determination criteria include: the controllability of data assets, the measurability of costs, and the expected inflow of economic benefits; data assets intended for external sale are mapped to inventory, non-monetary data assets intended for internal use and which are identifiable are mapped to intangible assets, and those that do not meet the recognition criteria are mapped to off-balance-sheet disclosed data resources.

7. An enterprise data asset classification management system, used to implement the enterprise data asset classification management method according to any one of claims 1-6, characterized in that, include: Rule configuration module: used to configure multi-dimensional integrated classification and grading rules, tag system, control strategy templates and confidence thresholds; Asset discovery and intelligent tagging module: used to automatically scan data resources across the entire domain, classify and identify them through a rule engine and semantic recognition model, generate classification tags with confidence, trigger the manual review process, and build a data asset catalog and ledger; Tag Flow and Management Module: Used to bind data lineage and category tags, enabling automatic tag carrying and inheritance across systems, and implementing differentiated permission, security, sharing, and lifecycle management based on tags; Asset value mapping module: used to assess the value of data assets and determine their accounting attributes, generate asset ledgers that are adapted to financial accounting, and support asset entry, valuation and transaction; Dynamic Audit Iteration Module: Used to periodically perform classification quality audits, receive feedback from business side regarding label objections, and drive dynamic updates of classification rules and asset catalogs.

8. The enterprise data asset classification management system according to claim 7, characterized in that, The asset discovery and intelligent tagging module includes a structured data recognition unit, an unstructured data recognition unit, a confidence calculation unit, and a manual review unit; The structured data recognition unit is used to collect metadata from databases and business systems, and to match business topics with sensitivity levels through a rule engine; The unstructured data recognition unit is used to perform content parsing and semantic recognition on documents, contracts, drawings, and audio / video materials, and to extract classification features. The confidence calculation unit is used to calculate the confidence of the classification label by combining the rule matching degree and semantic similarity. The manual review unit is used to push low-confidence labels to the corresponding data owners for review and confirmation.

9. The enterprise data asset classification management system according to claim 7, characterized in that, The tag circulation and control module includes a lineage tracking unit, a tag transmission unit, and a strategy execution unit; The lineage tracking unit is used to record the entire link flow path and copy relationship of data assets; The tag transmission unit is used to synchronously carry classification tag metadata during data flow, which is automatically inherited by downstream systems; The strategy execution unit is used to automatically invoke the corresponding access control, encryption and desensitization, approval process, and lifecycle management interfaces based on the category tags.

Images