Railway data classification and grading mechanism-oriented data desensitization strategy generation method and system

Abstract

The invention discloses a railway data classification and grading mechanism-oriented data desensitization strategy generation method and system, and belongs to the technical field of data security. The method comprises the following steps: 1, obtaining railway business data and extracting features; 2, the data obtained in the step 1 are matched with a railway information system data classification and grading catalog library, and multi-dimensional labels are marked on the data in combination with a service scene; 3, performing preliminary screening on the multi-dimensional label data in the step 2 based on a rule table; 4, according to an intensity coefficient range given by the rule matrix in the step 3, a final intensity coefficient is calculated by a quantitative evaluation model; and 5, generating and outputting a strategy based on the final strength coefficient in the step 4. The invention is used for solving the problem that the desensitization rule is separated from the security level and the service scene of the data and cannot be dynamically adjusted according to the context in the prior art.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data security, and particularly relates to a data desensitization strategy generation method and system for a railway data classification and grading mechanism. BACKGROUND

[0002] Data desensitization has become an essential link in data processing activities. The railway industry, as a national critical information infrastructure, has data with complex types, high value density, and strong real-time processing characteristics. Currently, the railway industry has established an internal data classification and grading mechanism. System data classification consists of system business attribution classification and extended classification. Business extended classification is to further classify data under the first or second business subclass by using line classification method after determining the information system business attribution, combining business management needs, and classifying data with the same or similar business attributes and characteristics to form multiple layers of data classes. Railway data grading classifies data into three levels of "core data" (such as passenger transport data and dispatching data), "important data" (such as freight transport data and maintenance data), and "general data" from the perspective of data security protection by determining two grading elements of influence object and influence degree. "General data" covers a wide range and is further classified into four levels from high to low (S4, S3, S2, and S1).

[0003] The existing general data desensitization products (such as IBM Optim and Oracle DSG) have the following significant shortcomings: first, static rigidity of strategy: the desensitization rules are separated from the security level and business scenario of data, and cannot be dynamically adjusted according to the context. Second, poor industry adaptability: the built-in identification rules and desensitization algorithms cannot effectively process the data formats specific to the railway industry. Third, lack of dynamics: once the strategy is established, it cannot be hot switched according to real-time risks or business scenarios, and manual intervention is required, which is inefficient and prone to errors.

[0004] Therefore, there is an urgent need for a desensitization strategy generation method that can be deeply adapted to the railway industry data classification and grading mechanism and dynamically intelligent, to solve the above problems. SUMMARY

[0005] The application provides a data desensitization strategy generation method and system for a railway data classification and grading mechanism, to solve the problems of the existing technology, such as the separation of desensitization rules from the security level and business scenario of data, the inability to dynamically adjust according to the context, the inability of built-in identification rules and desensitization algorithms to effectively process data formats specific to the railway industry, and the inability to hot switch according to real-time risks or business scenarios once the strategy is established, requiring manual intervention, which is inefficient and prone to errors.

[0006] The application is implemented by the following technical solutions: A method for generating data anonymization strategies for railway data classification and grading mechanisms, comprising the following steps: Step 1: Acquisition and feature extraction of railway business data; Step 2: Match the data obtained in Step 1 with the railway information system data classification and grading catalog, and add multi-dimensional tags to the data based on the business scenario; Step 3: Perform initial screening of the multidimensional label data from Step 2 based on the rule table; Step 4: Based on the intensity coefficients given by the rule matrix in Step 3 The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Step 5: Generate and output the strategy based on the final intensity coefficients from Step 4.

[0007] Furthermore, step 1 specifically involves the railway business data being structured data in a relational database and text data stored on an SFTP server; the feature extraction specifically involves extracting the metadata of the data, including the information system to which it belongs, database connection information, database name, table name, field name, and structure information.

[0008] Furthermore, the multidimensional labels in step 2 specifically include classification labels C, hierarchical labels L, and scene labels S; The classification label C specifically comprises six major business areas, primary, secondary, and extended categories; The hierarchical label L specifically determines the security level of the data; The scenario label S specifically marks the current business usage scenario of the data.

[0009] Furthermore, the rule table in step 3 is specifically as follows: If the category label C represents passenger ticket information, the hierarchical label L represents L-Core, and the scenario label S represents data outreach, then the recommended de-identification algorithm is a mask with a strength coefficient. The range is 0.9-1.0; If the category label C represents truck repair information, the hierarchical label L represents L-IMPORTANT, and the scenario label S represents data analysis, then the recommended de-identification algorithm is sampling and intensity coefficient. The range is 0.3-0.5; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents manufacturer collaboration, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.2-0.4; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents data outreach, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.6-0.8; If the classification label C represents train location information, the hierarchical label L represents L-S4, and the scenario label S represents emergency rescue, then the recommended de-identification algorithm is generalization with a strength coefficient. The range is 0.1-0.2; Based on the combination of the data's classification label C, hierarchical label L, and scene label S, a rule table is queried to obtain a basic algorithm selection and suggested intensity range.

[0010] Furthermore, the quantitative evaluation model in step 4 specifically includes:

[0011] In the formula For business parameters; The objective function is... For privacy protection strength function; These are weighting coefficients; This is a data utility function.

[0012] Furthermore, specific desensitization algorithms require the setting of actual operational parameters. The relationship is as follows:

[0013] In the formula Direction factor; These are the parameter values ​​corresponding to the weakest desensitization intensity; These are the parameter values ​​corresponding to the strongest desensitization intensity; Privacy protection strength function as follows:

[0014] In the formula For fitting and normalization constants; Data utility function as follows:

[0015] In the formula For fitting and normalization constants; Within the scope of business parameters [ , Within, by solving the objective function The maximum value is used to obtain the optimal business coefficient. And as parameters for the final de-identification algorithm to be used: .

[0016] Furthermore, step 5 specifically involves the task requiring the data in the category of "transportation production / production operation / freight operation / equipment fault information" to be anonymized. The business scenario is "vendor collaboration," and the anonymization strength factor is 0.22. The specific anonymization operation involves five columns of data: train number, train type, train number, fault code, and equipment code. Among them, the equipment code column is specified to use a mapping anonymization algorithm with a parameter of 902. The task also sets the processing speed to single-channel.

[0017] A data anonymization strategy generation system for railway data classification and grading mechanisms is provided. The system utilizes the aforementioned data anonymization strategy generation method for railway data classification and grading mechanisms. The system comprises: Data processing module: Acquisition and feature extraction of railway business data; Tag matching module: Matches the acquired data with the railway information system's data classification and hierarchical directory, and adds multi-dimensional tags to the data based on business scenarios; Filtering module: Labels multidimensional labeled data based on a rule table; Intensity coefficients given by the rule matrix The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Strategy generation and output module: Generates and outputs strategies based on the final strength coefficient.

[0018] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method described above.

[0019] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described above.

[0020] The beneficial effects of this invention are: This invention establishes a dynamic decision-making model driven by a three-element approach: "data classification attribute - data grading attribute - business scenario". It takes the classification labels, grading labels and business usage scenarios of railway data as inputs, and fine-tunes them through the built-in railway information system data classification and grading catalog, decision rule matrix and quantification model to automatically output the optimal and executable de-identification strategy.

[0021] This invention provides a method that can deeply understand the data classification and grading mechanism and business scenarios in the railway industry, and automatically generate accurate, dynamic, and quantifiable data desensitization strategies to achieve the optimal balance between data security and business utility.

[0022] This invention is the first to integrate the classification and grading mechanism of railway data, business scenario rules, and real-time risk quantification into the desensitization strategy generation process, solving the problem of general solutions being "unsuitable for local conditions".

[0023] The strategy of this invention is no longer statically configured, but dynamically generated based on the context, achieving precise matching of "data-strategy-scenario".

[0024] The strategy generation process of this invention introduces quantitative concepts such as "intensity coefficient," providing a data foundation for subsequent privacy protection effect evaluation and continuous strategy optimization. Attached Figure Description

[0025] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0026] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0027] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0028] It should also be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0029] The following is in conjunction with the appendix to this application specification. Figure 1 The technical solutions in the embodiments of this application are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0030] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0031] Implementation Method 1 This embodiment provides a data anonymization strategy generation method for railway data classification and grading mechanisms. The data anonymization strategy generation method includes the following steps: Step 1: Acquisition and feature extraction of railway business data; Step 2: Match the data obtained in Step 1 with the railway information system data classification and grading catalog, and add multi-dimensional tags to the data based on the business scenario; Step 3: Perform initial screening of the multidimensional label data from Step 2 based on the rule table; Step 4: Based on the intensity coefficients given by the rule matrix in Step 3 The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Step 5: Generate and output the strategy based on the final intensity coefficients from Step 4.

[0032] Furthermore, step 1 specifically involves the railway business data being structured data in a relational database and text data stored on an SFTP server; the feature extraction specifically involves extracting the metadata of the data, including the information system to which it belongs, database connection information (or SFTP connection information), database name, table name, field name, and structural information (field type, length, constraints).

[0033] Furthermore, the multidimensional labels in step 2 specifically include classification labels C, hierarchical labels L, and scene labels S; The classification label C specifically includes six major business areas, primary, secondary and extended categories; such as: strategic decision-making / business development / passenger transport operation (12306) / product sales / passenger ticket information.

[0034] The hierarchical label L specifically determines the security level of the data; "core data" is labeled L-CORE, "important data" is labeled L-IMPORTANT, and "general data" is labeled L-S1, L-S2, L-S3, and L-S4. The scenario label S specifically marks the current business usage scenario of the data, such as: data outsourcing, data analysis, test development, vendor collaboration, accident analysis, and emergency rescue.

[0035] Furthermore, the rule table in step 3 is specifically as follows: If the category label C represents passenger ticket information, the hierarchical label L represents L-Core, and the scenario label S represents data outreach, then the recommended de-identification algorithm is a mask with a strength coefficient. The range is 0.9-1.0; If the category label C represents truck repair information, the hierarchical label L represents L-IMPORTANT, and the scenario label S represents data analysis, then the recommended de-identification algorithm is sampling and intensity coefficient. The range is 0.3-0.5; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents manufacturer collaboration, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.2-0.4; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents data outreach, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.6-0.8; If the classification label C represents train location information, the hierarchical label L represents L-S4, and the scenario label S represents emergency rescue, then the recommended de-identification algorithm is generalization with a strength coefficient. The range is 0.1-0.2; Based on the combination of the data's classification label C, hierarchical label L, and scene label S, a rule table is queried to obtain a basic algorithm selection and suggested intensity range.

[0036]

[0037] Furthermore, step 4, the quantitative evaluation model, specifically aims to find the optimal balance between privacy protection strength and data availability. P(λ) +

[0038] In the formula For business parameters; Let P(λ) be the objective function; and let P(λ) be the privacy protection strength function. These are weighting coefficients; This is a data utility function.

[0039] Furthermore, specific de-identification algorithms require setting actual operational parameters (such as mapping table size, position offset distance, number of characters to retain, etc.). The relationship is as follows:

[0040] In the formula The direction factor (determined by the specific desensitization algorithm); These are the parameter values ​​corresponding to the weakest desensitization intensity; These are the parameter values ​​corresponding to the strongest desensitization intensity; =+1 ： Positive correlation; intensity coefficient The larger the value, the more business parameters... The larger the value; for example, in masking algorithms. Larger → Mask bit width The larger; = -1: and Negative correlation; intensity coefficient The larger the value, the more business parameters... The smaller the value; for example, in mapping replacement. Larger → Mapping table size The smaller.

[0041] The privacy protection strength function P(λ) is as follows:

[0042] In the formula For fitting and normalization constants; Data utility function as follows:

[0043] In the formula For fitting and normalization constants; 、 These two parameters are established by analyzing historical data. The correlation between actual privacy risks and data utility is obtained by curve fitting and normalization of λ.

[0044] and It is a weighting coefficient, and + = 1. Their values ​​are dynamically determined by the hierarchy L and the scene S. For example: For L-Core data and in scenarios where data is sent externally... =0.9, =0.1 (extremely safety-biased). For L-S4 driving data and in emergency rescue scenarios, =0.1, =0.2 (extremely biased towards availability).

[0045] Final decision: within the range of business parameters [ , Within this framework, the optimal business coefficients are obtained by solving for the maximum value of the objective function F(λ). And as parameters for the final de-identification algorithm to be used: = argmax_{λ ∈[ , ]} F(λ).

[0046] The example is: Desensitization through mapping and replacement of sensitive device codes. Business Background: The group company needs to share some equipment failure records with equipment manufacturers for joint failure analysis. The data contains sensitive equipment codes, and direct sharing poses a risk of leaking core asset information; therefore, anonymization is necessary. However, the anonymized data still needs to ensure that manufacturers can effectively analyze failure modes.

[0047] 1. Output of the regular matrix Based on the query rule matrix (data classification: equipment fault information, data grading: S4, scenario: manufacturer collaboration), the following results were obtained: Recommended desensitization algorithm: mapping replacement. = -1 (negative correlation).

[0048] Strength coefficient Range: [0.2, 0.4].

[0049] 2. Business Parameter Mapping The device code mapping table has a maximum size of 1000 and a minimum size of 20. Replacing with 1000 mapping codes results in a very fine-grained mapping, closer to the original data, but with weaker privacy protection; however, it is highly effective (it can better preserve device differences). Replacing all real device codes with 20 mapping codes results in a very coarse-grained mapping, strong privacy protection, but poor effectiveness (it is difficult to distinguish specific devices).

[0050] Parameter mapping: = 1000 + (20 - 1000) * ( (σ - 0.2) / (0.4 - 0.2) ) * (-1) Simplified version: = 1000 - 4900 * (σ - 0.2) 3. Define privacy function and data utility function Privacy functions : (Based on historical data fitting and) Quantity: ) Utility function : (Based on historical data fitting and) Quantity: ) 4. Define weights and construct the objective function. In this business scenario, a balance is struck between security and data availability, and the following weights are selected: = 0.5, = 0.5 Objective function: = 0.5 * + 0.5 *

[0051] 5. Find the optimal solution

[0052] in conclusion: exist The value is maximized when the value is 902. The optimal strategy is to use the desensitization intensity coefficient ( =0.22), using a mapping table of size 902.

[0053] Furthermore, step 5 specifically involves generating a set of specific policy instructions that can be executed by the de-identification engine, targeting the de-identification task instructions for the Oracle database in "HMIS System 1.0": The task requires de-identification of data in the category of "Transportation Production / Production Operation / Freight Operation / Equipment Fault Information" (level L-S4), with the business scenario being "vendor collaboration" and the de-identification strength factor being 0.22; the specific de-identification operation involves five columns of data: train number, vehicle type, vehicle number, fault code, and equipment code; among which, the equipment code column specifies the use of a mapping de-identification algorithm with a parameter of 902; the task also sets the processing speed to single channel (channel: 1).

[0054] Policy instructions are in JSON format, as shown in the example below: { "job": { "content": [{ "scource": { "system": "HMIS System 1.0", "type": "oracle", "url": " jdbc:oracle:thin: @10.16.xx / hmis", "user": "****", "pwd": "*********", "data_class": "Transportation Production / Production Operations / Freight Operations / Equipment Failure Information", / / Data Classification "data_level": "L-S4", / / Data level "scene": "vendor collaboration", / / business scenario "factor": 0.22, / / Desensitization intensity "column": [{ "index": 0, / / Train number "type": "string" }, { "index": 1, / / Vehicle type "type": "string" }, { "index": 2, / / Vehicle number "type": "string" }, { "index": 3, / / Fault code "type": "string" }, { "index": 4, / / Device code "type":"string", "transformer": { "name": "mapping", / / Desensitization algorithm: mapping replacement "param":

[902] } } ] }, "setting": { "speed": { "channel": "1" } } }] } } This invention was piloted in the process of externalizing ticket interface data at Harbin Railway Bureau Group Co., Ltd. Practice has proven that, compared to traditional static data masking solutions, it significantly improves policy management efficiency, achieving automated and intelligent generation and iteration of policies, reducing manual configuration work by 90%. Security and compliance are comprehensively enhanced, meeting the requirements for dynamic protection of important data.

[0055] Implementation Method 2 This embodiment provides a data anonymization strategy generation system for railway data classification and grading mechanisms. The data anonymization strategy generation method includes the following steps: Data processing module: Acquisition and feature extraction of railway business data; Tag matching module: Matches the acquired data with the railway information system's data classification and hierarchical directory, and adds multi-dimensional tags to the data based on business scenarios; Filtering module: Labels multidimensional labeled data based on a rule table; Intensity coefficients given by the rule matrix The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Strategy generation and output module: Generates and outputs strategies based on the final strength coefficient.

[0056] Implementation Method 3 This invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor. The memory stores software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory and processor are connected via a bus. Specifically, the processor implements any step in Embodiment 1 by running the computer program stored in the memory.

[0057] It should be understood that, in the embodiments of the present invention, the processor may be a Central Processing Unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0058] Memory may include read-only memory, flash memory, and random access memory, and provides instructions and data to the processor. Some or all of the memory may also include non-volatile random access memory.

[0059] It should be understood that if the integrated modules / units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods described above can also be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0060] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0061] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the above device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0062] It should be noted that the methods and detailed examples provided in the above embodiments can be incorporated into the apparatus and devices provided in the embodiments for mutual reference, and will not be repeated here.

[0063] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0064] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / device embodiments described above are merely illustrative. For instance, the division of modules or units described above is merely a logical functional division, and in actual implementation, it can be divided in other ways. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0065] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for generating data anonymization strategies for railway data classification and grading mechanisms, characterized in that, The data anonymization strategy generation method includes the following steps: Step 1: Acquisition and feature extraction of railway business data; Step 2: Match the data obtained in Step 1 with the railway information system data classification and grading catalog, and add multi-dimensional tags to the data based on the business scenario; Step 3: Perform initial screening of the multidimensional label data from Step 2 based on the rule table; Step 4: Based on the intensity coefficients given by the rule matrix in Step 3 The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Step 5: Generate and output the strategy based on the final intensity coefficients from Step 4.

2. The data anonymization strategy generation method according to claim 1, characterized in that, Step 1 specifically involves the following: the railway business data consists of structured data in a relational database and text data stored on an SFTP server; the feature extraction specifically involves extracting the metadata of the data, including the information system to which it belongs, database connection information, database name, table name, field name, and structure information.

3. The data anonymization strategy generation method according to claim 1, characterized in that, The multidimensional labels in step 2 specifically include classification labels C, hierarchical labels L, and scene labels S; The classification label C specifically comprises six major business areas, primary, secondary, and extended categories; The hierarchical label L specifically determines the security level of the data; The scenario label S specifically marks the current business usage scenario of the data.

4. The data anonymization strategy generation method according to claim 1, characterized in that, The rule table in step 3 is specifically as follows: If the category label C represents passenger ticket information, the hierarchical label L represents L-Core, and the scenario label S represents data outreach, then the recommended de-identification algorithm is a mask with a strength coefficient. The range is 0.9-1.0; If the category label C represents truck repair information, the hierarchical label L represents L-IMPORTANT, and the scenario label S represents data analysis, then the recommended de-identification algorithm is sampling and intensity coefficient. The range is 0.3-0.5; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents manufacturer collaboration, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.2-0.4; If the classification label C represents equipment fault information, the hierarchical label L represents L-S4, and the scenario label S represents data outreach, then the recommended de-identification algorithm is mapping and strength coefficient. The range is 0.6-0.8; If the classification label C represents train location information, the hierarchical label L represents L-S4, and the scenario label S represents emergency rescue, then the recommended de-identification algorithm is generalization with a strength coefficient. The range is 0.1-0.2; Based on the combination of the data's classification label C, hierarchical label L, and scene label S, a rule table is queried to obtain a basic algorithm selection and suggested intensity range.

5. The data anonymization strategy generation method according to claim 4, characterized in that, The quantitative evaluation model in step 4 is specifically as follows: In the formula For business parameters; The objective function is... For privacy protection strength function; These are weighting coefficients; This is a data utility function.

6. The data anonymization strategy generation method according to claim 5, characterized in that, The specific desensitization algorithm requires setting actual operational parameters. The relationship is as follows: In the formula Direction factor; These are the parameter values ​​corresponding to the weakest desensitization intensity; These are the parameter values ​​corresponding to the strongest desensitization intensity; Privacy protection strength function as follows: In the formula For fitting and normalization constants; Data utility function as follows: In the formula For fitting and normalization constants; Within the scope of business parameters [ , Within, by solving the objective function The maximum value is used to obtain the optimal business coefficient. And as parameters for the final de-identification algorithm to be used: 。 7. The data anonymization strategy generation method according to claim 1, characterized in that, Step 5 specifically involves the task of desensitizing data in the category of "transportation production / production operation / freight operation / equipment failure information". The business scenario is "vendor collaboration", and the desensitization strength factor is 0.

22. The specific desensitization operation involves five columns of data: train number, train type, train number, fault code, and equipment code. Among them, the equipment code column is specified to use a mapping desensitization algorithm with a parameter of 902. The task also sets the processing speed to single channel.

8. A data anonymization strategy generation system for railway data classification and grading mechanisms, characterized in that, The data anonymization strategy generation system uses a data anonymization strategy generation method for railway data classification and grading mechanisms as described in claims 1-7. The data anonymization strategy generation system includes: Data processing module: Acquisition and feature extraction of railway business data; Tag matching module: Matches the acquired data with the railway information system's data classification and hierarchical directory, and adds multi-dimensional tags to the data based on business scenarios; Filtering module: Labels multidimensional labeled data based on a rule table; Intensity coefficients given by the rule matrix The range, and the final intensity coefficient, are calculated by the quantitative evaluation model; Strategy generation and output module: Generates and outputs strategies based on the final strength coefficient.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1-7.

Images