Sensitive data grading desensitization method and system

Abstract

The invention discloses a sensitive data hierarchical desensitization method and system, belongs to the technical field of data security, and aims to solve the technical problem of how to improve the sensitive field identification accuracy and the correlation analysis availability of desensitized data so as to improve the system operation efficiency. According to the technical scheme, the method comprises the steps that sensitive data classification is conducted, wherein each sensitive data instance is divided into three levels of classification labels of core sensitivity, service sensitivity and basic sensitivity by constructing a feature table, extracting data features and calculating sensitivity; and desensitization processing: performing desensitization processing on the related data according to the desensitization mode and rule matched in the feature table in combination with the current user permission to obtain desensitized data.

Description

Technical Field

[0001] The present invention relates to the field of data security technology, and in particular to a sensitive data hierarchical desensitization method and system. Background Art

[0002] Traditional desensitization technologies employ globally unified desensitization rules (e.g., whole-field replacement and fixed masks), which fail to meet the differentiated needs of sensitive system data at different levels of confidentiality and for different roles. Existing classification methods are mostly based on static data classification (e.g., GB / T 35273-2020), lacking the ability to identify dynamic sensitive data in business scenarios. Conventional desensitization solutions can lead to the destruction of data relevance, hindering data analysis and clue mining in related work.

[0003] Therefore, how to improve the accuracy of sensitive field recognition and the availability of correlation analysis of desensitized data, thereby improving system operation efficiency, is a technical problem that needs to be solved urgently. Summary of the Invention

[0004] The technical task of the present invention is to provide a sensitive data hierarchical desensitization method and system to solve the problem of how to improve the accuracy of sensitive field recognition and the availability of association analysis of desensitized data, thereby improving the system operation efficiency.

[0005] The technical task of the present invention is achieved in the following manner: a hierarchical desensitization method for sensitive data, the method is as follows:

[0006] Sensitive data classification: By building a feature table, extracting data features, and calculating sensitivity, each sensitive data instance is classified into three levels: core sensitive, business sensitive, and basic sensitive.

[0007] Desensitization: Desensitize the relevant data based on the desensitization methods and rules matched in the feature table and the current user permissions to obtain the desensitized data.

[0008] As a preference, the feature table is constructed as follows:

[0009] Characteristic dimension: defines the classification basis of data sensitivity;

[0010] Regular expression value: pattern rules that accurately match specific data instances;

[0011] Normalized sensitivity calculation rule: Convert the original sensitivity to a standard value of [0,1] according to the preset rules.

[0012] More preferably, the preset rules include the following methods:

[0013] ①Preset value: directly set a constant between 0 and 1;

[0014] ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years);

[0015] ③Interval setting: directly set the sensitivity interval.

[0016] As a preference, the data features are extracted as follows:

[0017] The data to be desensitized is matched against all the data to be desensitized in the data to be desensitized and the normalized sensitivity calculation rule of each data instance to be desensitized by traversing the regular expression of the values ​​in the feature table;

[0018] During extraction, the existing multi-threaded + regular expression engine (Hyperscan) technology framework is used to improve matching efficiency, and a hot data preloading module is added to load frequently appearing feature table content (such as personal information, amount, etc.) into memory, reducing I / O overhead and improving efficiency.

[0019] As a preference, the sensitivity is calculated as follows:

[0020] Sensitivity calculation is performed based on the acquired data instances and corresponding calculation rules. The calculation method is executed according to the corresponding content of the feature table, and finally a three-level classification label is formed for each data instance: core sensitivity (≥0.8), business sensitivity (0.5-0.8), and basic sensitivity (<0.5);

[0021] Based on the user type and role permissions of the user who obtained the data, the desensitized data is returned to the calling system. For example, if the user's permission is 0.8, the content with a permission greater than 0.8 needs to be desensitized, and the content with a permission less than 0.8 is returned without being desensitized.

[0022] During the operation, the sensitivity calculation results are cached. When duplicate data is matched, no additional calculation is performed. Instead, the original calculation results are used directly, which effectively improves the operation efficiency.

[0023] As a preference, the desensitization treatment is as follows:

[0024] Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process;

[0025] Irreversible desensitization: including generalization (e.g., "500,000 yuan → [300,000-1,000,000] yuan"), permutation (maintaining data distribution), and K-anonymization (≥3-person group confusion);

[0026] Dynamic watermark: Inject invisible digital watermark (DWT-SVD algorithm) to record visitor information.

[0027] A sensitive data hierarchical desensitization system, the system comprising:

[0028] The sensitive data classification module is used to classify each sensitive data instance into three levels of classification labels: core sensitive, business sensitive, and basic sensitive by building a feature table, extracting data features, and calculating sensitivity;

[0029] The desensitization processing module is used to desensitize the relevant data according to the desensitization methods and rules matched in the feature table and combined with the current user permissions to obtain the desensitized data.

[0030] As a preference, the feature table is constructed as follows:

[0031] (1) Feature dimension: defines the classification basis of data sensitivity;

[0032] (2) Regular expression for value: pattern rules that accurately match specific data instances;

[0033] (3) Normalized sensitivity calculation rule: The original sensitivity is converted to a standard value of [0, 1] according to the preset rules; wherein the preset rules include the following methods:

[0034] ①Preset value: directly set a constant between 0 and 1;

[0035] ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years);

[0036] ③Interval setting: directly set the sensitivity interval;

[0037] The specific steps for extracting data features are as follows: the data to be desensitized is matched against the regular expressions in the feature table to all the data to be desensitized and the normalized sensitivity calculation rules for each instance of the data to be desensitized; during extraction, the existing multi-threaded + regular expression engine (Hyperscan) technology framework is used to improve matching efficiency, and a hot data preloading module is added to load frequently appearing feature table content (such as personal information, amount, etc.) into memory, reducing I / O overhead and improving efficiency;

[0038] The specific steps for calculating sensitivity are as follows: sensitivity calculation is performed based on the acquired data instances and the corresponding calculation rules. The calculation method is executed according to the corresponding content of the feature table, and finally a three-level classification label is formed for each data instance: core sensitivity (≥0.8), business sensitivity (0.5-0.8) and basic sensitivity (<0.5). Based on the user type who obtained the data and the role permissions, the desensitized data is returned to the caller system. For example, if the personnel permission is 0.8, the content >0.8 needs to be desensitized, and the content ≤0.8 is returned without desensitization. The sensitivity calculation results are cached during operation. When duplicate data is matched, no additional calculation is performed, but the original calculation results are directly used, which effectively improves operation efficiency.

[0039] The desensitization method of the desensitization processing module is as follows:

[0040] ① Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process;

[0041] ② Irreversible desensitization: including generalization (e.g., "500,000 yuan → [300,000-1,000,000] yuan"), permutation (maintaining data distribution), and K-anonymization (≥3-person group confusion);

[0042] ③ Dynamic watermark: inject invisible digital watermark (DWT-SVD algorithm) to record visitor information.

[0043] An electronic device comprising: a memory and at least one processor;

[0044] Wherein, the memory stores a computer program;

[0045] The at least one processor executes the computer program stored in the memory, so that the at least one processor performs the sensitive data classification and desensitization method as described above.

[0046] A computer-readable storage medium stores a computer program, which can be executed by a processor to implement the sensitive data hierarchical desensitization method as described above.

[0047] The sensitive data hierarchical desensitization method and system of the present invention have the following advantages:

[0048] (1) This invention solves the business availability problem caused by the "one-size-fits-all" desensitization of data in relevant data-sensitive systems. It also performs hierarchical processing of sensitive data based on business scenarios, selects corresponding desensitization rules according to the classification, and achieves precise desensitization of "data available but not visible". It protects privacy while retaining the associated value of data. It is suitable for privacy protection scenarios of sensitive data such as opinion information, case materials, personnel files, etc. in relevant data-sensitive system businesses;

[0049] (2) The present invention effectively improves data security and business compatibility, effectively improves the accuracy of sensitive field identification, greatly improves the usability of correlation analysis of desensitized data, and is less time-consuming than traditional classification methods, effectively improving system operation efficiency;

[0050] (3) The present invention is independently deployed and connected to the business system through the system interface. When data is accessed, it is automatically marked and the desensitization level is dynamically adjusted. Then, desensitization is performed in real time according to the corresponding desensitization level based on the authority of the operator information. The desensitized data is returned and the desensitization log is recorded for subsequent desensitization reversal. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The present invention will be further described below with reference to the accompanying drawings.

[0052] Attachment Figure 1 Schematic diagram of a hierarchical desensitization method for sensitive data. DETAILED DESCRIPTION

[0053] The sensitive data hierarchical desensitization method and system of the present invention are described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Example 1:

[0055] As attached Figure 1 As shown, this embodiment provides a sensitive data hierarchical desensitization method, the method is as follows:

[0056] S1. Sensitive Data Classification: Each sensitive data instance is classified into three levels: core sensitive, business sensitive, and basic sensitive by constructing a feature table, extracting data features, and calculating sensitivity.

[0057] S2. Desensitization: Desensitize the relevant data based on the desensitization methods and rules matched in the feature table and the current user permissions to obtain the desensitized data.

[0058] The construction of the feature table in step S1 of this embodiment is as follows:

[0059] S1-101, Feature Dimension: Define the classification basis of data sensitivity (such as field type, fund size, etc.);

[0060] S1-102, value regular expression: a pattern rule that accurately matches a specific data instance; for example, the identity information regular expression is "^[1-9]\d{5}(19|20)\d{2}(0[1-9]|1[0-2])(0[1-9]|

[12] \d|3

[01] )\d{3}[\dXx]$";

[0061] S1-103, Normalized sensitivity calculation rule: Convert the original sensitivity to the [0,1] standard value according to the preset rules.

[0062] The preset rules in step S1-103 of this embodiment include the following methods:

[0063] ① Default value: directly set a constant between 0 and 1; for example, the default value of identity information is 1.0 (core sensitive);

[0064] ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years);

[0065] Example: Credit card transaction records (initial sensitivity = 1.0): Current transaction: Sensitivity = 1.0 (core sensitivity); Transaction one year ago: Sensitivity = 1.0 / (1+1) = 0.5 (business sensitivity); Transaction five years ago: Sensitivity = 1.0 / (1+5) ≈ 0.17 (basic sensitivity);

[0066] ③Interval setting: directly set the sensitivity interval; example: if the capital scale is [0, 1 million) = 0.3, [1 million, 100 million) = 0.6, [100 million, +∞) = 0.9).

[0067] The specific features of the extracted data in step S1 of this embodiment are as follows:

[0068] S1-201, matching the data to be desensitized with the regular expression of the values ​​in the feature table to all the data to be desensitized and the normalized sensitivity calculation rule of each data instance to be desensitized;

[0069] S1-202: During extraction, the existing multi-threaded + regular expression engine (Hyperscan) technology framework is used to improve matching efficiency, and a hot data preloading module is added to load frequently appearing feature table content (such as personal information, amount, etc.) into memory, reducing I / O overhead and improving efficiency.

[0070] The calculation sensitivity in step S1 of this embodiment is specifically as follows:

[0071] S1-301. Calculate sensitivity based on the acquired data instances and corresponding calculation rules. The calculation method is performed according to the corresponding content of the feature table, and ultimately form three-level classification labels for each data instance: core sensitivity (≥0.8), business sensitivity (0.5-0.8), and basic sensitivity (<0.5);

[0072] S1-302: Based on the user type and role permissions of the user who obtained the data, the desensitized data is returned to the calling system. For example, if the user's permission is 0.8, the content with a permission level greater than 0.8 needs to be desensitized, and the content with a permission level less than 0.8 is returned without being desensitized.

[0073] S1-303. During operation, the sensitivity calculation results are cached. When duplicate data is matched, no additional calculation is performed. Instead, the original calculation results are directly used, effectively improving operation efficiency.

[0074] The desensitization process in step S2 of this embodiment is as follows:

[0075] ① Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process;

[0076] ② Irreversible desensitization: including generalization (e.g., "500,000 yuan → [300,000-1,000,000] yuan"), permutation (maintaining data distribution), and K-anonymization (≥3-person group confusion);

[0077] ③ Dynamic watermark: inject invisible digital watermark (DWT-SVD algorithm) to record visitor information.

[0078] Example 2:

[0079] This embodiment provides a sensitive data hierarchical desensitization system, which includes:

[0080] The sensitive data classification module is used to classify each sensitive data instance into three levels of classification labels: core sensitive, business sensitive, and basic sensitive by building a feature table, extracting data features, and calculating sensitivity;

[0081] The desensitization processing module is used to desensitize the relevant data according to the desensitization methods and rules matched in the feature table and combined with the current user permissions to obtain the desensitized data.

[0082] The construction feature table in this embodiment is specifically as follows:

[0083] (1) Feature dimension: defines the classification basis of data sensitivity;

[0084] (2) Regular expression for value: pattern rules that accurately match specific data instances;

[0085] (3) Normalized sensitivity calculation rule: The original sensitivity is converted to a standard value of [0, 1] according to the preset rules; wherein the preset rules include the following methods:

[0086] ①Preset value: directly set a constant between 0 and 1;

[0087] ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years);

[0088] ③Interval setting: directly set the sensitivity interval.

[0089] The data features extracted in this embodiment are specifically as follows: the data to be desensitized is matched with all the data to be desensitized in the data to be desensitized and the normalized sensitivity calculation rules of each data instance to be desensitized by traversing the regular expression of values ​​in the feature table; during extraction, the matching efficiency is improved through the existing multi-threaded + regular expression engine (Hyperscan) technology framework, and a hot data preloading module is added to load the frequently appearing feature table content (such as personal information, amount, etc.) into the memory, reducing I / O overhead and improving efficiency.

[0090] The calculation sensitivity in this embodiment is specifically as follows: sensitivity calculation is performed based on the acquired data instance and the corresponding calculation rules, and the calculation method is executed according to the corresponding content of the feature table, and finally a three-level classification label of core sensitivity (≥0.8), business sensitivity (0.5-0.8) and basic sensitivity (<0.5) is formed for each data instance; and based on the type of user who obtains the data and based on the role authority, the desensitized data is returned to the calling system. For example, if the personnel authority is 0.8, the content >0.8 needs to be desensitized, and the content ≤0.8 is returned without desensitization; the sensitivity calculation results are cached during operation, and when duplicate data is matched, no additional calculation is performed, but the original calculation results are directly used, which effectively improves the operation efficiency.

[0091] The desensitization method of the desensitization processing module in this embodiment is as follows:

[0092] ① Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process;

[0093] ② Irreversible desensitization: including generalization (e.g., "500,000 yuan → [300,000-1,000,000] yuan"), permutation (maintaining data distribution), and K-anonymization (≥3-person group confusion);

[0094] ③ Dynamic watermark: inject invisible digital watermark (DWT-SVD algorithm) to record visitor information.

[0095] Example 3:

[0096] This embodiment also provides an electronic device, including: a memory and a processor;

[0097] wherein the memory stores computer-executable instructions;

[0098] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the sensitive data hierarchical desensitization method in any embodiment of the present invention.

[0099] The processor may be a central processing unit (CPU), 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. The processor may be a microprocessor or any conventional processor, etc.

[0100] The memory can be used to store computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created based on the use of the terminal, etc. In addition, the memory can also include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash memory card, at least one disk storage period, a flash memory device, or other volatile solid-state memory devices.

[0101] Example 4:

[0102] This embodiment further provides a computer-readable storage medium storing a plurality of instructions, which are loaded by a processor to cause the processor to execute the sensitive data hierarchical desensitization method according to any embodiment of the present invention. Specifically, a system or device equipped with a storage medium can be provided, wherein the storage medium stores software program code that implements the functions of any of the above-described embodiments, and a computer (or CPU or MPU) of the system or device can read and execute the program code stored in the storage medium.

[0103] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute part of the present invention.

[0104] Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-Rs, CD-RWs, DVD-ROMs, DVD-RYMs, DVD-RWs, DVD+RWs), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer via a communications network.

[0105] In addition, it should be clear that the functions of any of the above embodiments can be achieved not only by executing the program code read by the computer, but also by enabling the operating system operating on the computer to complete part or all of the actual operations based on the instructions of the program code.

[0106] In addition, it can be understood that the program code read from the storage medium is written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion unit connected to the computer, and then based on the instructions of the program code, the CPU installed on the expansion board or expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above embodiments.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A sensitive data hierarchical desensitization method, characterized in that: The method is as follows: Sensitive data classification: By building a feature table, extracting data features, and calculating sensitivity, each sensitive data instance is classified into three levels: core sensitive, business sensitive, and basic sensitive. Desensitization: Desensitize the relevant data based on the desensitization methods and rules matched in the feature table and the current user permissions to obtain the desensitized data.

2. The sensitive data hierarchical desensitization method according to claim 1, characterized in that: The construction of the feature table is as follows: Characteristic dimension: defines the classification basis of data sensitivity; Regular expression value: pattern rules that accurately match specific data instances; Normalized sensitivity calculation rule: Convert the original sensitivity to a standard value of [0,1] according to the preset rules.

3. The sensitive data hierarchical desensitization method according to claim 2, characterized in that: The preset rules include the following: ①Preset value: directly set a constant between 0 and 1; ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years); ③Interval setting: directly set the sensitivity interval.

4. The sensitive data hierarchical desensitization method according to claim 1, characterized in that: The extracted data features are as follows: The data to be desensitized is matched against all the data to be desensitized in the data to be desensitized and the normalized sensitivity calculation rule of each data instance to be desensitized by traversing the regular expression of the values ​​in the feature table; During extraction, the existing multi-threaded + regular expression engine technology framework is used to improve matching efficiency, and a hot data preloading module is added to load frequently appearing feature table contents into memory, reducing I / O overhead and improving efficiency.

5. The sensitive data hierarchical desensitization method according to claim 1, characterized in that: The sensitivity is calculated as follows: Sensitivity calculation is performed based on the acquired data instances and corresponding calculation rules. The calculation method is executed according to the corresponding content of the feature table, and finally a three-level classification label of core sensitivity, business sensitivity and basic sensitivity is formed for each data instance; Based on the user type and role permissions of the user who obtained the data, the desensitized data is returned to the calling system; During the operation, the sensitivity calculation results are cached. When duplicate data is matched, no additional calculation is performed. Instead, the original calculation results are used directly, which effectively improves the operation efficiency.

6. The sensitive data hierarchical desensitization method according to claim 1, characterized in that: The desensitization treatment methods are as follows: Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process; Irreversible desensitization: including generalization, substitution, and K-anonymization; Dynamic watermark: Inject invisible digital watermark to record visitor information.

7. A sensitive data hierarchical desensitization system, characterized by: The system includes: The sensitive data classification module is used to classify each sensitive data instance into three levels of classification labels: core sensitive, business sensitive, and basic sensitive by building a feature table, extracting data features, and calculating sensitivity; The desensitization processing module is used to desensitize the relevant data according to the desensitization methods and rules matched in the feature table and combined with the current user permissions to obtain the desensitized data.

8. The sensitive data hierarchical desensitization system according to claim 7, characterized in that: The construction of the feature table is as follows: (1) Feature dimension: defines the classification basis of data sensitivity; (2) Regular expression for value: pattern rules that accurately match specific data instances; (3) Normalized sensitivity calculation rule: The original sensitivity is converted to a standard value of [0, 1] according to the preset rules; wherein the preset rules include the following methods: ①Preset value: directly set a constant between 0 and 1; ② Built-in calculation rules: The calculation formula for time-attenuated sensitivity of time-sensitive data is: Sensitivity = initial sensitivity × attenuation coefficient, attenuation coefficient = 1 / (1 + number of years); ③Interval setting: directly set the sensitivity interval; Extracting data features involves: matching the data to be desensitized with the regular expressions in the feature table to all the data to be desensitized and the normalized sensitivity calculation rules for each instance of the data to be desensitized; during extraction, the existing multi-threaded + regular expression engine technology framework is used to improve matching efficiency, and a hot data preloading module is added to load frequently appearing feature table content into memory, reducing I / O overhead and improving efficiency; The specific steps for calculating sensitivity are: sensitivity calculation is performed based on the acquired data instances and the corresponding calculation rules. The calculation method is executed according to the corresponding content of the feature table, and finally a three-level classification label of core sensitivity, business sensitivity and basic sensitivity is formed for each data instance; and based on the type of user who obtained the data and the role permissions, the desensitized data is returned to the caller system; during the operation, the sensitivity calculation results are cached, and when duplicate data is matched, no additional calculation is performed, but the original calculation results are directly used, which effectively improves the operation efficiency; The desensitization method of the desensitization processing module is as follows: ① Reversible desensitization: Encrypted using the national secret SM4 algorithm, which can be decrypted after approval process; ② Irreversible desensitization: including generalization, substitution, and K-anonymization; ③ Dynamic watermark: inject invisible digital watermark to record visitor information.

9. An electronic device, characterized in that: include: memory and at least one processor; Wherein, the memory stores a computer program; The at least one processor executes the computer program stored in the memory, so that the at least one processor performs the sensitive data hierarchical desensitization method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which can be executed by a processor to implement the sensitive data hierarchical desensitization method according to any one of claims 1 to 6.

Images