Data desensitization method and device, equipment and medium
By establishing a unified desensitization abstraction layer and an effect verification mechanism, the problem of poor scalability in existing technologies is solved, enabling flexible data desensitization operations and efficient data processing, thereby improving data security and usability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGDIAN DATA IND CO LTD
- Filing Date
- 2025-12-17
- Publication Date
- 2026-04-24
AI Technical Summary
Existing data masking technologies have poor scalability, making it difficult to quickly respond to constantly changing data types, business scenarios, and security requirements. Furthermore, their masking strategies are limited and cannot meet diverse data processing needs.
It adopts a unified desensitization abstraction layer, encapsulates the interfaces of various sensitive information identification modes and desensitization strategies, automatically matches and executes desensitization operations through mapping relationships, and provides a desensitization effect verification mechanism.
It achieves high scalability and flexibility in data anonymization methods, enabling rapid adaptation to new data types and business scenarios, ensuring the accuracy and consistency of anonymization results, and improving data security and availability.
Smart Images
Figure CN121919902A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a data desensitization method, apparatus, device, and medium. Background Technology
[0002] With the development of big data and artificial intelligence technologies, data has become a core production factor driving innovation and development across various industries. Throughout the entire lifecycle of data collection, storage, processing, sharing, and application, data needs to be anonymized to maximize its commercial and social value while ensuring data security and privacy protection.
[0003] Data anonymization technology primarily works by transforming sensitive data, allowing it to retain specific data characteristics or business rules while losing its original sensitive information, thereby protecting privacy. Depending on the timing and application scenario, it is mainly divided into static anonymization and dynamic anonymization. Static anonymization typically involves batch processing of data before it is migrated from the production environment to testing, development, or analysis environments, while dynamic anonymization is performed in real-time when data is queried or accessed. Currently, the mainstream data anonymization implementation methods in the industry include: Rule-based data masking: This involves identifying sensitive data using predefined rules and applying masking strategies. For example, regular expressions can be used to match sensitive information such as ID card numbers and mobile phone numbers, and then masking operations such as replacement and masking can be applied.
[0004] Machine learning-based desensitization: Utilizing natural language processing techniques such as Named Entity Recognition (NER) to automatically identify sensitive information in text. For example, using pre-trained models like BERT to identify sensitive entities such as names and addresses in text.
[0005] Dictionary-based desensitization: Construct a sensitive dictionary and quickly identify sensitive information using string matching algorithms (such as the Aho-Corasick algorithm).
[0006] However, existing solutions are mostly designed for fixed types of sensitive data, and the de-identification strategies are relatively simple and rigid. When a new sensitive data type or a new de-identification algorithm is needed, it often requires in-depth modification of the system code, resulting in poor scalability and difficulty in quickly responding to constantly changing data types, business scenarios and security requirements. Summary of the Invention
[0007] This invention provides a data anonymization method, apparatus, device, and medium, which solves the problem of scalability in data anonymization methods.
[0008] To achieve the above objectives, this application adopts the following technical solution: Firstly, a data anonymization method is provided, including: Configure a unified desensitization abstraction layer to extensibly define and encapsulate interfaces for various sensitive information identification modes and various desensitization strategies; Receive data to be processed; in response to the data to be processed, optionally invoke at least one sensitive information identification mode through the desensitization abstraction layer to identify sensitive data fields and their types in the data to be processed; Based on the mapping relationship between the type of sensitive data fields and the pre-configured desensitization strategy, the desensitization abstraction layer matches the corresponding desensitization strategy for each identified sensitive data field. According to the matched desensitization strategy, the sensitive data fields are desensitized to generate desensitized data; The desensitization effect of the desensitized data was verified.
[0009] Secondly, a data anonymization device is provided, comprising: The abstract interface module is used to configure a unified desensitization abstraction layer, and to define and encapsulate interfaces for various sensitive information identification modes and various desensitization strategies in an extensible manner. The identification module is used to receive data to be processed; in response to the data to be processed, at least one sensitive information identification mode is optionally invoked through the desensitization abstraction layer to identify sensitive data fields and their types in the data to be processed; The strategy matching module is used to match the corresponding desensitization strategy for each identified sensitive data field through the desensitization abstraction layer, based on the mapping relationship between the type of the sensitive data field and the desensitization strategy pre-configured. The desensitization processing module is used to perform desensitization operations on the sensitive data fields according to the matched desensitization strategy, and generate desensitized data; The desensitization effect verification module is used to verify the desensitization effect of the desensitized data.
[0010] Thirdly, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the data desensitization method as described in the first aspect.
[0011] Fourthly, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the data desensitization method as described in the first aspect. Attached Figure Description
[0012] Figure 1 This is a schematic flowchart illustrating a data anonymization method provided in an embodiment of this application. Detailed Implementation
[0013] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions in the embodiments of this application are clearly described. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art are within the scope of protection of this application.
[0014] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.
[0015] The steps described in this application and the flowcharts in the accompanying drawings are not necessarily strictly executed according to the step numbers; the execution order of the steps can be changed. Furthermore, certain steps can be omitted, multiple steps can be combined into one step, and / or one step can be broken down into multiple steps.
[0016] In the context of big data, data anonymization faces multiple challenges: Diverse data types: covering structured data (such as database tables), semi-structured data (such as JSON and XML), and unstructured data (such as text and images).
[0017] The de-identification strategy is becoming more complex: de-identification methods need to be flexibly selected and combined according to data types, business scenarios and security requirements.
[0018] High performance requirements: With large amounts of data, the de-identification process needs to be highly efficient to avoid becoming a bottleneck in data processing.
[0019] Verifying the effectiveness of data anonymization is difficult: it is necessary to ensure that the anonymized data can effectively protect privacy while meeting the needs of business analysis.
[0020] Based on this, this specification provides a data desensitization method, and also relates to a data desensitization device, a computer device, and a computer-readable storage medium, which will be described in detail below with reference to the accompanying drawings and preferred embodiments.
[0021] Please see Figure 1 This application provides a data anonymization method, including: Step S1: Configure a unified desensitization abstraction layer, which is used to define and encapsulate interfaces for various sensitive information identification modes and various desensitization strategies in an extensible manner.
[0022] Step S2: Receive data to be processed; in response to the data to be processed, optionally invoke at least one sensitive information identification mode through the desensitization abstraction layer to identify sensitive data fields and their types in the data to be processed.
[0023] Step S3: Based on the mapping relationship between the type of sensitive data field and the pre-configured desensitization strategy, the desensitization abstraction layer matches the corresponding desensitization strategy for each identified sensitive data field.
[0024] Step S4: Perform desensitization operation on the sensitive data fields according to the matched desensitization strategy to generate desensitized data.
[0025] Step S5: Verify the desensitization effect of the desensitized data.
[0026] This invention encapsulates diverse sensitive information identification patterns (such as regular expressions, trie, and NER) and desensitization strategies (such as masking, hashing, forgery, and generation) by designing a unified desensitization abstraction layer and standardized interfaces. This allows the system to easily integrate new algorithms, significantly reducing the complexity of system integration and maintenance, and enabling rapid adaptation to new data types and compliance requirements in the future. Simultaneously, it can select the most suitable identification method based on data characteristics, effectively balancing processing speed and identification accuracy. The various desensitization strategies provided can be flexibly configured and combined according to different business scenarios (such as development and testing, data analysis, and data sharing) and security levels, achieving an optimal balance between security and data availability.
[0027] Furthermore, the sensitive information identification pattern includes: Regular expression-based recognition patterns are used to identify sensitive data with fixed formats. Pattern recognition based on Aho-Corasick dual-array trie is used for multi-pattern matching using a pre-built sensitive dictionary; The recognition pattern based on the named entity recognition model is used to identify sensitive entities in unstructured text.
[0028] The desensitization strategy includes: Masking desensitization strategy: Replace some characters in sensitive data fields with preset mask characters; Hash-based data masking strategy uses a hash algorithm to convert sensitive data fields into irreversible hash values; The Fake desensitization strategy replaces sensitive data fields with simulated data that conforms to the original data format and semantic features; The Generator generates desensitization strategies by creating simulated data based on the characteristic distribution of the original data to replace sensitive data fields.
[0029] The unified desensitization abstraction layer configuration includes: Define an abstract base class that provides a unified initialization interface, a desensitization processing interface, and a result assembly interface; The regular expression-based recognition class, the trie-based recognition class, and the named entity recognition class inherit from the abstract base class and implement their respective recognition logic. The masking desensitization strategy, hash desensitization strategy, forgery desensitization strategy, and generation desensitization strategy are adapted to the desensitization abstraction layer.
[0030] Specifically, this application provides a unified abstract base class, SensitivePattern, as the base class for all de-identification algorithms. This base class defines a common interface and process for de-identification processing, including methods for initialization configuration, de-identification processing, and result assembly. Through this abstraction layer, all specific de-identification algorithms adhere to a unified interface specification, facilitating integration and management. An example is shown below: public abstract class SensitivePattern { / / Salinity, used to increase the safety of desensitization. protected String salt; / / Configuration information for precise control of desensitization behavior protected List <config>configs; / / Unified desensitization interface method public abstract String desensitive(String text, boolean longText); / / Result assembly method, used to assemble the recognition results into the final de-identified text. public static String processHIT(List <hit>res, String text); } Regular expression-based recognition: Implemented by inheriting the abstract class `SensitivePatternRegex`, this method is suitable for recognizing sensitive data with fixed formats, such as ID card numbers and mobile phone numbers. This class defines the regular expression for the sensitive pattern and the reassembled format. An example is shown below: public abstract class SensitivePatternRegex extends SensitivePattern{ / / Define sensitive pattern regular expression public abstract String pattern(); / / Define the reassembled format public abstract String target(String source); } Based on trie recognition, this class uses the Aho-Corasick double-array trie algorithm, suitable for keyword matching scenarios such as sensitive word filtering and name recognition. It constructs a trie by loading a sensitive word dictionary, achieving efficient multi-pattern matching.
[0031] Named Entity Recognition (NER) based recognition is suitable for identifying sensitive information in complex text, such as addresses and organization names. This method uses a pre-trained NER model to identify sensitive entities in the text and then performs de-identification processing.
[0032] Masking and desensitization hides sensitive information by replacing some characters with specific symbols (such as *). For example, the phone number "1xx12345678" can be masked as "1xx****5678". This strategy is suitable for scenarios where it is necessary to preserve the data format but hide some information.
[0033] Hash-based data masking uses an irreversible hash algorithm (such as MD5) to convert sensitive information into hash values, making it suitable for scenarios where data consistency needs to be maintained. For example, when sharing data, hash values can be compared to verify data consistency.
[0034] Fake data anonymization involves replacing original sensitive data with fake but seemingly real values while preserving the data's format and semantic characteristics. For example, replacing a real name with a randomly generated name while maintaining the name's length and format. This strategy is suitable for scenarios where it's necessary to retain the statistical properties of the data.
[0035] Data anonymization is generated by creating new simulated data based on the original data, suitable for test data generation scenarios. For example, simulated age data can be generated based on the age distribution of real users for system testing.
[0036] Furthermore, for multiple sensitive data fields that are related, assign them related desensitization labels; The associated desensitization tag is used as a seed to be associated with the matching logic of the desensitization strategy, so that all sensitive data fields with the same associated desensitization tag within the same data record or across data records will ultimately obtain desensitization results based on the same desensitization strategy.
[0037] A configuration management mechanism supports associated data masking, ensuring consistency of related fields within the same record after masking. For example, fields such as a person's name, ID number, and mobile phone number can use the same seed for masking, guaranteeing the correlation of the masked data. This mechanism is implemented by maintaining a global seed mapping table, ensuring that related fields use the same random seed during the masking process.
[0038] Further, step S3, based on the mapping relationship between the type of sensitive data fields and the pre-configured de-identification strategy, matches a corresponding de-identification strategy for each identified sensitive data field through the de-identification abstraction layer, specifically including: Step S301: Match the types of the identified sensitive data fields with a preset expert knowledge base, which stores the mapping relationship between data types and recommended desensitization strategies; Step S302: Based on the matching results, automatically recommend at least one desensitization strategy for the sensitive data fields.
[0039] In some possible implementations, machine learning models can be used to replace expert knowledge bases, automatically identifying sensitive data and recommending desensitization strategies; the advantage is that they are highly adaptable; however, the disadvantage is that they require a large amount of labeled data for training.
[0040] Further, step S5 verifies the desensitization effect of the desensitized data, including: Rule configuration verification: Check whether corresponding de-identification strategies have been configured for the types of all sensitive data fields; Desensitization verification: Check whether all identified sensitive data fields have been desensitized. Secondary identification and verification: Perform a second identification of sensitive data fields to check whether sensitive data fields still exist.
[0041] In this step, the rule configuration verification checks whether de-identification rules have been configured for all sensitive fields to ensure no omissions; the de-identification execution verification checks whether all sensitive fields have been de-identified to ensure no omissions; and the secondary recognition verification performs a secondary recognition on the de-identified results to confirm whether de-identification is complete and to ensure the quality of de-identification. In some possible implementations, regular expression matching can be used instead of secondary recognition, which has the advantage of being simple to implement, but the disadvantage is that it cannot cover all sensitive patterns.
[0042] Furthermore, the data anonymization method can optionally be executed in a static or dynamic manner. Static anonymization involves anonymizing data stored in batches in a non-production environment; dynamic anonymization involves anonymizing the results returned by data queries or access requests in real time in a production environment.
[0043] In practice, the masking SDK supports both static masking via Spark and dynamic masking within the application. In some possible implementations, Flink can also be used instead of Spark for static masking; its advantage is strong streaming processing capabilities, but its disadvantage is that its ecosystem is not as robust as Spark's.
[0044] In addition, the following performance optimization measures are provided: 1. Trie algorithm optimization: The Aho-Corasick double-array trie algorithm is used to improve the efficiency of multi-pattern matching and reduce the matching time.
[0045] 2. Caching mechanism: Commonly used de-identification rules and configurations are cached to reduce redundant calculations and improve processing speed.
[0046] 3. Parallel processing support: Supports parallel de-identification processing of large-scale data, utilizing multi-core processors to improve processing efficiency.
[0047] Based on the above solution, this invention, in addition to achieving high integration and scalability and improving the flexibility and accuracy of data masking, ensures consistent masking of related sensitive information of the same entity scattered across different tables and fields by introducing associated masking tags and a seed mechanism. This fundamentally solves the data association breakage problem caused by independent masking in existing technologies, greatly improving the usability and value of masked data in downstream businesses such as testing and analysis. It provides a complete masking effect verification mechanism, ensuring the integrity and effectiveness of the masking operation, significantly reducing the risk of sensitive information leakage due to human error or policy defects, and improving the overall data security level. In scenarios with massive data and high concurrency access, it can meet both the throughput requirements of static batch masking and the low latency requirements of dynamic real-time masking.
[0048] Corresponding to the above-described data anonymization method embodiments, this application provides a data anonymization apparatus, comprising: The abstract interface module is used to configure a unified desensitization abstraction layer, and to define and encapsulate interfaces for various sensitive information identification modes and desensitization strategies in an extensible manner.
[0049] The identification module is used to receive data to be processed; in response to the data to be processed, it can optionally call at least one sensitive information identification mode through the desensitization abstraction layer to identify sensitive data fields and their types in the data to be processed.
[0050] The strategy matching module is used to match the corresponding desensitization strategy for each identified sensitive data field through the desensitization abstraction layer, based on the mapping relationship between the type of the sensitive data field and the desensitization strategy pre-configured.
[0051] The desensitization module is used to perform desensitization operations on the sensitive data fields according to the matched desensitization strategy, and generate desensitized data.
[0052] The desensitization effect verification module is used to verify the desensitization effect of the desensitized data.
[0053] Furthermore, the sensitive information identification pattern includes: Recognition patterns based on regular expressions, recognition patterns based on Aho-Corasick double-array trie, and recognition patterns based on named entity recognition models; The desensitization strategy includes: Masking desensitization strategy, hash desensitization strategy, forgery desensitization strategy, and generation desensitization strategy; The abstract interface module is specifically used for: Define an abstract base class that provides a unified initialization interface, a desensitization processing interface, and a result assembly interface; The regular expression-based recognition class, the trie-based recognition class, and the named entity recognition class inherit from the abstract base class and implement their respective recognition logic. The masking desensitization strategy, hash desensitization strategy, forgery desensitization strategy, and generation desensitization strategy are adapted to the desensitization abstraction layer.
[0054] Furthermore, the strategy matching module is specifically used for: The types of the identified sensitive data fields are matched with a pre-set expert knowledge base, which stores the mapping relationship between data types and recommended de-identification strategies. Based on the matching results, at least one de-identification strategy is automatically recommended for the sensitive data fields.
[0055] Furthermore, the strategy matching module is also used for: For multiple sensitive data fields that are related, assign them related desensitization labels; The associated desensitization tag is used as a seed to be associated with the matching logic of the desensitization strategy, so that all sensitive data fields with the same associated desensitization tag within the same data record or across data records will ultimately obtain desensitization results based on the same desensitization strategy.
[0056] Furthermore, the desensitization effect verification module is specifically used for: Rule configuration verification: Check whether corresponding de-identification strategies have been configured for the types of all sensitive data fields; Desensitization verification: Check whether all identified sensitive data fields have been desensitized. Secondary identification and verification: Perform a second identification of sensitive data fields to check whether sensitive data fields still exist.
[0057] The above-described data desensitization device implements the steps and processes of the above-described data desensitization method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.
[0058] Corresponding to the above-described data anonymization method embodiments, this application provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps and processes of the above-described data anonymization method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.
[0059] Memory can be used to store software programs and various data. Memory can primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area can store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, memory can include volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM). The memory in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.
[0060] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and applications, while the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor.
[0061] Corresponding to the above-described data desensitization method embodiments, this application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the steps and processes of the above-described data desensitization method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.
[0062] The processor is the processor in the electronic device described in the above embodiments of this application. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.
[0063] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
[0064] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0065] It is understood that the embodiments of this application have been described above in conjunction with the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. As those skilled in the art will know, various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, those skilled in the art, under the guidance or instruction of this application, can modify these features and embodiments to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, this invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of this invention.< / hit> < / config>
Claims
1. A data anonymization method, characterized in that, include: Configure a unified desensitization abstraction layer to extensibly define and encapsulate interfaces for various sensitive information identification modes and various desensitization strategies; Receive data to be processed; In response to the data to be processed, at least one sensitive information identification mode can be optionally invoked through the desensitization abstraction layer to identify the sensitive data fields and their types in the data to be processed; Based on the mapping relationship between the type of sensitive data fields and the pre-configured desensitization strategy, the desensitization abstraction layer matches the corresponding desensitization strategy for each identified sensitive data field. According to the matched desensitization strategy, the sensitive data fields are desensitized to generate desensitized data; The desensitization effect of the desensitized data was verified.
2. The data anonymization method according to claim 1, characterized in that, The sensitive information identification mode includes: Recognition patterns based on regular expressions, recognition patterns based on Aho-Corasick double-array trie, and recognition patterns based on named entity recognition models; The desensitization strategy includes: Masking desensitization strategy, hash desensitization strategy, forgery desensitization strategy, and generation desensitization strategy; The unified desensitization abstraction layer configuration includes: Define an abstract base class that provides a unified initialization interface, a desensitization processing interface, and a result assembly interface; The regular expression-based recognition class, the trie-based recognition class, and the named entity recognition class inherit from the abstract base class and implement their respective recognition logic. The masking desensitization strategy, hash desensitization strategy, forgery desensitization strategy, and generation desensitization strategy are adapted to the desensitization abstraction layer.
3. The data anonymization method according to claim 1, characterized in that, For multiple sensitive data fields that are related, assign them related desensitization labels; The associated desensitization tag is used as a seed to be associated with the matching logic of the desensitization strategy, so that all sensitive data fields with the same associated desensitization tag within the same data record or across data records will ultimately obtain desensitization results based on the same desensitization strategy.
4. The data anonymization method according to claim 1, characterized in that, The process of matching a corresponding desensitization strategy to each identified sensitive data field through the desensitization abstraction layer, based on the mapping relationship between the type of the sensitive data field and the pre-configured desensitization strategy, includes: The types of the identified sensitive data fields are matched with a pre-set expert knowledge base, which stores the mapping relationship between data types and recommended de-identification strategies. Based on the matching results, at least one de-identification strategy is automatically recommended for the sensitive data fields.
5. The data anonymization method according to claim 1, characterized in that, The verification of the desensitization effect of the desensitized data includes: Rule configuration verification: Check whether corresponding de-identification strategies have been configured for the types of all sensitive data fields; Desensitization verification: Check whether all identified sensitive data fields have been desensitized. Secondary identification and verification: Perform a second identification of sensitive data fields to check whether sensitive data fields still exist.
6. The data anonymization method according to claim 1, characterized in that, The data desensitization method can be performed in either static or dynamic desensitization mode.
7. A data anonymization device, characterized in that, include: The abstract interface module is used to configure a unified desensitization abstraction layer, and to define and encapsulate interfaces for various sensitive information identification modes and various desensitization strategies in an extensible manner. The identification module is used to receive data to be processed; In response to the data to be processed, at least one sensitive information identification mode can be optionally invoked through the desensitization abstraction layer to identify the sensitive data fields and their types in the data to be processed; The strategy matching module is used to match the corresponding desensitization strategy for each identified sensitive data field through the desensitization abstraction layer, based on the mapping relationship between the type of the sensitive data field and the desensitization strategy pre-configured. The desensitization processing module is used to perform desensitization operations on the sensitive data fields according to the matched desensitization strategy, and generate desensitized data; The desensitization effect verification module is used to verify the desensitization effect of the desensitized data.
8. The data desensitization device according to claim 7, characterized in that, The strategy matching module is also used for: For multiple sensitive data fields that are related, assign them related desensitization labels; The associated desensitization tag is used as a seed to be associated with the matching logic of the desensitization strategy, so that all sensitive data fields with the same associated desensitization tag within the same data record or across data records will ultimately obtain desensitization results based on the same desensitization strategy.
9. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the data desensitization method as described in any one of claims 1 to 6.
10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the data desensitization method as described in any one of claims 1 to 6.