Data desensitization method and device for consanguinity algorithm, medium and product

Abstract

The invention provides a data desensitization method for a consanguinity algorithm, and the method comprises the steps: determining the association strength of a target enterprise and an associated enterprise based on the association information of the target enterprise and the associated enterprise; on the basis of enterprise information of the target enterprise, the association information and the association strength, an association value is calculated, and a blood relationship ID is generated and stored; based on the association strength, desensitizing the enterprise information of the target enterprise and the association information by using different desensitization rules; and performing correlation value verification on the desensitized enterprise information and the correlation information.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a data desensitization method, device, medium, and product for lineage algorithms. Background Technology

[0002] In credit reporting, data platforms need to process massive amounts of data, such as highly sensitive information like personal identification information, corporate equity structures, and transaction information. This data must strictly comply with relevant laws and regulations. Therefore, when conducting credit reporting, this sensitive information must be de-identified or anonymized—a process known as data desensitization—to prevent privacy leaks.

[0003] Currently, traditional data anonymization methods have the following problems.

[0004] The core contradiction between data anonymization and association: Traditional data anonymization techniques (such as fixed masks, hashing, generalization, etc.) have a polarization problem. Either excessive anonymization destroys data association (e.g., after anonymizing corporate shareholder information, it cannot be matched with the equity penetration relationship of related companies, resulting in a break in the lineage algorithm model); or the anonymization intensity is reduced to retain association, which leads to the risk of privacy leakage.

[0005] Lack of dynamic adaptation mechanism: Existing desensitization methods are static rules that do not take into account the differences in the strength of blood relations (such as direct shareholding versus indirect shareholding). The uniform desensitization rules lead to "breaks in important relationships and insufficient privacy protection for secondary relationships", which cannot meet the core needs of the credit reporting platform for "enterprise relationship investigation".

[0006] Lack of consistency verification: There is no effective mechanism to verify the integrity of blood relations after desensitization. If the desensitization operation causes the relationship to break, manual investigation and repair are required, which is inefficient and prone to omissions, affecting the accuracy of downstream businesses such as credit reports and risk monitoring. Summary of the Invention

[0007] To address the contradiction that data anonymization in lineage algorithms may disrupt data correlation, this application proposes a data anonymization method for lineage algorithms.

[0008] The data anonymization method includes: determining the association strength between the target company and the associated companies based on the association information between the target company and the associated companies; calculating the association value and generating and storing the lineage ID based on the company information of the target company, the association information, and the association strength; anonymizing the company information of the target company and the association information using different anonymization rules based on the association strength; and verifying the association value of the anonymized company information and the association information.

[0009] Optionally, the correlation strength includes strong correlation and weak correlation.

[0010] Optionally, the correlation value is calculated using the following formula: Rtotal = R1 × W1 + R2 × W2, where Rtotal is the correlation value, R1 is the equity correlation value, which is determined based on the equity percentage, R2 is the position correlation value, which is determined based on the position held, W1 is the weight of the equity correlation value, and W2 is the weight of the position correlation value.

[0011] Optionally, the lineage ID includes the identifier of the target enterprise, the association type code, the timestamp of generating the lineage ID, and a check digit.

[0012] Optionally, the identifier of the target enterprise is the unified social credit code; the association type code indicates the association method between the target enterprise and the associated enterprise, including equity (GQ), employment (RZ), and industry chain (CY); the check bit is obtained by performing a hash operation on the unified social credit code and the association type code, and taking a preset decimal number from the resulting hash value.

[0013] Optionally, when the association strength is strong, partial masking and specific value range-based desensitization are used for desensitization; when the association strength is weak, full masking is used for desensitization.

[0014] Optionally, the lineage ID is stored in a first database associated with the target company and a second database associated with the related company.

[0015] Optionally, the check bits of the lineage ID stored in the first database and the second database are compared.

[0016] This application also proposes an electronic device comprising a memory storing computer-executable instructions and a processor; when the instructions are executed by the processor, the device causes the device to perform the method according to any of the preceding claims.

[0017] This application also proposes a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the method described in any of the preceding claims.

[0018] This application also proposes a computer program product, including a computer program that, when executed by a processor, implements the method as described in any of the preceding claims.

[0019] This data anonymization method for kinship algorithms achieves compatibility between data correlation and data privacy, and can adaptively use different anonymization rules based on the strength of the correlation, as well as automatically verify the accuracy of the anonymized data. Attached Figure Description

[0020] Figure 1 This is a flowchart of a data desensitization method for a lineage algorithm proposed according to an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0022] The data desensitization method for the lineage algorithm proposed in this application will now be described in detail with reference to the accompanying drawings.

[0023] Figure 1 This is a flowchart of a data desensitization method for a lineage algorithm proposed according to an embodiment of this application.

[0024] In step 11, the strength of the association between the target company and its related companies is determined based on the association information between the target company and its related companies.

[0025] The target company refers to the target entity in the credit investigation business, and related companies refer to companies that have various connections with the target company. Connection information includes various related information such as the relationship between the parties. In this application, the relationship between the target company and related companies includes two types: actual control and employment. Actual control can be determined through equity penetration, while employment indicates that an employee of the target company holds a position in the related company. Therefore, the strength of the connection can be categorized as strong or weak. A strong connection indicates that the target company holds a shareholding in the related company exceeding a threshold percentage, or that an employee of the target company holds a key position in the related company. This threshold percentage is, for example, 30%, and key positions include, for example, legal representatives. A weak connection indicates that the target company holds a shareholding in the related company below the threshold percentage and / or that an employee holds a non-key position in the related company, or that there is only an employee holding a position in the related company without any equity connection.

[0026] In one example, Company A holds 51% of the shares in Company B, therefore Company A and Company B are strongly related. Although an employee of Company B serves as the legal representative of Company C, there is no equity relationship between Company B and Company C, therefore Company B and Company C are weakly related. Company C holds 5% of the shares in Company D, and employees of Company C hold ordinary positions in Company D, therefore Company C and Company D are weakly related.

[0027] Therefore, by combining the analysis of dual blood ties through equity penetration and employment relationships, data can be classified and protected according to the strength of the association in subsequent processes.

[0028] In step 12, based on the target company's enterprise information, association information, and association strength, the association value is calculated and the lineage ID is generated and stored.

[0029] In one embodiment, the association value is related to dual blood ties, and therefore only applies to situations where the target company and its affiliated companies have a strong relationship. Thus, the formula for calculating the association value can be Rtotal = R1 × W1 + R2 × W2. Where Rtotal is the association value, R1 is the equity association value (based on equity percentage), R2 is the employment association value (based on position), W1 is the weight of the equity association value, and W2 is the weight of the employment association value. Since equity control has a higher priority than employment influence in credit reporting scenarios, W1 is usually higher than W2, for example, W1 is 60% and W2 is 40%.

[0030] In one embodiment, the values ​​of equity-related value R1 and employment-related value R2 can be determined according to the table below.

[0031] Relationship Dimension Assignment Logic Equity Related Value R1 Shareholding percentage: 50% or more (absolute control): 9-10 points (9 points for 50%-67%, 10 points for over 67%); 30%-50% (relative control): 7-9 points; less than 30% (partial stake): 3-7 points Job-related value R2 Job Relevance: Key Positions (Legal Representative, Executive Director, Manager): 7-9 points (Directly impacts corporate decisions); Senior Management (Chief Financial Officer, Supervisor): 5-7 points; General Staff: 1-3 points

[0032] The lineage ID records the target company's own corporate information and its association information with related companies. In one embodiment, the lineage ID includes the target company's identifier, association type code, timestamp of the generation of the lineage ID, and a check digit. The target company's identifier can be a unique identifier such as the Unified Social Credit Code (USCC). The association type code indicates the association method between the target company and related companies, such as GQ for equity association, QZ for employment association, and CY for being in the same industry chain. The check digit is generated by hashing the target company's identifier (Unified Social Credit Code) and association type code, and then taking a preset number of decimal digits from the resulting hash value. For example, performing a SHA-256 hash operation on the USCC and association type code, and taking the last 8 decimal digits of the hash value as the check digit, thus covering up to one hundred million companies.

[0033] In relevant embodiments, the format of the lineage ID is USCC-Association Type Code-Timestamp-8-bit Check Bit. For example, the lineage ID associated with enterprise A and enterprise B is 91320000MA1YLUX123-GQ-20251001-15678902, and the lineage ID associated with enterprise B and enterprise C is 91320000MA1YLUX456-RZ-20251001-28934567.

[0034] After generating the lineage ID, it needs to be stored for subsequent relationship tracing and anonymization verification. The lineage ID can be stored in a first database associated with the target company and a second database associated with related companies.

[0035] In step 13, based on the association strength, different desensitization rules are used to desensitize the target company's corporate information and related information.

[0036] When the target company and its related companies are strongly related, since it is necessary to use relevant information to calculate the association value and other data used for credit scoring, partial masking and specific value range-based methods are used for anonymization, thus achieving compatibility between numerical calculation and privacy protection. For example, in the relationship between target company A and company B, the shareholder name of company A (as an example of company information) is masked as "Zhang San", and the specific value of 51% equity share is ranged as "50%-55%" (as an example of specific value range-based methods).

[0037] When the target company and its related companies have a weak relationship, the anonymization logic prioritizes privacy protection while also ensuring the integrity of the relationship. Therefore, a full mask is used to completely hide sensitive identity information, preventing the leakage of non-core related identity data. This adheres to the principle of "minimum necessary" use of credit data, thus avoiding privacy leaks. Furthermore, the potential value of the related data is ensured through lineage IDs. For example, in the relationship between target company B and company C, the legal representative of company B is anonymized as "***".

[0038] In addition, regardless of whether the relationship is strong or weak, the lineage ID must be fully retained in order to trace the lineage relationship of the enterprise in subsequent business.

[0039] In step 14, the association value is verified on the de-identified enterprise information and related information.

[0040] After data anonymization, it is also necessary to perform consistency verification based on the check bit of the bloodline ID for strong associations. The check bit of the bloodline ID stored in the first database is matched with the check bit of the bloodline ID stored in the second database to determine whether the bloodline ID has been tampered with or whether information has been lost during storage.

[0041] Furthermore, the anonymized data is used for numerical calculations required for corporate credit investigation, such as the correlation value Rtotal. Taking Company A as an example, since the anonymized shareholding ratio of Company A is 50%-55%, as shown in the aforementioned table, the score assigned to the shareholding ratio is consistent with that when the shareholding is 51%, both being 9 points. Only the names of the personnel in charge are anonymized. Therefore, the anonymized data can still be used for the correlation value calculation, and the data is not distorted.

[0042] When the above consistency checks and numerical calculation verifications fail, it is necessary to investigate whether there are any issues with the determination of the association strength. For example, a 51% equity stake may be considered a strong association, but in reality, the target company does not actually control the related company, or the anonymization process may have disrupted the core calculation factors for a strong association (e.g., equity range, leading to distortion of the Rtotal parameter). In this case, the target company's equity stake in the related company and its actual control relationship should be re-verified.

[0043] If, after verification, the correlation strength is deemed correct (i.e., a strong correlation is confirmed), and only the anonymization factor is distorted, the corresponding data should be adjusted. For example, the anonymized equity range of 50%-55% can be expanded to 50%-60%, while ensuring the integrity of the bloodline ID.

[0044] If, after verification, the association strength is deemed incorrect (i.e., it is actually a weak association), the strong association rule will be upgraded (i.e., the degree of desensitization will be increased) to a weak association rule. For example, the specific value range desensitization of the target company will be changed to full mask desensitization, retaining only the lineage ID.

[0045] Finally, the anonymized correlation values ​​are substituted into the credit risk scoring formula to prove that the anonymized data can still support the accurate calculation of the formula. The credit risk scoring formula is, for example, Risk_Score=100−[α×Rtotal×Risklink+β×Riskself+γ×(M−S) / 550×30]. Where Risk_Score is the risk control score, Risklink is the risk transmission coefficient of the related enterprise (e.g., Risklink=0.8 when the related enterprise has overdue payments, Risklink=1.2 when involved in litigation, and Risklink=0.1 when there is no risk), Riskself is the target enterprise's own risk value (e.g., 0-30 points, 15 points for litigation, 10 points for administrative penalties, and 5 points for transaction fluctuations, which can be accumulated), α, β, and γ are weights customized based on different credit reporting scenarios, S is the credit score of the target enterprise by the risk control platform, and M is the maximum value of this credit score. Because the anonymized data retains the data used for numerical values, without excessively anonymizing the data needed to calculate the risk control scoring formula, the anonymized data can still support the accurate calculation of the formula.

[0046] The aforementioned data anonymization methods achieve a dynamic balance between data privacy protection and the integrity of data relationships in the credit reporting field. By dynamically binding anonymization rules to objectively quantified kinship strength, the traditional one-size-fits-all static anonymization model is changed. It also ensures the quality and accuracy of data input for downstream business models, solving the problem of inaccurate or ineffective business models caused by traditional anonymization methods. Furthermore, verification and review enhance the reliability of the anonymization process.

[0047] Now for reference Figure 2 The diagram shown is a block diagram of an electronic device 200 according to an embodiment of the present application. The electronic device 200 may include one or more processors 202, system control logic 208 connected to at least one of the processors 202, system memory 204 connected to the system control logic 208, non-volatile memory (NVM) 206 connected to the system control logic 208, and network interface 210 connected to the system control logic 208.

[0048] Processor 202 may include one or more single-core or multi-core processors. Processor 202 may include any combination of general-purpose processors and special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In embodiments herein, processor 202 may be configured to perform one or more embodiments of various embodiments of the data desensitization method for lineage algorithms proposed in this application.

[0049] In some embodiments, system control logic 208 may include any suitable interface controller to provide any suitable interface to at least one of the processors 202 and / or any suitable device or component communicating with system control logic 208.

[0050] In some embodiments, system control logic 208 may include one or more memory controllers to provide an interface to system memory 204. System memory 204 may be used to load and store data and / or instructions. In some embodiments, system memory 204 of electronic device 200 may include any suitable volatile memory, such as suitable dynamic random access memory (DRAM).

[0051] The nonvolatile memory 206 may include one or more tangible, non-transitory computer-readable storage media for storing data and / or instructions. In some embodiments, the nonvolatile memory 206 may include any suitable nonvolatile memory such as flash memory and / or any suitable nonvolatile storage device, such as at least one of HDD (Hard Disk Drive), CD (Compact Disc) drive, and DVD (Digital Versatile Disc) drive.

[0052] The non-volatile memory 206 may include a portion of the storage resources installed on the device of the electronic device 200, or it may be accessible by the device but is not necessarily part of the device. For example, the non-volatile memory 206 may be accessed over a network via the network interface 210.

[0053] Specifically, system memory 204 and non-volatile memory 206 may each include a temporary copy and a permanent copy of instructions 220. Instructions 220 may include instructions that, when executed by at least one of processors 202, cause electronic device 200 to implement the methods provided in this application. In some embodiments, instructions 220, hardware, firmware, and / or their software components may additionally / alternatively reside in system control logic 208, network interface 210, and / or processor 202.

[0054] In some embodiments, network interface 210 may be integrated into other components of electronic device 200. For example, network interface 210 may be integrated into at least one of processor 202, system memory 204, non-volatile memory 206, and firmware device (not shown) with instructions that, when at least one of processor 202 executes the instructions, electronic device 200 implements one or more embodiments of the various embodiments of the data desensitization method for lineage algorithms described herein. Network interface 210 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface.

[0055] In one embodiment, at least one of the processors 202 may be packaged together with the logic of one or more controllers for system control logic 208 to form a system package (SiP). In another embodiment, at least one of the processors 202 may be integrated on the same die with the logic of one or more controllers for system control logic 208 to form a system on chip (SoC).

[0056] Electronic device 200 may further include: input / output (I / O) device 212. Input / output (I / O) device 212 may include a user interface that enables a user to interact with device 200; the design of peripheral component interfaces enables peripheral components to also interact with electronic device 200.

[0057] In some embodiments, the user interface may include, but is not limited to, a display (e.g., a liquid crystal display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., a light-emitting diode flash), and a keyboard.

[0058] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.

[0059] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the electronic device 200. In other embodiments of this application, the electronic device 200 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0060] Program code can be applied to input instructions to perform the functions described herein and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application-specific integrated circuit (ASIC), or a microprocessor.

[0061] The program code can be implemented using a high-level procedural language or an object-oriented programming language to communicate with the processing system. Assembly language or machine language can also be used when needed. In fact, the mechanisms described in this paper are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0062] One or more aspects of at least one embodiment can be implemented by representational instructions stored on a computer-readable storage medium, the instructions representing various logics in a processor, which, when read by a machine, cause the machine to create logic for performing the techniques described herein. These representations, referred to as “IP cores,” can be stored on a tangible computer-readable storage medium and provided to multiple customers or production facilities for loading into manufacturing machines that actually manufacture the logic or processor.

[0063] One embodiment of this application discloses a computer-readable storage medium storing one or more programs executable by one or more processors to implement the data desensitization method for lineage algorithm of this application.

[0064] One embodiment of this application discloses a computer program product, including a computer program that, when executed by a processor, implements the data desensitization method for lineage algorithm of this application.

[0065] The specific embodiments described above illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Although the description of this application is presented in conjunction with preferred embodiments, this does not mean that the features of this invention are limited to these embodiments. Furthermore, to avoid confusion or obscuring the focus of this application, some specific details will be omitted in the description. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0066] Furthermore, the various operations will be described as multiple discrete operations in a manner most conducive to understanding the illustrative embodiments; however, the order of description should not be construed as implying that these operations must depend on the order. In particular, these operations do not need to be performed in the order presented.

[0067] Unless the context otherwise specifies, the terms “contains,” “has,” and “includes” are synonyms. The phrase “A / B” means “A or B.” The phrase “A and / or B” means “(A and B) or (A or B).”

[0068] As used herein, the terms “module” or “unit” may refer to, be, or include: application-specific integrated circuits (ASICs), electronic circuits, (shared, dedicated, or group) processors and / or memories that execute one or more software or firmware programs, combinational logic circuits, and / or other suitable components that provide the described functionality.

[0069] In the accompanying drawings, certain structural or methodological features are shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be necessary. In some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. Furthermore, the inclusion of structural or methodological features in a particular figure does not imply that such features are required in all embodiments, and in some embodiments, these features may be omitted or may be combined with other features.

[0070] It should be understood that although terms such as "first," "second," etc., may be used herein to describe various units or data, these units or data should not be limited by these terms. These terms are used merely to distinguish one feature from another. For example, without departing from the scope of the exemplary embodiments, a first feature may be referred to as a second feature, and similarly, a second feature may be referred to as a first feature.

[0071] It should be noted that in this specification, similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0072] Although the invention has been illustrated and described with reference to certain preferred embodiments thereof, those skilled in the art will understand that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.

Claims

1. A data anonymization method for lineage algorithms, characterized in that, include: Based on the association information between the target company and its affiliated companies, the strength of the association between the target company and its affiliated companies is determined; Based on the target company's enterprise information, the association information, and the association strength, calculate the association value and generate and store the lineage ID; Based on the association strength, different desensitization rules are used to desensitize the target company's enterprise information and the association information; The association value is verified for the anonymized enterprise information and the associated information.

2. The data anonymization method according to claim 1, characterized in that, The correlation strength includes strong correlation and weak correlation.

3. The data anonymization method according to claim 2, characterized in that, The correlation value is calculated using the following formula: Rtotal = R1 × W1 + R2 × W2, Wherein, Rtotal is the correlation value, R1 is the equity correlation value, which is determined according to the equity percentage, R2 is the position correlation value, which is determined according to the position held, W1 is the weight of the equity correlation value, and W2 is the weight of the position correlation value.

4. The data anonymization method according to claim 3, characterized in that, The lineage ID includes the identifier of the target enterprise, the association type code, the timestamp of the generation of the lineage ID, and a check digit.

5. The data anonymization method according to claim 4, characterized in that, The target enterprise is identified by its unified social credit code. The association type code indicates the association method between the target enterprise and the associated enterprise, including equity (GQ), employment (RZ), and industry chain (CY). The verification bit is obtained by performing a hash operation on the unified social credit code and the associated type code, and taking a preset decimal number from the resulting hash value.

6. The data anonymization method according to claim 5, characterized in that, When the association strength is strong, partial masking and specific value range-based desensitization are used for desensitization; when the association strength is weak, full masking is used for desensitization.

7. The data anonymization method according to claim 6, characterized in that, The lineage IDs are stored in a first database associated with the target company and a second database associated with the related company.

8. The data anonymization method according to claim 7, characterized in that, The check digits of the bloodline ID stored in the first database and the second database are compared.

9. An electronic device, characterized in that, The device includes a memory storing computer-executable instructions and a processor; when the instructions are executed by the processor, the device performs the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the method of any one of claims 1 to 8.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 8.

Images