Data protection processing method, apparatus, device, and storage medium

Abstract

The application discloses a data protection processing method and device, equipment and a storage medium, and belongs to the field of data processing. The method comprises the following steps: obtaining business data of multiple real businesses, extracting data items from the business data according to data domains in a data category; obtaining an initial data domain correlation result parameter based on the data items corresponding to each real business, wherein the initial data domain correlation result parameter represents the correlation between the data domains in the business data; calling a generator and a discriminator to perform data processing iteration based on the initial data domain correlation result parameter until the data processing iteration meets an iteration termination condition; and selecting synthetic data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration as analysis base data for business analysis, wherein the data domain correlation result parameter represents the relationship between the data domains in the data input into the discriminator. According to the embodiment of the application, the data security can be improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of data processing, and particularly relates to a data protection processing method and device, equipment and a storage medium. BACKGROUND

[0002] With the continuous development of digital technology, data has become one of the important production factors. More and more business analysis needs to rely on a large amount of data as analysis basis data for modeling or other operations. However, the data as analysis basis data may involve data privacy. In order to ensure the privacy and security of data, the data can be protected by encryption. However, the encrypted data can still be traced back to the source data, and if the encrypted data is leaked, the data privacy and security will still be harmed, and the data security will be reduced. SUMMARY

[0003] The embodiments of the present application provide a data protection processing method, device, equipment and storage medium, which can improve the data security.

[0004] In a first aspect, the embodiments of the present application provide a data protection processing method, comprising: obtaining business data of a plurality of real businesses, and extracting data items from the business data according to data domains in a data category; obtaining an initial data domain correlation result parameter based on the data items corresponding to each real business, the initial data domain correlation result parameter being used to represent the correlation relationship between the data domains in the business data; calling a generator and a discriminator to perform data processing iteration based on the initial data domain correlation result parameter until the data processing iteration meets an iteration stop condition, and selecting synthetic data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration as analysis basis data for business analysis, the data domain correlation result parameter being used to represent the relationship between the data domains in the input data of the discriminator.

[0005] In a second aspect, the embodiments of the present application provide a data protection processing device, comprising: an extraction module configured to obtain business data of a plurality of real businesses, and extract data items from the business data according to data domains in a data category; a calculation module configured to obtain an initial data domain correlation result parameter based on the data items corresponding to each real business, the initial data domain correlation result parameter being used to represent the correlation relationship between the data domains in the business data; and a training calling module configured to call a generator and a discriminator to perform data processing iteration based on the initial data domain correlation result parameter until the data processing iteration meets an iteration stop condition, and select synthetic data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration as analysis basis data for business analysis, the data domain correlation result parameter being used to represent the relationship between the data domains in the input data of the discriminator.

[0006] In a third aspect, an electronic device is provided, and includes a processor and a memory storing computer program instructions; and the processor implements the data protection processing method of the first aspect when executing the computer program instructions.

[0007] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores computer program instructions; and the computer program instructions are executed by a processor to implement the data protection processing method of the first aspect.

[0008] The embodiments of the present application provide a data protection processing method and device, equipment and a storage medium. The data protection processing method can be used for processing data according to the correlation between data domains in real business data, using a generator and a discriminator to perform data processing iteration, so that the generator can generate synthetic data according to the data domain correlation result parameter output by the discriminator, which represents the correlation between data domains of the data, the discriminator outputs the data domain correlation result parameter according to the synthetic data to update the generator, so that the generator outputs new synthetic data, and the process is repeated until the data processing iteration meets the iteration processing condition, and the synthetic data output by the generator is selected as analysis base data for business analysis. The synthetic data is obtained by the generator and the discriminator, and has similar appearance, meaning and statistical characteristics to the real business data, but is not real business data, and there is no real business corresponding to the synthetic data. The synthetic data is used as analysis base data for business analysis, which can meet the demand of business analysis, avoid the leakage of private data, and improve the data security. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. Those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0010] Figure 1 The flow chart of the data protection processing method provided by an embodiment of the present application;

[0011] Figure 2 The flow chart of the data protection processing method provided by another embodiment of the present application;

[0012] Figure 3 The flow chart of an example of the data protection processing flow provided by an embodiment of the present application;

[0013] Figure 4 The structural schematic diagram of the data protection processing device provided by an embodiment of the present application;

[0014] Figure 5 The structural schematic diagram of the data protection processing device provided by another embodiment of the present application;

[0015] Figure 6 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 1. DETAILED DESCRIPTION

[0016] In order to make the purposes, technical solutions and advantages of the present application clearer, the following further describes the present application in detail with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, but not to limit the present application. The present application can be implemented without some of the specific details by those skilled in the art. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0017] With the continuous development of digital technology, data has become one of the important production factors. More and more business analysis needs to rely on a large amount of data as analysis basis data for modeling or other operations. However, the data as analysis basis data may involve data privacy. In order to ensure the privacy and security of the analysis basis data, the analysis basis data can be protected by encryption. However, the encrypted analysis basis data can still be traced back to the source data. If the encrypted analysis basis data is leaked, the data privacy and security will still be endangered, and the data security will be reduced.

[0018] The present application provides a data protection processing method, device and equipment and a storage medium, which can use a generator and a discriminator to perform data processing iteration according to the correlation of the data domains in the business data of the real business, and through mutual updating of the generator and the discriminator, the generator can output synthetic data of the real business that is similar to the appearance, meaning and statistical characteristics of the business data of the real business, but does not exist and does not correspond to the real business. According to the synthetic data, the business data of the real business cannot be traced back, that is, the real data cannot be traced back, the data privacy and security can be protected, and the data security can be improved.

[0019] The data protection processing method, device, equipment and storage medium provided by the present application are described below respectively.

[0020] The first aspect of the present application provides a data protection processing method, which can be applied to a scenario of providing analysis basis data for business analysis and other operations. The data protection processing method is executed by a data protection processing device, equipment and the like, and the type and number of the data protection processing device, equipment and the like are not limited herein. Figure 1 A flowchart of the data protection processing method provided by an embodiment of the present application is shown in FIG. 2, which can include steps S101 to S103. Figure 1

[0021] ​In step S101, the business data of multiple real businesses is acquired, and data items are extracted from the business data according to data domains in data categories.

[0022] The business data of the real business is the business data generated by the real business, and can be regarded as source data. For example, the real business is a transaction business, and the transaction data of the transaction business can include user age data, transaction amount data, transaction region data, transaction currency data, transaction category data, and data about whether a coupon is used, etc. If these data are leaked, risks to data security can be caused.

[0023] The business data of each real business can include multiple data items, and at least part of the data items have certain correlation. The data items can be extracted from the business data according to data domains in data categories. The business data can include data of at least one data category. The data category can be a category on a data structure, for example, the data category can include one or more than two of the following: Boolean type, numerical type, classification type, and continuous character type. The data of the Boolean type can include data having only two values, such as data with values of true or false, or data with values of 0 or 1. The data of the numerical type includes data with values that can be represented by numbers, such as user age and transaction amount. The data of the classification type includes data with values that can be classified into several categories, such as transaction region and transaction currency. The data of the continuous character type includes data with values that can be represented by continuous characters, such as data with values including language and character descriptions. Each data category can include at least one data domain, and the data domain can be understood as an item under the data category. For example, the data domain can include user age, transaction amount, transaction region, transaction currency, transaction category, and whether a coupon is used, etc., which are not limited herein. The data items can be extracted from the business data according to the data domains. The data item can be regarded as the smallest meaningful data segment in the data.

[0024] In some examples, the data category includes continuous character types. Since data of continuous character types is not convenient for extracting data items, the data of continuous character types can be split to obtain data of data subcategories, and then data items are extracted from the data of data subcategories. Specifically, in the case where the data category includes continuous character types, data of continuous character types in the business data can be split according to data fields in the data subcategories to obtain data items. The data of continuous character types can be split into data of at least one data subcategory. Each data subcategory includes at least one data field. The data subcategory includes one or more of the following: a Boolean type, a numerical type, and a classification type. The Boolean type, the numerical type, and the classification type can refer to the related descriptions in the above embodiments, which will not be repeated here. For example, the continuous character type is a transaction description, and the data of the transaction description is "XX City XX District XX Avenue Debit Card Consumption". The data of the transaction description can be split into data items corresponding to data fields of data subcategories, and the split data items include "XX City", "XX District", "XX Avenue", and "Debit Card Consumption". The data field of "XX City", "XX District", and "XX Avenue" is a transaction area, and the data field of "Debit Card Consumption" is a transaction type. The data items obtained according to the data fields in the data category also include data items split according to the data fields in the data subcategory, that is, the data items split according to the data fields in the data subcategory also participate in the operation of the initial data field association result parameter and the data field association result parameter.

[0025] In some examples, in the case where the data category includes continuous character types, the data of continuous character types can be first screened according to a preset continuous character screening condition. The data of continuous character types that meet the continuous character screening condition is retained, and the data of continuous character types that do not meet the continuous character screening condition is discarded. The continuous character screening condition is used to determine whether the data of continuous character types is data required for business analysis. The data of continuous character types that meet the continuous character screening condition indicates that the data of continuous character types is data required for business analysis and has a beneficial effect on business analysis; the data of continuous character types that do not meet the continuous character screening condition indicates that the data of continuous character types is not data required for business analysis and has no beneficial effect on business analysis, and does not need to participate in the extraction of data items.

[0026] In step S102, based on the data items corresponding to each real business, an initial data field association result parameter is obtained.

[0027] There is a certain association relationship between at least part of the data items corresponding to each real business. The association relationship between the data domains to which the data items corresponding to each real business belong can be obtained according to the association relationship between the data items of the business data of each real business. Then, the association relationship between the data domains in the business data in the case of multiple businesses can be obtained according to the association relationship between the data domains to which the data items corresponding to each real business belong. The initial data domain association result parameter is used to represent the association relationship between the data domains in the business data. The association relationship here can include relationship type, association degree, and other information that can describe the association relationship. The initial data domain association result parameter can be implemented as characters, strings, tables, relationship views, etc., which are not limited here.

[0028] In step S103, the generator and the discriminator perform data processing iteration based on the initial data domain association result parameter until the data processing iteration meets the iteration stopping condition. The synthetic data generated by the generator configured with the data domain association result parameter obtained by the discriminator in the last data processing iteration is selected as the analysis basis data for business analysis.

[0029] The initial data domain association result parameter can be used as the model parameter of the generator, so that the generator generates synthetic data according to the initial data domain association result parameter. The synthetic data is input into the discriminator as input data of the discriminator. The discriminator generates a data domain association result parameter according to the input synthetic data. The data domain association result parameter is used to represent the relationship between the data domains in the data input into the discriminator. In the case where the data processing iteration does not meet the iteration stopping condition, the data domain association result parameter generated by the discriminator can be transmitted to the generator, so that the generator generates synthetic data according to the transmitted data domain association result parameter. The synthetic data is input into the discriminator again, so that the discriminator generates a data domain association result parameter again. The above steps performed by the generator and the discriminator are repeated until the data processing iteration meets the iteration stopping condition. At this time, it is considered that the synthetic data generated by the generator in the last time is very similar to the business data of the real business in appearance, meaning, statistical characteristics, etc., and the synthetic data is not the business data of the real business, i.e., the synthetic data does not exist corresponding to the real business, and can be used as analysis basis data for business analysis.

[0030] In some embodiments, between step S102 and step S103, data requirement information of the business analysis can also be acquired, and based on the data requirement information, the initial data field correlation result parameter is adjusted to make the initial data field correlation result parameter tend to the data requirement represented by the data requirement information. The data requirement information can represent the data requirement of the business analysis. Different business analyses can have different data requirements. For example, a large-amount transaction business analysis can have a data requirement that is more biased towards data with larger transaction amounts, and a small-amount transaction business analysis can have a data requirement that is more biased towards data with smaller transaction amounts. Based on the data requirement information, the initial data field correlation result parameter can be adjusted to make the initial data field correlation result parameter tend to the data requirement represented by the data requirement information. When the initial data field correlation result parameter tends to the data requirement represented by the data requirement information, the data items corresponding to the data fields in the generated synthetic data are also closer to the data requirement represented by the data requirement information.

[0031] Through the adjustment of the initial data field correlation result parameter, in the case that the business data of the real business has bias and incompleteness, synthetic data with lower bias and lower incompleteness and closer to the data requirement represented by the data requirement information can also be obtained, and at least part of the bias and incompleteness of the source data itself can be eliminated.

[0032] In the embodiments of the present application, the generator and the discriminator can be used for data processing iteration according to the correlation relationship of the data fields in the business data of the real business, so that the generator can generate synthetic data according to the data field correlation result parameter output by the discriminator to represent the correlation relationship of the data fields of the data, the discriminator can output the data field correlation result parameter according to the synthetic data to update the generator, the generator can output new synthetic data, the process is repeated until the data processing iteration meets the iteration processing condition, and the synthetic data output by the generator is selected as the analysis base data for the business analysis. The synthetic data is obtained by the cooperation of the generator and the discriminator, and has similar appearance, meaning, statistical characteristics, etc. to the business data of the real business, but is not the business data of the real business, and there is no real business corresponding to the synthetic data. The synthetic data is used as the analysis base data for the business analysis, which can not only meet the requirements of the business analysis, but also avoid the leakage of private data and improve the data security.

[0033] Furthermore, the business data in the embodiments of this application is often discrete data, not continuous data like images in image processing. Therefore, it is not possible to use generative adversarial networks (GANs) that process continuous data for processing. If a GAN that processes continuous data is used to directly process discrete data, i.e., the discriminator takes the derivative of discrete data like business data, it will cause a serious pattern collapse problem. The discriminator will not be able to generate the correct feedback parameters, and the generator will not be able to generate data based on the correct feedback parameters. In other words, even if the generator generates data, the data will not reflect the appearance, meaning, statistical characteristics, etc. of the real business data, and will not have any value for business analysis.

[0034] In the data protection processing method of this application embodiment, instead of differentiating the business data, it samples the business data according to the data domain to obtain data items. Using these data items, it obtains data domain correlation result parameters that characterize the relationships between data domains. These data domain correlation result parameters are then provided as feedback parameters to the generator, enabling the generator to generate synthetic data based on these parameters. The generator and discriminator dynamically iterate through the above process, completing the dynamic iterative optimization of the generator. This ensures that the synthetic data generated by the generator can reflect the appearance, meaning, and statistical characteristics of real business data. Even in scenarios with discrete data, it can generate high-quality synthetic data that meets the needs of business analysis.

[0035] In some embodiments, during the first data processing iteration, the generator outputs synthesized data based on the initial data domain correlation result parameters. During the subsequent data processing iterations after the first data processing iteration, i.e., the i-th data processing iteration, the generator can output synthesized data based on the data domain correlation result parameters obtained in the (i-1)-th data processing iteration, where i is an integer greater than 1. Figure 2 A flowchart illustrating a data protection processing method provided in another embodiment of this application. Figure 2 and Figure 1 The difference is that, Figure 1 Step S103 can be further refined as follows: Figure 2 Steps S1031 to S1037 are included. Steps S1031 and S1032 are the steps of the first data processing iteration, and steps S1034 and S1035 are the steps of the i-th data processing iteration.

[0036] In step S1031, the generator is invoked to output synthesized data based on the initial data domain correlation result parameters.

[0037] Here, the initial data domain correlation result parameters are used as the feedback parameters required by the generator, so that the generator can output synthesized data based on the initial data domain correlation result parameters.

[0038] In step S1032, the discriminator outputs the data field correlation result parameter according to the input first mixed data.

[0039] The first mixed data includes the synthetic data and the business data of the plurality of real businesses. Here, the business data of the plurality of real businesses is obtained again, and the business data is mixed with the synthetic data and input into the discriminator. The business data of the real businesses in the first mixed data can be the same as or different from the business data of the real businesses in step S101, which is not limited here. Through the mixing of the data, the amount of data input into the discriminator is expanded, so that the data field correlation result parameter generated by the discriminator has more data support.

[0040] The discriminator can extract data items from the business data according to the data fields in the data categories from the first mixed data, randomly combine the data items to obtain a plurality of combined data. The number of combined data is greater than or equal to the sum of the number of synthetic data and the number of business data in the first mixed data. The initial data field correlation result parameter can be used to score the combined data. In the case where the score of the combined data does not meet the score cutoff condition, the discriminator continues to randomly combine data items, increases the combined data, that is, randomly combines to obtain more combined data, and scores the combined data again. Repeat the process until the combined data meets the score cutoff condition. The discriminator can generate the data field correlation result parameter corresponding to the data items in the increased combined data based on the increased combined data. The score obtained by scoring the combined data can represent the closeness of the combined data to the authenticity of the real data. In some cases, there may be a phenomenon that the score of part of the combined data is even higher than the score of the business data, which indicates that the closeness of the combined data to the authenticity of the real data is higher than the closeness of the business data to the authenticity of the real data.

[0041] The content of extracting data items from the first mixed data can refer to the related description of extracting data items from the business data in the above embodiments, which is not repeated here. In some examples, in the case where the data categories include continuous character types, the data belonging to the continuous character types in the first mixed data is split according to the data fields in the data subcategories to obtain data items. For specific content, refer to the related description of extracting data items from the data belonging to the continuous character types in the business data in the above embodiments, which is not repeated here.

[0042] In step S1033, it is judged whether the data processing iteration meets the iteration cutoff condition. If the data processing iteration does not meet the iteration cutoff condition, step S1034 is executed; if the data processing iteration meets the iteration cutoff condition, step S1037 is executed.

[0043] After the first data processing iteration, it is determined whether the data processing iteration meets the iteration termination condition. If the data processing iteration meets the iteration termination condition, the next data processing iteration is not needed, i.e., the data processing iteration is terminated. If the data processing iteration does not meet the iteration termination condition, the next data processing iteration, i.e., the second data processing iteration, is needed.

[0044] In the following, for the convenience of description of other data processing iterations after the first data processing iteration, any one of the data processing iterations after the first data processing iteration is selected for description, i.e., the ith data processing iteration is described.

[0045] In step S1034, the generator outputs the synthetic data according to the data domain correlation result parameter obtained in the (i-1)th data processing iteration.

[0046] The data domain correlation result parameter generated by the discriminator in the (i-1)th data processing iteration is used as the feedback parameter required by the generator in the ith data processing iteration, and the generator generates the synthetic data according to the data domain correlation result parameter generated in the (i-1)th data processing iteration.

[0047] In step S1035, the discriminator generates the data domain correlation result parameter according to the second mixed data.

[0048] The second mixed data includes the synthetic data obtained in the ith data processing iteration and the synthetic data obtained in the (i-1)th data processing iteration. That is, the second mixed data can be formed by mixing the synthetic data obtained in the ith data processing iteration and the synthetic data obtained in the (i-1)th data processing iteration. Through the mixing of data, the amount of data input to the discriminator is increased, and the influence of the synthetic data of the last iteration, i.e., the (i-1)th data processing iteration, is introduced, so that the data domain correlation result parameter generated by the discriminator has more data support, and the association relationship between the data domains represented by the generated data domain correlation result parameter is closer to the association relationship between the data domains of the real data according to the last iteration, i.e., the (i-1)th data processing iteration.

[0049] The discriminator can be called to extract data items from the second mixed data according to the data domains in the data category; the discriminator is called to randomly combine the extracted data items to obtain a plurality of combined data; the discriminator is called to score the combined data based on the initial data domain correlation result parameter; in the case that the combined data does not meet the scoring termination condition, the discriminator is called to randomly combine the extracted data items again to increase the combined data, and score the combined data, until the combined data meets the scoring termination condition, and the discriminator is called to generate the data domain correlation result parameter corresponding to the data items in the combined data based on the combined data.

[0050] The discriminator obtains the number of the combined data based on the second mixed data, which is greater than or equal to the sum of the number of the synthetic data obtained in the i th data processing iteration and the number of the synthetic data obtained in the i-1 th data processing iteration in the second mixed data. The combined data can be scored by using the initial data domain correlation result parameter. In a case where the score of the combined data does not satisfy the score cutoff condition, the discriminator continues to randomly combine data items, increases new combined data, that is, randomly combines more combined data, and scores the combined data again. The score obtained by scoring the combined data can represent the closeness of the combined data to the authenticity of the real data. In some cases, the score of part of the combined data can even be higher than the score of the business data, which indicates that the closeness of the combined data to the authenticity of the real data is higher than the closeness of the business data to the authenticity of the real data. The specific content of this part can also be referred to the related description of the discriminator generating the data domain correlation result parameter based on the input first mixed data in the above-mentioned embodiments, which will not be repeated here.

[0051] The content of the data item extracted from the second mixed data can refer to the related description of the extraction of the data item from the business data in the above-mentioned embodiments, which will not be repeated here. In some examples, in a case where the data category includes the continuous character type, the data belonging to the continuous character type in the second mixed data is split according to the data domain in the data subcategory to obtain the data item. The specific content can be referred to the related description of the extraction of the data item from the data belonging to the continuous character type in the business data in the above-mentioned embodiments, which will not be repeated here.

[0052] In some embodiments, after calling the discriminator to output the data domain correlation result parameter corresponding to the data item in the combined data based on the combined data, the data requirement information of the business analysis can also be obtained, and the data domain correlation result parameter is adjusted based on the data requirement information, so that the data domain correlation result parameter tends to the data requirement represented by the data requirement information. The content of the data requirement information and the adjustment of the data domain correlation result parameter can be referred to the related description of the data requirement information and the adjustment of the initial data domain correlation result parameter in the above-mentioned embodiments, which will not be repeated here.

[0053] The score cutoff condition in the above-mentioned embodiments can include: the proportion of the combined data whose score is greater than or equal to the target score in the combined data is higher than or equal to the preset proportion threshold; and / or, the number of the combined data reaches the target number.

[0054] The target score is a score threshold that meets the synthetic data requirement, which can be set according to scenarios, requirements, experience, etc., and is not limited herein. The preset proportion threshold is a proportion threshold that meets the synthetic data requirement, which can be set according to scenarios, requirements, experience, etc., and is not limited herein. The combined data with a score greater than or equal to the target score is data that meets the requirement in terms of the degree of similarity in appearance, meaning, statistical characteristics, etc. between the combined data and the business data of the real business. The proportion of the combined data with a score greater than or equal to the target score in the combined data is higher than or equal to the preset proportion threshold, indicating that the discriminator has generated a better data domain correlation result parameter that can be generated in this data processing iteration.

[0055] The target number is a number threshold that meets the synthetic data requirement, which in some cases can be the maximum number of combined data that can be generated. When the number of combined data reaches the target number, it indicates that the discriminator has generated a better data domain correlation result parameter that can be generated in this data processing iteration.

[0056] In step S1036, it is determined whether the data processing iteration meets the iteration stop condition. If the data processing iteration does not meet the iteration stop condition, return to step S1034; if the data processing iteration meets the iteration stop condition, execute step S1037.

[0057] After the i-th data processing iteration, it can be determined whether the data processing iteration meets the iteration stop condition. If the data processing iteration meets the iteration stop condition, the next data processing iteration is not needed, i.e., the data processing iteration ends. If the data processing iteration does not meet the iteration stop condition, the next data processing iteration, i.e., the (i+1)-th data processing iteration, needs to be performed, and steps S1034 and S1035 are executed.

[0058] In step S1037, the combined data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration is selected as the analysis base data for business analysis.

[0059] The data processing iteration meeting the iteration stop condition indicates that the data domain correlation result parameter generated in the last data processing iteration, i.e., the current data processing iteration, meets the expected requirement, and the combined data generated by the generator using the data domain correlation result also meets the expected requirement, which can be used as the business analysis base data for business analysis.

[0060] The iteration stop condition in the above embodiment is a discrimination condition for the data domain correlation result parameter generated by the discriminator to achieve the expected effect. The data processing iteration meeting the iteration stop condition indicates that the data domain correlation result parameter generated by the discriminator in the current data processing iteration can achieve the expected effect, i.e., the combined data generated using the data domain correlation result parameter that cannot be traced back to the source data can meet the business analysis requirement.

[0061] In some embodiments, the iteration stopping condition comprises: the number of data processing iterations reaching a preset iteration number; and / or, the quality evaluation score of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration being higher than or equal to a qualified evaluation score.

[0062] The preset iteration number can be set according to scenarios, requirements, experience, etc., which are not limited herein, for example, the preset iteration number is 10 times.

[0063] The quality of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration, i.e., the current data processing iteration, can be evaluated to obtain a quality evaluation score, and the quality evaluation score can represent the degree of adaptation of the synthetic data as analysis basis data. The higher the quality evaluation score, the higher the degree of adaptation of the synthetic data as analysis basis data. The qualified evaluation score is a score threshold for judging whether the degree of adaptation of the synthetic data as analysis basis data meets the requirements. The quality evaluation score higher than or equal to the qualified evaluation score indicates that the degree of adaptation of the synthetic data as analysis basis data meets the requirements. The quality evaluation score lower than the qualified evaluation score indicates that the degree of adaptation of the synthetic data as analysis basis data does not meet the requirements.

[0064] In some examples, the iteration stopping condition comprises the number of data processing iterations reaching a preset iteration number; correspondingly, if the number of data processing iterations reaches the preset iteration number, it can be considered that the data processing iteration meets the iteration stopping condition.

[0065] In some examples, the iteration stopping condition comprises the quality evaluation score of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration being higher than or equal to a qualified evaluation score; correspondingly, if the quality evaluation score of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration is higher than or equal to the qualified evaluation score, it can be considered that the data processing iteration meets the iteration stopping condition.

[0066] In some examples, the iteration stopping condition comprises: the number of data processing iterations reaching a preset iteration number, and the quality evaluation score of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration being higher than or equal to a qualified evaluation score; correspondingly, only in the case that the number of data processing iterations reaches the preset iteration number, and the quality evaluation score of the synthetic data generated by the generator configured with the discriminator for the data domain relevance result parameter obtained in the last data processing iteration is higher than or equal to the qualified evaluation score, it can be considered that the data processing iteration meets the iteration stopping condition.

[0067] The quality evaluation score can be obtained based on one or more of the statistical index parameter, the likelihood index parameter, the detection index parameter, the learning performance index parameter, and the privacy index parameter of the synthetic data. Specifically, the quality evaluation score of the synthetic data can be obtained by using a weight algorithm according to the index parameters of the synthetic data and the corresponding weights of the index parameters.

[0068] The statistical index parameter represents the similarity of the statistical characteristics of the synthetic data and the business data. The statistical index parameter can be obtained by comparing the difference between the statistical characteristics of the synthetic data and the business data of the real business.

[0069] The likelihood index parameter represents the likelihood value of the synthetic data and the business data. The likelihood index parameter can include the likelihood value and / or the logarithmic likelihood value. The likelihood index parameter can be obtained by fitting a probability model of the business data of the real business, calculating the likelihood value of the synthetic data, and obtaining the likelihood index parameter based on the likelihood value.

[0070] The detection index parameter represents the possibility of the synthetic data being classified into the same category as the business data by regression classification. The detection index parameter can be obtained by mixing the synthetic data and the business data of the real business, using a regression classification model to obtain a classification result, and obtaining the detection index parameter based on the classification result.

[0071] The learning performance index parameter represents the effect of the synthetic data as analysis basis data for business analysis. The learning performance index parameter is used to evaluate whether the synthetic data can be used to replace the real data for business analysis.

[0072] The privacy index parameter represents the possibility that the real data cannot be traced based on the synthetic data. The privacy index parameter is used to evaluate whether the synthetic data can inversely identify the related information of the real data and assess the possibility of leaking the information of the real data.

[0073] To facilitate understanding, the overall process of the data protection processing is described below with an example. Figure 3 An example of a flowchart of the data protection processing process provided by the embodiments of the present application is shown in FIG. 2. As shown in FIG. 2, the data protection processing process includes steps S201 to S217. Figure 3

[0074] In step S201, the real data is analyzed to obtain the data items in the real data and the data domains to which the data items belong.

[0075] In step S202, the data categories corresponding to the data domains are obtained, and the data domains of the non-continuous character type and the data domains of the continuous character type are determined.

[0076] In step S203, the correlation between the data domains of the non-continuous character type is analyzed.

[0077] ​In step S204, it is judged whether the data corresponding to the data field of the continuous character type is a key field. The key field is data having a beneficial effect on service analysis. If the data corresponding to the data field of the continuous character type is a key field, step S205 is executed; if the data corresponding to the data field of the continuous character type is a key field, step S207 is executed.

[0078] In step S205, the data corresponding to the data field of the continuous character type is split to obtain a data field of a data subcategory.

[0079] In step S206, the association between the data fields of the data subcategory and the association between the data field of the data subcategory and the data field of the non-continuous character type are analyzed.

[0080] In step S207, a data field association result parameter is obtained based on the association between the data fields.

[0081] In step S208, the data field association result parameter is adjusted in combination with data requirement information.

[0082] In step S209, a synthesizer is called to generate synthetic data based on the data field association result parameter.

[0083] In step S210, it is judged whether the data processing iteration reaches a preset iteration number. If the preset iteration number is not reached, step S211 is executed; if the preset iteration number is reached, step S217 is executed.

[0084] In step S211, the synthetic data is mixed with other data and input to a discriminator.

[0085] In step S212, the data item of the mixed data is obtained by calling the discriminator.

[0086] In step S213, the data item is randomly combined by calling the discriminator to obtain combined data.

[0087] In step S214, the combined data is scored according to the data field association result parameter by calling the discriminator.

[0088] In step S215, it is judged whether the combined data meets a scoring cutoff condition. If the scoring cutoff condition is not met, step S213 is returned; if the scoring cutoff condition is met, step S216 is executed.

[0089] In step S216, the data field association result parameter corresponding to the combined data is generated, and step S208 is executed.

[0090] In step S217, the synthetic data is quality evaluated, and it is determined whether the quality evaluation passes. If yes, the synthetic data is taken as analysis base data, and the process ends; if no, the process returns to step S211.

[0091] The specific content in steps S201 to S217 can be referred to the related description in the above embodiment, and will not be repeated here.

[0092] It should be noted that the acquisition, storage, use, processing, etc. of information and data in the embodiments of the present application are authorized by the user or relevant institution and comply with relevant provisions of national laws and regulations.

[0093] The second aspect of the present application provides a data protection processing device. Figure 4 As shown in the structural schematic diagram of the data protection processing device provided in an embodiment of the present application, the data protection processing device 300 can include an extraction module 301, a calculation module 302, and a training calling module 303. Figure 4 The calculation module 302 can be configured to obtain an initial data domain correlation result parameter based on the data item corresponding to each real business, and the initial data domain correlation result parameter is used to represent the correlation between the data domains in the business data.

[0094] The extraction module 301 can be configured to obtain business data of a plurality of real businesses, and extract data items from the business data according to data domains in a data category.

[0095] In some examples, the data category includes one or more of the following: Boolean type, numerical type, classification type, and continuous character type.

[0096] The calculation module 302 can be configured to obtain an initial data domain correlation result parameter based on the data item corresponding to each real business, and the initial data domain correlation result parameter is used to represent the correlation between the data domains in the business data.

[0097] The training calling module 303 can be configured to call the generator and the discriminator to perform data processing iterations based on the initial data domain correlation result parameter until the data processing iterations meet an iteration stop condition, and select synthetic data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration as analysis base data for business analysis, and the data domain correlation result parameter is used to represent the relationship between the data domains in the input data of the discriminator.

[0098] In some examples, the iteration stop condition includes: the number of data processing iterations reaches a preset iteration number; and / or, the quality evaluation score of the synthetic data generated by the generator configured with the data domain correlation result parameter obtained by the discriminator in the last data processing iteration is higher than or equal to a qualified evaluation score.

[0099] The quality evaluation score is obtained based on one or more of a statistical index parameter, a likelihood index parameter, a detection index parameter, a learning performance index parameter, and a privacy index parameter of the synthetic data.

[0100] The statistical index parameter represents the similarity of statistical characteristics between the synthetic data and the business data. The likelihood index parameter represents the likelihood value of the synthetic data and the business data. The detection index parameter represents the possibility of the synthetic data being classified into the same category as the business data through regression classification. The learning performance index parameter represents the effect of the synthetic data as analysis base data for business analysis. The privacy index parameter represents the possibility that the real data cannot be traced based on the synthetic data.

[0101] In the embodiments of the present application, the generator and the discriminator are used for data processing iteration according to the correlation between the data domains of the business data of the real business, so that the generator can generate synthetic data according to the data domain correlation result parameter representing the correlation between the data domains of the data output by the discriminator, the discriminator outputs the data domain correlation result parameter according to the synthetic data to update the generator, the generator outputs new synthetic data, and the process is repeated until the data processing iteration meets the iteration processing condition, and the synthetic data output by the generator is selected as the analysis base data for business analysis. The synthetic data is obtained by the generator and the discriminator, and has similar appearance, meaning, statistical characteristics, etc. to the business data of the real business, but is not the business data of the real business, and there is no real business corresponding to the synthetic data. The synthetic data is used as the analysis base data for business analysis, which can meet the demand of business analysis and avoid the leakage of privacy data, and improve the data security.

[0102] Moreover, the business data in the embodiments of the present application is often discrete data, not continuous data such as images in image processing, so it cannot be processed by the generative adversarial network for processing continuous data. If the generative adversarial network for processing continuous data is directly used to process discrete data, that is, the discriminator is used to derive the discrete data such as business data, it will cause a serious mode collapse problem, the discriminator cannot generate correct feedback parameters, and the generator cannot generate data according to the correct feedback parameters, that is, the generator generates data, but the data cannot represent the appearance, meaning, statistical characteristics, etc. of the real business data, and has no use value for business analysis.

[0103] The data protection processing method in the embodiments of the present application is not to derive the business data, but to sample data items from the business data according to the data domains, to obtain data domain correlation result parameters capable of representing the correlation between the data domains by using the data items, to take the data domain correlation result parameters as feedback parameters provided to the generator, and to make the generator generate synthetic data according to the data domain correlation result parameters. The generator and the discriminator dynamically cycle the above process to complete dynamic cycle optimization of the generator, so that the synthetic data generated by the generator can reflect the appearance, meaning, statistical characteristics and the like of the real business data, and in the scene of discrete data, the generator can also generate high-quality synthetic data meeting the business analysis requirements.

[0104] In some embodiments, the data processing iteration is the first data processing iteration, and the training calling module 303 can be configured to: call the generator to output synthetic data according to the initial data domain correlation result parameter; and call the discriminator to generate the data domain correlation result parameter according to the input first mixed data, the first mixed data including the synthetic data and the business data of the plurality of real businesses.

[0105] In some embodiments, the data processing iteration is the i-th data processing iteration, and the training calling module 303 can be configured to: call the generator to output synthetic data according to the data domain correlation result parameter obtained in the (i-1)-th data processing iteration; and call the discriminator to generate the data domain correlation result parameter according to the second mixed data, the second mixed data including the synthetic data obtained in the i-th data processing iteration and the synthetic data obtained in the (i-1)-th data processing iteration, wherein i is an integer greater than 1.

[0106] In some examples, the training calling module 303 can be configured to: call the discriminator to extract data items from the second mixed data according to the data domains in the data categories; call the discriminator to randomly combine the extracted data items to obtain a plurality of combined data; call the discriminator to score the combined data based on the initial data domain correlation result parameter; in the case that the combined data does not meet the scoring cutoff condition, call the discriminator to randomly combine the extracted data items again to increase the combined data, and score the combined data until the combined data meets the scoring cutoff condition, and call the discriminator to generate the data domain correlation result parameter corresponding to the data items in the combined data based on the combined data.

[0107] In some examples, the scoring cutoff condition includes: the proportion of the combined data with a score greater than or equal to the target score in the combined data is higher than or equal to a preset proportion threshold; and / or, the number of the combined data reaches a target number.

[0108] In some embodiments, the extraction module 301 can also be configured to, in the case that the data category comprises continuous character type, split the data belonging to the continuous character type in the business data or the second mixed data according to the data field in the data subcategory, to obtain the data item.

[0109] The data subcategory comprises at least one data field. The data subcategory comprises one or more of the following: Boolean type, numerical type, categorical type.

[0110] Figure 5 A structural schematic diagram of a data protection processing apparatus provided by another embodiment of the present application. Figure 5 Different from Figure 4 , the data protection processing apparatus 300 shown in FIG. 4 further comprises a parameter adjustment module 304. Figure 5 The parameter adjustment module 304 can be configured to: obtain data requirement information of the business analysis, and adjust the initial data field correlation result parameter based on the data requirement information, so that the initial data field correlation result parameter tends to the data requirement represented by the data requirement information.

[0111] The parameter adjustment module 304 can be configured to: obtain data requirement information of the business analysis, and adjust the initial data field correlation result parameter based on the data requirement information, so that the initial data field correlation result parameter tends to the data requirement represented by the data requirement information.

[0112] The parameter adjustment module 304 can be configured to: obtain data requirement information of the business analysis, and adjust the initial data field correlation result parameter based on the data requirement information, so that the initial data field correlation result parameter tends to the data requirement represented by the data requirement information.

[0113] The third aspect of the present application provides an electronic device. Figure 6 A structural schematic diagram of an electronic device provided by an embodiment of the present application. As shown in Figure 6 The electronic device 400 comprises a memory 401, a processor 402, and a computer program stored in the memory 401 and executable on the processor 402.

[0114] In some examples, the processor 402 described above can comprise a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement one or more embodiments of the present application.

[0115] The memory 401 can include a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Therefore, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions that, when executed (e.g., by one or more processors), are operable to perform the operations described with reference to the data protection processing method according to the embodiments of the present application.

[0116] The processor 402 runs a computer program corresponding to the executable program code stored in the memory 401 by reading the executable program code, for implementing the data protection processing method in the above-described embodiments.

[0117] In some examples, the electronic device 400 can further include a communication interface 403 and a bus 404. As shown, the memory 401, the processor 402, and the communication interface 403 are connected through the bus 404 and complete communication among each other. Figure 6

[0118] The communication interface 403 is mainly used to implement the communication among the modules, devices, units, and / or equipment in the embodiments of the present application. The input device and / or the output device can also be accessed through the communication interface 403.

[0119] ​Bus 404 includes a hardware, software, or both that couples components of electronic device 400 to each other. As an example and not by way of limitation, bus 404 can include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand™ interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-E) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or another suitable bus or a combination of two or more of these. Where appropriate, bus 404 can include one or more buses. Although this application describes and shows a particular bus, this application contemplates any suitable bus or interconnect.

[0120] The fourth aspect of the application further provides a computer readable storage medium, and computer program instructions are stored on the computer readable storage medium. The computer program instructions are executed by a processor to implement the data protection processing method in the above embodiments, and the same technical effects can be achieved. To avoid repetition, details are not described here. The computer readable storage medium can include a non-transitory computer readable storage medium, such as a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, which is not limited here.

[0121] The embodiments of the application provide a computer program product, and instructions in the computer program product are executed by a processor of an electronic device to enable the electronic device to execute the data protection processing method in the above embodiments, and the same technical effects can be achieved. To avoid repetition, details are not described here.

[0122] It should be clarified that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. For the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments, the relevant parts can be referred to the description section of the method embodiments. This application is not limited to the specific steps and structures described above and shown in the figures. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application. Furthermore, for the sake of brevity, detailed descriptions of known methods and techniques are omitted here.

[0123] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0124] Those skilled in the art will understand that the above embodiments are exemplary and not restrictive. Different technical features appearing in different embodiments can be combined to achieve beneficial effects. Based on a study of the drawings, specification, and claims, those skilled in the art should be able to understand and implement other variations of the disclosed embodiments. In the claims, the term "comprising" does not exclude other means or steps; the quantifier "a" does not exclude a plurality; the terms "first" and "second" are used to identify names and not to indicate any particular order. No reference numerals in the claims should be construed as limiting the scope of protection. The functionality of multiple parts appearing in the claims can be implemented by a single hardware or software module. The appearance of certain technical features in different dependent claims does not mean that these technical features cannot be combined to achieve beneficial effects.

Claims

1. A data protection processing method, characterized in that, include: Acquire business data from multiple real business transactions, and extract data items from the business data according to the data fields in the data categories. The data fields are the items under the data categories, and the data items are the smallest meaningful data segments in the business data. Based on the data items corresponding to each real business transaction, an initial data domain correlation result parameter is obtained. The initial data domain correlation result parameter is used to characterize the correlation relationship between data domains in the business data. The generator and discriminator are invoked to perform data processing iterations based on the initial data domain correlation result parameters until the data processing iteration meets the iteration cutoff condition. The synthetic data generated by the generator, which is configured with the data domain correlation result parameters obtained by the discriminator in the most recent data processing iteration, is selected as the basic data for business analysis. The data domain correlation result parameters are used to characterize the relationship between data domains in the data input to the discriminator. Each data processing iteration includes: the generator generating the synthetic data based on the data domain correlation result parameters, using the synthetic data as input data for the discriminator, and the discriminator generating the data domain correlation result parameters based on the input synthetic data.

2. The method according to claim 1, characterized in that, The first data processing iteration includes: The generator is invoked to output the synthesized data based on the initial data domain correlation result parameters; The discriminator is invoked to generate the data domain correlation result parameters based on the input first mixed data. The first mixed data includes the synthetic data and business data from multiple real business transactions.

3. The method according to claim 1, characterized in that, The i-th data processing iteration includes: The generator is invoked to output the synthesized data based on the data domain correlation result parameters obtained in the (i-1)th data processing iteration; The discriminator is invoked to generate the data domain correlation result parameters based on the second mixed data. The second mixed data includes the synthetic data obtained from the i-th data processing iteration and the synthetic data obtained from the (i-1)-th data processing iteration, where i is an integer greater than 1.

4. The method according to claim 3, characterized in that, The call discriminator generates the data domain correlation result parameters based on the second mixed data, including: The discriminator is invoked to extract data items from the second mixed data according to the data fields in the data category; The discriminator is invoked to randomly combine the extracted data items to obtain multiple combined data; The discriminator is invoked to score the combined data based on the initial data domain correlation result parameters; If the combined data does not meet the scoring cutoff condition, the discriminator is invoked to randomly combine the extracted data items again, adding the combined data and scoring the combined data until the combined data meets the scoring cutoff condition. Then, the discriminator is invoked to generate the data domain correlation result parameters corresponding to the data items in the combined data based on the combined data.

5. The method according to claim 1 or 4, characterized in that, The data categories include one or more of the following: Boolean type, numeric type, categorical type, and continuous character type.

6. The method according to claim 3, characterized in that, Also includes: When the data category includes a continuous character type, the data belonging to the continuous character type in the business data or the second mixed data is split according to the data fields in the data subcategory to obtain data items. The data subcategory includes at least one data field. The data subcategories include one or more of the following: Boolean type, numeric type, and category type.

7. The method according to claim 4, characterized in that, Before the call generator and discriminator perform data processing iterations based on the initial data domain correlation result parameters, the method further includes: Obtain data requirement information for business analysis, and based on the data requirement information, adjust the initial data domain correlation result parameters so that the initial data domain correlation result parameters tend to the data requirements represented by the data requirement information; After the discriminator is invoked to output the data domain correlation result parameter corresponding to the data item in the combined data based on the combined data, the method further includes: Obtain the data requirement information for business analysis, and based on the data requirement information, adjust the data domain correlation result parameters so that the data domain correlation result parameters tend to the data requirements represented by the data requirement information.

8. The method according to claim 1, characterized in that, The iteration cutoff condition includes: The number of data processing iterations has reached the preset number of iterations; And / or, The quality assessment score of the synthetic data generated by the generator, which is configured with the data domain correlation result parameter obtained by the discriminator in the most recent data processing iteration, is higher than or equal to the qualified assessment score. The quality assessment score is obtained based on one or more of the statistical index parameters, likelihood index parameters, detection index parameters, learning efficiency index parameters, and privacy index parameters of the synthetic data. The statistical indicator parameter characterizes the similarity of the statistical features of the synthetic data and the business data; the likelihood indicator parameter characterizes the likelihood value of the synthetic data and the business data; the detection indicator parameter characterizes the probability that the synthetic data belongs to the same class as the business data through regression classification; the learning effectiveness indicator parameter characterizes the effectiveness of the synthetic data as the basis for analysis in business analysis; and the privacy indicator parameter characterizes the possibility that the real data cannot be traced back based on the synthetic data.

9. The method according to claim 4, characterized in that, The scoring cutoff criteria include: The proportion of the combined data with scores greater than or equal to the target score is higher than or equal to a preset proportion threshold. And / or, The number of combined data has reached the target number.

10. A data protection processing device, characterized in that, include: The extraction module is used to acquire business data from multiple real business transactions, and extract data items from the business data according to the data fields in the data categories. The data fields are the items under the data categories, and the data items are the smallest meaningful data segments in the business data. The calculation module is used to obtain initial data domain correlation result parameters based on the data items corresponding to each real business transaction. The initial data domain correlation result parameters are used to characterize the correlation relationship between data domains in the business data. The training and calling module is used to call the generator and discriminator to perform data processing iterations based on the initial data domain correlation result parameters until the data processing iteration meets the iteration cutoff condition. The synthetic data generated by the generator, which is configured with the data domain correlation result parameters obtained by the discriminator in the most recent data processing iteration, is selected as the basic data for business analysis. The data domain correlation result parameters are used to characterize the relationship between data domains in the data input to the discriminator. Each data processing iteration includes: the generator generating the synthetic data based on the data domain correlation result parameters, using the synthetic data as input data for the discriminator, and the discriminator generating the data domain correlation result parameters based on the input synthetic data.

11. An electronic device, characterized in that, include: Processor and memory storing computer program instructions; When the processor executes the computer program instructions, it implements the data protection processing method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the data protection processing method as described in any one of claims 1 to 9.

Images