User attribute data volume adjustment method, electronic equipment and storage medium

Abstract

The invention relates to the technical field of data processing, in particular to a user attribute data volume adjusting method, electronic equipment and a storage medium, and the method comprises the following steps: processing data in each received user attribute data table according to a preset compliance processing rule, and packaging the processed data into a user attribute data packet, according to the data packet size of the user attribute data packet, calculating to obtain a data priority corresponding to the user attribute data packet, when the data priority is greater than a preset data priority threshold, deleting a target user in the user attribute data packet and user attribute data corresponding to the target user, otherwise, deleting the target user in the user attribute data packet. Adding a plurality of historical users and a plurality of user attributes corresponding to each historical user into a user attribute data packet; according to the method and the device, the user information can be mixed by adjusting the user attribute data volume, so that the user with changed information is not easy to identify, and the information leakage of the specific user is further prevented.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method for adjusting the amount of user attribute data, an electronic device, and a storage medium. Background Art

[0002] At present, several third-party platforms, such as various app platforms, will periodically or irregularly upload a large amount of current user attribute information and app usage information to other data analysis platforms, or use it to generate various analysis reports, or user advertising, etc. In this process, in order to prevent the leakage of user information, the user attribute data will be encrypted in advance. However, each time the user information is uploaded, some user information will change, such as new registered users and deregistered users. If the specific changed user data is cracked, it will cause targeted leakage of user data. Therefore, it is very important to know how to obfuscate the information of users with changes. Summary of the Invention

[0003] In response to the above technical problems, the present invention provides a method for adjusting the amount of user attribute data, an electronic device and a storage medium, which can obfuscate user information by adjusting the amount of user attribute data, making it difficult to identify users whose information has changed, thereby preventing the leakage of specific user information.

[0004] According to a first aspect of the present invention, there is provided a method for adjusting the amount of user attribute data, comprising the following steps:

[0005] S100, receiving a user attribute data set sent by a third-party platform, processing the data in each user attribute data table in the user attribute data set according to preset compliance processing rules, and packaging the processed user attribute data tables into a user attribute data packet; the user attribute data table includes several target users and several encrypted user attributes corresponding to each target user.

[0006] S200 , calculating a data priority D corresponding to the user attribute data packet according to the data packet size of the user attribute data packet.

[0007] The data priority Y corresponding to the user attribute data packet meets the following conditions:

[0008] Y=δ+ψ; wherein δ is the historical data packet loss rate corresponding to the data packet size of the user attribute data packet, and ψ is the normalized value of the historical transmission duration corresponding to the data packet size of the user attribute data packet.

[0009] S300 , when the data priority Y corresponding to the user attribute data packet is greater than Y0, r1 target users and a number of user attributes corresponding to each deleted target user are deleted from the user attribute data packet, where Y0 is a preset data priority threshold.

[0010] S400 : When the data priority Y corresponding to the user attribute data packet is Y0, r2 historical users and a number of user attributes corresponding to each historical user are added to the user attribute data packet.

[0011] According to the second aspect of the present invention, a non-transitory computer-readable storage medium is provided, in which at least one instruction or at least one program is stored. The at least one instruction or the at least one program is loaded and executed by a processor to implement the above-mentioned user attribute data amount adjustment method.

[0012] According to a third aspect of the present invention, there is provided an electronic device comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0013] The present invention has at least the following beneficial effects:

[0014] The present invention provides a method for adjusting the amount of user attribute data. First, data in each received user attribute data table is encrypted in compliance according to a preset compliance processing rule, and is packaged into a user attribute data packet after processing. According to the data packet size of the user attribute data packet, the data priority corresponding to the user attribute data packet is calculated, wherein the data priority is obtained according to two indicators: the historical data packet loss rate and the normalized value of the historical transmission time corresponding to the data packet size. Reliable data priority can be obtained, which is conducive to reliable processing of the data packet to improve the accuracy of subsequent data analysis; when the data priority is greater than a preset data priority threshold, the target user and the user attribute data corresponding to the target user in the user attribute data packet are deleted, which can reduce the data error rate. The deletion of data can make the changed user difficult to be identified, thereby preventing user information leakage. Conversely, several historical users and several user attributes corresponding to each historical user are added to the user attribute data packet, which can ensure sufficient data volume and perform obfuscation processing on the user data in the data table, making the user with changed information difficult to be identified, thereby preventing information leakage of specific users. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 This is a flow chart of a method for adjusting the amount of user attribute data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0018] The embodiment of the present invention provides a method for adjusting the amount of user attribute data, such as Figure 1 As shown, the method includes the following steps:

[0019] S100: Receive a user attribute data set sent by a third-party platform, process the data in each user attribute data table in the user attribute data set according to preset compliance processing rules, and package the processed user attribute data tables into a user attribute data packet; the user attribute data table includes a plurality of target users and a plurality of encrypted user attributes corresponding to each target user. For example, the user attributes may be encrypted names, unique identification numbers, SIM card numbers, MAC addresses, ages, and other attributes.

[0020] In a specific embodiment, step S100 includes the following steps:

[0021] S101, extract the i-th user attribute data table A i Several field names in A i Corresponding target field name list B i ={B i1 , B i2 ,……,B ij ,……,B in}, where i = 1, 2, ..., m, where m is the number of user attribute data tables in the user attribute data set, B ij Refers to A i The jth target field name in A, j = 1, 2, ..., n, n is i The number of target field names in ; the target field name is any user attribute.

[0022] S102, according to B ij The corresponding preset encryption judgment rules, from B ij Determine the encrypted target field value from the corresponding initial field values ​​and calculate B ij The corresponding target encryption level F ij .

[0023] Specifically, B ij The corresponding target encryption level F ij Meet the following conditions:

[0024] F ij =H 0 ij / H ij , where H 0 ij B ij The number of corresponding target field values, H ij B ij The total number of corresponding initial field values.

[0025] As mentioned above, by calculating the ratio of the number of encrypted initial field values ​​under each target field name to all initial field values ​​under the target field name, the target encryption level of the target field name can be obtained, thereby accurately understanding the data encryption situation, which is conducive to adopting different processing strategies according to the level of encryption, so as to effectively reduce data leakage in the platform.

[0026] Furthermore, the preset encryption judgment rule is set based on a given decision tree; wherein, the root node and each internal node of the given decision tree correspond to preset attribute judgment conditions, and the leaf node of the given decision tree is the encryption judgment result, and the encryption judgment result includes the encryption status and the corresponding node encryption score; it can be understood that: the encryption status includes not encrypted and encrypted, wherein the encrypted leaf node corresponds to the preset node encryption score.

[0027] For ease of understanding, let's take an example: For a SIM card number, the fourth to seventh characters are usually encrypted during encryption and represented by asterisks. Based on the encryption judgment rules pre-set in a given decision tree, the root node first determines whether it contains 11 characters. If so, the result is that encryption is not performed. If not, the first three characters are determined to be numeric values. If not, it is considered that the encryption does not comply with the pre-set encryption rules and is marked as not encrypted. If so, the next internal node of the root node again determines whether the fourth to seventh characters contain asterisks. If not, it is marked as not encrypted. If there are asterisks, the next internal node determines how many asterisks are contained. This is repeated layer by layer. If the number of asterisks is small, it indicates weak encryption, and the corresponding node has a low encryption score. If the number is greater than the threshold, it indicates strong encryption, and the corresponding node has a high encryption score. The corresponding leaf node outputs are all encrypted, obtaining the final encryption judgment result corresponding to the input data. It should be noted that the pre-set encryption judgment rules are set according to the encryption rules corresponding to different target field names. For example, for some target field names, multiple encryption rules with different encryption strengths are used. When constructing encryption judgment rules, it can be determined which encryption rule it belongs to. According to the different encryption strengths of different encryption rules, corresponding node encryption scores are set at the corresponding leaf nodes, and the node encryption scores are obtained according to the encryption strength of the determined encryption rules.

[0028] In a specific embodiment, S102 further includes the following steps:

[0029] S1021, obtain B from several historical user attribute data tables ij Several historical field values ​​under the corresponding target field name, and input each historical field value into a given decision tree to obtain the number of historical field values ​​corresponding to each preset leaf node in the given decision tree.

[0030] S1022: Calculate the encryption cracking probability corresponding to each preset leaf node in a given decision tree based on the number of historical field values ​​corresponding to each preset leaf node.

[0031] Among them, the encryption cracking probability corresponding to the preset leaf node of any layer meets the following conditions:

[0032] P d =(∑ d c＝1 K c ) / K, where P d The encryption cracking probability corresponding to the preset leaf node of the dth layer, K c is the number of historical field values ​​corresponding to the preset leaf nodes of the cth layer, and K is the sum of the number of historical field values ​​corresponding to the preset leaf nodes of each layer.

[0033] S1023, normalize the encryption cracking probability corresponding to each layer of preset leaf nodes to between 0 and 1, and obtain the node encryption score corresponding to each layer of preset leaf nodes; it can be understood as: using the normalized result corresponding to each layer of preset leaf nodes as the node encryption score corresponding to the preset leaf nodes themselves.

[0034] S1024, for B ij Corresponding to any initial field value, the initial field value is input into a given decision tree for judgment to obtain the target leaf node corresponding to the initial field value.

[0035] S1025, obtain the node encryption score corresponding to the target leaf node, and when the node encryption score is greater than the preset score threshold, determine that the initial field value is the encrypted target field value; in addition, when setting the preset score threshold, it is generally related to the number of layers corresponding to the leaf node. The more layers the judgment is performed, the higher the degree of encryption and the larger the preset score threshold. In specific implementation, the score threshold can also be manually set according to the specific judgment conditions.

[0036] As mentioned above, since the given decision tree contains multiple layers of judgment conditions, when it reaches the internal node but not the leaf node, it is still not determined whether the final result is encrypted. However, when it reaches the internal node, it means that a partial judgment has been made, and it is believed that the data is partially encrypted or the encryption format does not comply with the preset encryption rules. Therefore, the leaf nodes of this layer also have a certain encryption score. The node encryption score is used to determine whether it is in an encrypted state. This can accurately monitor encrypted data and unencrypted data, thereby achieving effective data screening.

[0037] S103, when B ij The corresponding target encryption level F ij ≥F 0 ij When B ij Get all the initial field values ​​except the target field value from the corresponding initial field values ​​and process them to get B ij Corresponding final field values ​​to obtain compliant data, where F 0 ij The default encryption level threshold.

[0038] In a specific embodiment, B is obtained by the following steps: ij The corresponding final field values ​​are:

[0039] S1031, targeting B ijFor any initial field value other than the target field value among the corresponding initial field values, the initial field value is identified to determine whether the initial field value meets the preset format requirements; it can be understood that: meeting the preset format requirements means that the content of the initial field value can be identified and the content of the initial field value meets the preset requirements. For example, the input format requirement for age is 30, and the number counted in the user data table is thirty. This can be identified and the corresponding preset encryption rule can be called to perform encryption again; or for another target field name, it can be identified that the encryption rule used for encryption has a low encryption strength, resulting in the output being unencrypted. After identifying the field value, an encryption rule with a higher encryption strength is used for encryption.

[0040] S1032, when the initial field value meets the preset format requirements, based on the target field corresponding to the initial field value, call the preset encryption rule corresponding to the target field and encrypt the initial field value to obtain the final field value corresponding to the initial field value.

[0041] S1033, when the initial field value does not meet the preset format requirements, the record corresponding to the initial field value is deleted to obtain B ij The corresponding final field values.

[0042] As described above, the initial field values ​​in the unencrypted state in the judgment results are identified, and the identified initial field values ​​are encrypted again. The initial field values ​​that cannot be identified may be data loss or data errors, etc., and they are deleted to achieve effective screening of the data and avoid affecting subsequent data analysis.

[0043] Further, F is obtained by the following steps 0 ij :

[0044] S10, obtain several historical user attribute data tables, and fit a relationship curve between the number of users and data accuracy based on the historical number of users and data accuracy corresponding to each historical user attribute data table; it can be understood that: data accuracy refers to the ratio of the number of correct initial field values ​​in the user attribute data table to the number of all initial field values.

[0045] S20, based on the relationship curve between the number of users and data accuracy, according to A i The number of users in A i The corresponding data accuracy η.

[0046] S30, obtain B in each historical user attribute data table ij The encryption probability λ of the corresponding target field name is calculated, and the average encryption probability is obtained; it can be understood as: the target encryption degree corresponding to the target field name is used as the encryption probability corresponding to the target field name.

[0047] S40, according to A i The corresponding data accuracy η, average encryption probability λ and the preset weights corresponding to η and λ are calculated by weighted sum to obtain F 0 ij .

[0048] As mentioned above, since the encryption level is not only affected by data collection, but also by errors in the data transmission process, the encryption level threshold is calculated based on the data accuracy and average encryption probability corresponding to the historical attribute data table, so that the encryption level threshold is more reasonable.

[0049] S104, when B ij The corresponding target encryption level F ij <F 0 ij When the data is not in compliance, a warning message will be sent to the third-party platform so that the third-party platform can i After verifying the data in the file, upload the data again until A is obtained. i Corresponding compliance data.

[0050] S105: Packing the obtained user attribute data tables with compliant data into a user attribute data packet.

[0051] As mentioned above, when the target encryption level is less than the preset encryption level threshold, it indicates that there are many field values ​​in the user attribute data table that are in an unencrypted state, which may also contain a lot of erroneous data, which may lead to data leakage or insufficient data volume in the subsequent data analysis process. Therefore, it is necessary to provide feedback to the third-party platform that sends the user attribute data table so that it can verify and correct the data to obtain compliant data that meets the requirements.

[0052] S200 , calculating a data priority D corresponding to the user attribute data packet according to the data packet size of the user attribute data packet.

[0053] The data priority Y corresponding to the user attribute data packet meets the following conditions:

[0054] Y=δ+ψ; wherein δ is the historical data packet loss rate corresponding to the data packet size of the user attribute data packet, and ψ is the normalized value of the historical transmission duration corresponding to the data packet size of the user attribute data packet.

[0055] In a specific embodiment, δ is obtained by the following steps:

[0056] K201, obtain several historical user attribute data packets, and obtain the historical data packet loss rate corresponding to each historical user attribute data packet.

[0057] K202, the data packet sizes of several historical user attribute data packets are used as independent variables, and the packet loss rate of each corresponding historical data is used as the dependent variable, and are substituted into the pre-set initial prediction model z = αx + β, and the target prediction model is obtained after linear fitting using the least squares method.

[0058] K203, substituting the data packet size of the user attribute data packet as a dependent variable into the target prediction model to obtain the historical data packet loss rate δ corresponding to the data packet size of the user attribute data packet.

[0059] In addition, when obtaining ψ, we first obtain the historical transmission duration corresponding to each historical user attribute data packet and normalize them to between 0 and 1. The packet sizes of several historical user attribute data packets are used as independent variables, and the corresponding normalized historical transmission durations are used as dependent variables. The subsequent processing is consistent with the process of obtaining δ and will not be repeated here.

[0060] As mentioned above, considering that the larger the packet loss rate and sending time are, the larger the data packet is, the packet loss rate and sending time are introduced when calculating the data priority, which can accurately characterize the data packet size and obtain a reliable data priority of the data packet, which can also be understood as a data packet score, and is conducive to reliable processing of the data packet according to the data packet score, so as to improve the accuracy of subsequent data analysis.

[0061] S300, when the data priority Y corresponding to the user attribute data packet is greater than Y0, r1 target users and several user attributes corresponding to each deleted target user are deleted from the user attribute data packet, where Y0 is a preset data priority threshold; those skilled in the art can set r1 and r2 according to actual needs; in addition, deleting r1 target users and several user attributes corresponding to each deleted target user can be understood as: when Y>Y0, r1 records are deleted from the user attribute data packet, where a row in the data table represents one record.

[0062] Furthermore, r1 can be obtained by the following steps:

[0063] S301, Y0 is used as the new data priority Y and substituted into Y = δ + ψ, and a new data packet size is obtained based on the obtained new δ and ψ and used as the data packet reference size; it can be understood as: δ and ψ are respectively converted into the formulas of the corresponding target prediction model, at this time the independent variable in Y = δ + ψ is changed to the data packet size, and the obtained data packet size is used as the data packet reference size.

[0064] S302: When the data priority Y corresponding to the user attribute data packet is greater than Y0, the difference between the data packet size of the original user attribute data packet and the data packet reference size is used as a reference amount of data to be deleted.

[0065] S303, based on the reference amount of data to be deleted, calculate the number of records to be deleted from the user attribute data packet, and use the calculation result as r1; each record includes a target user and several user attributes corresponding to the target user; it can be understood that the total data volume of the deleted r1 records is closest to the reference amount of data to be deleted.

[0066] As mentioned above, when the data packet is too large, the data in the user attribute data packet needs to be deleted, which can reduce the data error rate. Since a record corresponds to several attribute information of a user, the data in the entire record needs to be deleted when deleting. Moreover, since the data packet is sent periodically, for example, all registered users of an app, the changed users can be screened out based on the change information therein. Random deletion of data can make the changed users difficult to be identified, prevent user information leakage, and ensure that the remaining data will not affect big data analysis.

[0067] S400 : When the data priority Y corresponding to the user attribute data packet is Y0, r2 historical users and a number of user attributes corresponding to each historical user are added to the user attribute data packet.

[0068] Furthermore, r2 can be obtained by the following steps:

[0069] S401 : When the data priority Y corresponding to the user attribute data packet is less than or equal to Y0, the difference between the data packet reference size and the data packet size of the original user attribute data packet is used as a reference amount of data to be added.

[0070] S402, calculating the number of records to be added from the user attribute data packet according to the reference amount of data to be added, and taking the calculation result as r2; it can be understood that the total data amount of the deleted r1 records is closest to the reference amount of data to be deleted.

[0071] As mentioned above, when the data packet is too small, in order to ensure sufficient data volume and make it difficult to identify the changed users, the data packet needs to be increased. By randomly adding historical data to the user attribute data table, the user data in the data table can be obfuscated, thereby preventing the leakage of specific user information.

[0072] An embodiment of the present invention also provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a method in a method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0073] An embodiment of the present invention further provides an electronic device including a processor and the aforementioned non-transitory computer-readable storage medium.

[0074] Although some specific embodiments of the present invention have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A method for adjusting the amount of user attribute data, characterized in that: The method comprises the following steps: S100, receiving a user attribute data set sent by a third-party platform, processing the data in each user attribute data table in the user attribute data set according to preset compliance processing rules, and packaging the processed user attribute data tables into a user attribute data packet; the user attribute data table includes a plurality of target users and a plurality of encrypted user attributes corresponding to each target user; S200, calculating a data priority D corresponding to the user attribute data packet according to the data packet size of the user attribute data packet; The data priority Y corresponding to the user attribute data packet meets the following conditions: Y = δ + ψ; where δ is the historical data packet loss rate corresponding to the data packet size of the user attribute data packet, and ψ is the normalized historical transmission duration corresponding to the data packet size of the user attribute data packet; S300, when the data priority Y corresponding to the user attribute data packet is greater than Y0, deleting r1 target users and a number of user attributes corresponding to each deleted target user from the user attribute data packet, where Y0 is a preset data priority threshold; S400 : When the data priority Y corresponding to the user attribute data packet is Y0, r2 historical users and a number of user attributes corresponding to each historical user are added to the user attribute data packet.

2. The method for adjusting the amount of user attribute data according to claim 1, wherein: Get r1 by following the steps below: S301, Y0 is substituted into Y=δ+ψ as the new data priority Y, and a new data packet size is obtained based on the obtained new δ and ψ and used as the data packet reference size; S302, when the data priority Y corresponding to the user attribute data packet is greater than Y0, the difference between the data packet size of the original user attribute data packet and the data packet reference size is used as a reference amount of data to be deleted; S303 , calculating the number of records to be deleted from the user attribute data packet according to the reference amount of data to be deleted, and taking the calculation result as r1 ; each record includes a target user and several user attributes corresponding to the target user.

3. The method for adjusting the amount of user attribute data according to claim 2, wherein: Obtain r2 by following the steps below: S401, when the data priority Y corresponding to the user attribute data packet is less than or equal to Y0, the difference between the data packet reference size and the data packet size of the original user attribute data packet is used as a reference amount of data to be added; S402: Calculate the number of records to be added from the user attribute data packet according to the reference amount of data to be added, and use the calculation result as r2.

4. The method for adjusting the amount of user attribute data according to claim 1, wherein: Step S100 includes the following steps: S101, extract the i-th user attribute data table A i Several field names in A i Corresponding target field name list B i ={B i1 , B i2 ,……,B ij ,……,B in }, where i = 1, 2, ..., m, where m is the number of user attribute data tables in the user attribute data set, B ij Refers to A i The jth target field name in A, j = 1, 2, ..., n, n is i The number of target field names in ; the target field name is any user attribute; S102, according to B ij The corresponding preset encryption judgment rules, from B ij Determine the encrypted target field value from the corresponding initial field values ​​and calculate B ij The corresponding target encryption level F ij ; Among them, B ij The corresponding target encryption level F ij Meet the following conditions: F ij =H 0 ij / H ij , where H 0 ij For B ij The number of corresponding target field values, H ij For B ij The total number of corresponding initial field values; S103, when B ij The corresponding target encryption level F ij ≥F 0 ij When B ij Get all the initial field values ​​except the target field value from the corresponding initial field values ​​and process them to get B ij Corresponding final field values ​​to obtain compliant data, where F 0 ij is the preset encryption level threshold; S104, when B ij The corresponding target encryption level F ij <F 0 ij When the data is not in compliance, a warning message will be sent to the third-party platform so that the third-party platform can i After verifying the data in the file, upload the data again until A is obtained. i Corresponding compliance data; S105: Packing the obtained user attribute data tables with compliant data into a user attribute data packet.

5. The method for adjusting the amount of user attribute data according to claim 4, characterized in that: The preset encryption judgment rule is set based on a given decision tree; The root node and each internal node of the given decision tree correspond to preset attribute judgment conditions, and the leaf nodes of the given decision tree are encryption judgment results, which include encryption status and corresponding node encryption scores.

6. The method for adjusting the amount of user attribute data according to claim 5, characterized in that: According to B ij The corresponding preset encryption judgment rules, from B ij Determining the encrypted target field value from the corresponding initial field values ​​includes the following steps: S1021, obtain B from several historical user attribute data tables ij A number of historical field values ​​under the corresponding target field name, and input each historical field value into a given decision tree to obtain the number of historical field values ​​corresponding to each preset leaf node in the given decision tree; S1022, calculating the encryption cracking probability corresponding to each preset leaf node in a given decision tree based on the number of historical field values ​​corresponding to each preset leaf node; Among them, the encryption cracking probability corresponding to the preset leaf node of any layer meets the following conditions: P d =(∑ d c＝1 K c ) / K, where P d The encryption cracking probability corresponding to the preset leaf node of the dth layer, K c is the number of historical field values ​​corresponding to the preset leaf nodes in the cth layer, and K is the sum of the number of historical field values ​​corresponding to the preset leaf nodes in each layer; S1023, normalizing the encryption cracking probability corresponding to each layer of preset leaf nodes to between 0 and 1 to obtain a node encryption score corresponding to each layer of preset leaf nodes; S1024, for B ij Corresponding to any initial field value, the initial field value is input into a given decision tree for judgment, and the target leaf node corresponding to the initial field value is obtained; S1025 , obtaining a node encryption score corresponding to the target leaf node, and when the node encryption score is greater than a preset score threshold, determining that the initial field value is the encrypted target field value.

7. The method for adjusting the amount of user attribute data according to claim 4, characterized in that: The from B ij Get all the initial field values ​​except the target field value from the corresponding initial field values ​​and process them to get B ij The corresponding final field values ​​include the following steps: S1031, targeting B ij For any initial field value other than the target field value among the corresponding initial field values, the initial field value is identified to determine whether the initial field value meets the preset format requirement; S1032, when the initial field value meets the preset format requirements, based on the target field corresponding to the initial field value, calling the preset encryption rule corresponding to the target field and encrypting the initial field value to obtain the final field value corresponding to the initial field value; S1033, when the initial field value does not meet the preset format requirements, the record corresponding to the initial field value is deleted to obtain B ij The corresponding final field values.

8. The method for adjusting the amount of user attribute data according to claim 4, wherein: Obtain F by following the steps below 0 ij : S10, obtaining several historical user attribute data tables, and fitting a relationship curve between the number of users and the data accuracy rate according to the number of historical users and the data accuracy rate corresponding to each historical user attribute data table; S20, based on the relationship curve between the number of users and data accuracy, according to A i The number of users in A i The corresponding data accuracy η; S30, obtain B in each historical user attribute data table ij The encryption probability λ of the corresponding target field name is calculated, and the average encryption probability is obtained; S40, according to A i The corresponding data accuracy η, average encryption probability λ and the preset weights corresponding to η and λ are calculated by weighted sum to obtain F 0 ij .

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, characterized in that: The at least one instruction or the at least one program segment is loaded and executed by the processor to implement the user attribute data amount adjustment method as described in any one of claims 1-8.

10. An electronic device, characterized in that: The device comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Images