User cluster data privacy protection sharing method, system, equipment and medium
By conducting sensitivity analysis and cluster screening of the behavioral communication data of target users, combining data desensitization and analysis models, and optimizing the sensitivity threshold, the problem of difficulty in balancing privacy protection and data sharing in the existing technology is solved, and more flexible and efficient data privacy protection and sharing is achieved.
Patent Information
- Application Number
- CN202510060965.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-15
- Publication Date
- 2025-04-25
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology lacks a flexible dynamic optimization mechanism, making it difficult to meet the balance between privacy data protection and data sharing at the same time, especially in the adjustment of sensitive thresholds.
By obtaining the behavioral communication data of the target user, conducting sensitivity analysis to obtain sensitive entropy values, clustering analysis to screen initial sensitivity thresholds, randomly generate sensitivity thresholds to be screened, data desensitization, analyze privacy strength and share flexibility data, and optimize sensitivity thresholds using the desensitized data analysis model.
The flexible activity optimization of sensitive thresholds is achieved, which can find the best balance between protecting privacy and promoting data sharing, improving data security and utilization efficiency.
Smart Images
Figure CN119893500A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of privacy data protection and sharing technology, and in particular to a user cluster data privacy protection and sharing method, system, device and medium. Background Art
[0002] In traditional 5G networks, when performing privacy protection and sharing of user cluster data, it is important to choose appropriate technologies and strategies to balance the needs of privacy protection and data sharing. Through methods such as data desensitization, differential privacy, encryption technology, and federated learning, user privacy can be effectively protected while supporting data analysis and sharing. Among them, the sensitivity threshold plays a vital role in the privacy protection and sharing of user cluster data. The level of the sensitivity threshold directly affects the intensity of privacy protection. When the sensitivity threshold is low, more data will be considered sensitive information, thus imposing stronger privacy protection measures. In this case, the risk of data leakage is low, but the flexibility of data utilization and sharing may be limited. If the sensitivity threshold is high, it means that only a few data are considered sensitive information, and more data can be shared and utilized. This improves the flexibility of data sharing, but it may also increase the risk of data leakage. Therefore, it is necessary to find the best balance between protecting privacy and promoting data sharing according to actual needs by dynamically adjusting the threshold.
[0003] The current existing technologies lack a flexible dynamic optimization mechanism for sensitive thresholds, making it difficult to simultaneously balance privacy data protection and data sharing. Summary of the invention
[0004] The present invention provides a user cluster data privacy protection sharing method, system, device and medium, the main purpose of which is to solve the problem of lack of flexible dynamic optimization mechanism for sensitive thresholds and difficulty in simultaneously satisfying the balance between privacy data protection and data sharing.
[0005] In a first aspect, to achieve the above-mentioned purpose, the present invention provides a user cluster data privacy protection sharing method, comprising:
[0006] Acquire behavioral communication data of a number of target users, perform sensitivity analysis on the behavioral communication data, and obtain a number of sensitive entropy values;
[0007] Performing cluster analysis and screening on a number of the sensitive entropy values to obtain an initial sensitive threshold;
[0008] Generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor;
[0009] Desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened, and obtaining privacy strength data and sharing flexibility data in the data desensitization process;
[0010] Analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened;
[0011] The application effect value and the sensitivity threshold to be screened are used to perform data analysis to obtain an optimized sensitivity threshold, which is then applied to user cluster data privacy protection and sharing.
[0012] In a second aspect, the present invention further provides a user cluster data privacy protection sharing system, the system comprising:
[0013] A sensitivity analysis module is used to obtain the behavior communication data of several target users, perform sensitivity analysis on the behavior communication data, and obtain several sensitivity entropy values;
[0014] An initial threshold screening module is used to perform cluster analysis and screening on a plurality of the sensitive entropy values to obtain an initial sensitive threshold;
[0015] A threshold random generation module, used to generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor;
[0016] A data desensitization module, used to perform data desensitization on the behavioral communication data of each target user using the sensitivity threshold to be screened, and obtain privacy strength data and sharing flexibility data during the data desensitization process;
[0017] A model analysis module, used to analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened;
[0018] The data analysis module is used to perform data analysis using the application effect value and the sensitivity threshold to be screened, obtain an optimized sensitivity threshold, and apply it to user cluster data privacy protection and sharing.
[0019] In a third aspect, the present invention further provides an electronic device, the electronic device comprising:
[0020] at least one processor; and,
[0021] a memory communicatively connected to the at least one processor; wherein,
[0022] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned user cluster data privacy protection sharing method.
[0023] In a fourth aspect, the present invention further provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is executed by a processor in an electronic device to implement the above-mentioned user cluster data privacy protection sharing method.
[0024] The present invention obtains the behavior communication data of several target users, performs sensitivity analysis on the behavior communication data to obtain several sensitive entropy values, performs cluster analysis and screening on the several sensitive entropy values to obtain an initial sensitive threshold, uses the initial sensitive threshold and a preset change factor to generate several different random sensitive thresholds to be screened, uses the sensitive threshold to be screened to perform data desensitization on the behavior communication data of each of the target users, obtains privacy strength data and sharing flexibility data in the data desensitization process, uses a preset desensitized data analysis model to analyze the privacy strength data and the sharing flexibility data to obtain application effect values of the sensitive threshold to be screened, uses the application effect values and the sensitive threshold to be screened to perform data analysis to obtain optimized sensitive thresholds, and applies them to privacy protection and sharing of user cluster data, thereby effectively solving the problem of lack of a flexible dynamic optimization mechanism for sensitive thresholds and difficulty in simultaneously satisfying the balance between privacy data protection and data sharing. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0026] Figure 1 A flowchart of a user cluster data privacy protection sharing method provided by an embodiment of the present invention;
[0027] Figure 2 A schematic diagram of a module of a user cluster data privacy protection sharing system provided by an embodiment of the present invention;
[0028] Figure 3 A schematic diagram of the structure of an electronic device for implementing a user cluster data privacy protection sharing method provided by an embodiment of the present invention;
[0029] Figure 4It is another structural schematic diagram of an electronic device for implementing a user cluster data privacy protection sharing method provided by an embodiment of the present invention.
[0030] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0031] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, and to fully understand and implement how the present disclosure applies technical means to solve technical problems and achieve the corresponding technical effects, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only embodiments of a part of the present disclosure, not all of the embodiments. The embodiments of the present disclosure and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.
[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.
[0033] The embodiment of the present application provides a user cluster data privacy protection sharing method, and the execution subject of the user cluster data privacy protection sharing method includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the system provided by the embodiment of the present application. In other words, the user cluster data privacy protection sharing method can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.
[0034] Reference Figure 1 FIG. 1 is a flow chart of a method for sharing user cluster data with privacy protection provided by an embodiment of the present invention. In this embodiment, the method for sharing user cluster data with privacy protection includes:
[0035] S1. Obtaining behavior communication data of several target users, performing sensitivity analysis on the behavior communication data, and obtaining several sensitive entropy values.
[0036] In an embodiment of the present invention, relevant behavioral communication data are collected according to the target user group, including: user click records, browsing history, social media interactions, purchase records, search queries, etc., and the behavioral communication data is subjected to sensitivity analysis through Shannon information entropy to obtain the sensitive entropy value of each behavioral communication data.
[0037] In detail, the sensitivity analysis is performed on the behavior communication data to obtain several sensitive entropy values, including:
[0038] Acquire several target data features of the behavioral communication data;
[0039] Converting the target data features into binary feature data;
[0040] Counting the occurrence probabilities of the binary feature data one by one;
[0041] The occurrence probability is used to calculate the sensitivity entropy value of the behavior communication data.
[0042] In detail, the target data feature of behavioral communication data refers to an item in the behavioral communication data of the target user, including: user click records, browsing history, social media interaction, etc. The target data feature is converted into binary feature data. The specific operation is as follows: map each target data feature to a number, for example, the code for "user click record" is 0, the code for "browsing history" is 1, and the code for "social media interaction" is 2, and 2 is converted to binary as 10.
[0043] In detail, the occurrence probabilities of the binary feature data are counted one by one, and the sensitive entropy value of the behavior communication data is calculated using the occurrence probabilities. The calculation formula is as follows:
[0044]
[0045]
[0046] Among them, f(x i ) represents the number of occurrences of the i-th binary feature data, p(x i ) represents the occurrence probability of each binary feature data, x i represents the i-th binary feature data, n represents the total number of binary feature data, and H(X) represents the sensitive entropy value of the behavioral communication data.
[0047] If the sensitive entropy value of the data is high, it means that the information content is rich and difficult to predict, and therefore has a higher sensitivity; if the sensitive entropy value of the data is low, it means that the information content is relatively certain, and therefore has a lower sensitivity.
[0048] By performing sensitivity analysis on behavioral communication data, the privacy risk and sensitivity of the data can be accurately assessed, thus providing a scientific basis for data protection. Converting target data features into binary features and calculating the sensitive entropy value of behavioral communication data can help quantify data uncertainty and the possibility of information leakage. In this way, the sensitivity of data can be dynamically identified and classified, thereby more effectively supporting data sharing and utilization while ensuring privacy protection.
[0049] S2. Perform cluster analysis and screening on a plurality of the sensitive entropy values to obtain an initial sensitive threshold.
[0050] In the embodiment of the present invention, K- means The algorithm divides the sensitive entropy values into a number of clusters, and selects the cluster with the most sensitive entropy value data as the initial sensitive threshold.
[0051] In detail, cluster analysis and screening of a plurality of the sensitive entropy values to obtain an initial sensitive threshold value includes:
[0052] S21, obtaining the number of clusters, and randomly selecting a number of sensitive entropy values corresponding to the number of clusters as initial cluster centers;
[0053] S22, randomly selecting one of the sensitive entropy values as the target entropy value;
[0054] S23, calculating the distance between the target entropy value and each of the initial cluster centers, and selecting the initial cluster center corresponding to the minimum distance;
[0055] S24, assigning the target entropy value to the initial cluster center corresponding to the minimum distance to obtain an updated cluster;
[0056] S25, calculating the update cluster center according to the sensitive entropy value in the update cluster;
[0057] Iterate steps S22-S25, and when the center of each updated cluster no longer changes, several final clusters are obtained;
[0058] Counting the total number of sensitive entropy values in the final clusters, sorting the total number from large to small to select the final cluster with the largest total number;
[0059] The final cluster center among the final clusters with the largest total number is used as an initial sensitive threshold.
[0060] Specifically, the distance between the target entropy value and each of the initial cluster centers is calculated using the following calculation formula:
[0061]
[0062] Among them, H(X m ) represents the horizontal coordinate of the data point of the mth target entropy value, H(Y m ) represents the ordinate of the data point of the mth target entropy value, F(X m , Y m ) represents the distance between the target entropy value and the center of each initial cluster.
[0063] The final cluster center is calculated based on the sensitive entropy value in the final cluster. The calculation formula is as follows:
[0064]
[0065] Among them, w m The weight of the data point representing the mth target entropy value, H(X m ) represents the horizontal coordinate of the data point of the mth target entropy value, M represents the total number of sensitive entropy values in the final cluster, C H ′ represents the initial sensitivity threshold.
[0066] By systematically grouping and screening the sensitive entropy values, representative sensitive thresholds can be effectively identified. First, randomly selecting cluster centers and iteratively updating them can ensure accurate division of the data and avoid bias in the results caused by the initial selection. By finally screening out the cluster with the largest total number, it is ensured that the selected sensitive threshold can represent the overall characteristics of the data, which improves the accuracy and reliability of the threshold, reduces human intervention, improves the objectivity of the analysis results, and helps to extract valuable key information from complex data.
[0067] S3. Generate a number of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor.
[0068] In an embodiment of the present invention, based on the screened initial sensitivity threshold, multiple different sensitivity thresholds to be screened are generated according to preset change factors of a small range of numerical values. These randomly generated thresholds provide multiple candidate values for further screening and optimization, ensuring that the final sensitivity threshold can be better applied to user cluster data privacy protection and sharing, thereby improving the accuracy and reliability of the processing results.
[0069] In detail, the method of generating a plurality of different random sensitivity thresholds to be screened by using the initial sensitivity threshold and a preset variation factor includes:
[0070] Randomly generate a number of changes according to a preset change factor;
[0071] The initial sensitivity threshold is multiplied by the change amount to obtain a plurality of sensitivity thresholds to be screened.
[0072] In detail, assuming that the variation factor is α, which is a small range of values, such as 0.05≤α≤0.2, the variation factor controls the fluctuation range of the generated random sensitivity threshold. Based on the variation factor, several variation quantities are randomly generated, and the calculation formula is as follows:
[0073] ΔT=α*C H ′*random_factor
[0074] Among them, α represents the preset change factor, C H ′ represents the initial sensitivity threshold, and random_factor represents the value generated in the random distribution.
[0075] The initial sensitivity threshold is multiplied by the change amount to obtain a plurality of sensitivity thresholds to be screened, and the calculation formula is as follows:
[0076] C new =C H ′*(1+ΔT)
[0077] Among them, C H′ represents the initial sensitivity threshold, and ΔT represents the change amount.
[0078] By using the initial sensitive threshold and the preset variation factor to generate multiple random sensitive thresholds to be screened, the diversity and randomness of the thresholds can be effectively increased, thereby improving the robustness and flexibility of the screening process. The randomly generated variation can avoid the limitations of fixed thresholds, optimize the selection range of sensitive thresholds, and improve screening accuracy and adaptability.
[0079] S4. Desensitize the behavioral communication data of each target user using the sensitivity threshold to be screened, and obtain privacy strength data and sharing flexibility data during the data desensitization process.
[0080] In an embodiment of the present invention, if the sensitive entropy value of the behavioral communication data is greater than the sensitive threshold to be screened, Laplace noise is used to perturb the behavioral communication data to complete data desensitization; if the sensitive entropy value of the behavioral communication data is less than or equal to the sensitive threshold to be screened, the behavioral communication data is anonymized to complete data desensitization. The balance of the data desensitization process can be analyzed through privacy strength data and sharing flexibility data. There is a trade-off between privacy strength and sharing flexibility: excessive desensitization may lead to increased privacy strength, but reduced sharing flexibility; excessive sharing flexibility may lead to an increased risk of privacy leakage.
[0081] In detail, the method of desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened includes:
[0082] Randomly select a sensitive threshold to be screened and behavioral communication data of a target user;
[0083] Determining whether the sensitive entropy value of the behavior communication data is greater than the sensitivity threshold to be screened;
[0084] If the sensitive entropy value of the behavior communication data is greater than the sensitive threshold to be screened, obtaining a disturbance intensity parameter and a scale parameter;
[0085] generating a noise value using the disturbance intensity parameter and the scale parameter;
[0086] Randomly adding the noise value to the behavior communication data whose sensitive entropy value is greater than the sensitive threshold to be screened, to obtain desensitized communication data;
[0087] If the sensitive entropy value of the behavior communication data is less than or equal to the sensitivity threshold to be screened, dividing the behavior communication data whose sensitive entropy value is less than or equal to the sensitivity threshold to be screened into sensitive attribute data and identification attribute data;
[0088] The identification attribute data is generalized to obtain desensitized communication data.
[0089] Specifically, the disturbance intensity parameter and the scale parameter are obtained, and the noise value is generated using the disturbance intensity parameter and the scale parameter. The calculation formula is as follows:
[0090] NNN=λLap(b)
[0091] Among them, λ represents the perturbation intensity parameter, b represents the scale parameter of the Laplace distribution, and NNN represents the noise value.
[0092] The noise value is randomly added to the behavioral communication data whose sensitive entropy value is greater than the sensitive threshold to be screened. The specific operation is as follows: randomly select part of the behavioral communication data, add the noise value to the randomly selected part of the data to obtain updated data, and re-fill the updated data into the behavioral communication data to complete data desensitization.
[0093] In detail, the behavior communication data whose sensitive entropy value is less than or equal to the sensitive threshold to be screened is divided into sensitive attribute data and identification attribute data. The specific operation is as follows: the identifiable user privacy data in the behavior communication data, such as user name, user address, etc., is used as identification attribute data, and the data that can be used for analysis in the behavior communication data, such as user click records, purchase records, etc., is used as sensitive attribute data. Generalization is to convert certain specific values into more vague categories, generalize the identification attribute data, and complete data desensitization. The specific operation is as follows: for example, the user name is converted to only retain the user's surname, and the user address is converted to only retain the user's province.
[0094] By randomly adding noise, the risk of sensitive information leakage can be effectively reduced, and malicious users can be prevented from reversely deducing personal privacy data. At the same time, sensitive attributes and identification attributes are processed separately using sensitive thresholds, further enhancing data security. This method ensures data availability and applicability while protecting data privacy, providing a diversified process for obtaining privacy strength data and sharing flexibility data.
[0095] In detail, the privacy strength data and sharing flexibility data obtained during the data desensitization process include:
[0096] Counting the number of occurrences of target characteristic data of each of the desensitized communication data;
[0097] Calculate the desensitization entropy value of each desensitized communication data using the number of occurrences;
[0098] Taking the desensitized entropy value and the sensitive entropy value as quotient to obtain an entropy value ratio;
[0099] Generate a number of privacy strength data using the entropy value ratio;
[0100] Performing logistic regression on each of the behavioral communication data and the corresponding desensitized communication data to obtain a first probability and a second probability;
[0101] Each group of shared flexibility data is obtained by taking the quotient of the first probability and the second probability.
[0102] In detail, privacy strength measures the degree of privacy protection during the data desensitization process, and usually uses the entropy ratio as an indicator to measure the degree of desensitization and privacy strength. The entropy ratio reflects the difference in the amount of information between desensitized communication data and behavioral communication data. A lower entropy ratio means that the desensitized data has lost more information and the privacy protection strength may be stronger; a higher entropy ratio means that the desensitized data still retains more information and the privacy protection strength is lower.
[0103] Furthermore, the number of occurrences of the target feature data of each desensitized communication data is counted, and the desensitization entropy value of each desensitized communication data is calculated using the number of occurrences. The calculation formula is as follows:
[0104]
[0105]
[0106] Among them, f(r g ) represents the number of occurrences of the g-th target feature data, p(r g ) represents the occurrence probability of each target feature data, r g represents the g-th target feature data, r h represents the total number of target feature data, and H(R) represents the desensitized entropy value of the desensitized communication data.
[0107] The desensitized entropy value and the sensitive entropy value are divided to obtain an entropy value ratio, and the entropy value ratio is used to generate a number of privacy strength data. The calculation formula is as follows:
[0108]
[0109] Among them, H(R) represents the desensitized entropy value of desensitized communication data, H(X) represents the sensitive entropy value of behavioral communication data, and H RX Represents the entropy ratio.
[0110] P H =1-H RX
[0111] Among them, H RX represents the entropy ratio, P H Represents privacy strength data.
[0112] In detail, sharing flexibility reflects the impact of desensitization on the predictability and information sharing of behavioral communication data. If the desensitization affects the prediction accuracy of the logistic regression model, the ratio between the first probability and the second probability will change significantly. A high sharing flexibility value indicates that the desensitized communication data still retains more predictive information and is close to the probability of the behavioral communication data; a low sharing flexibility value indicates that the desensitized communication data has lost a lot of information, and the predicted probability is quite different from the original data. The main purpose of desensitized data is to protect the user's privacy information, but excessive privacy protection may lose the availability of data and affect the prediction ability of the model. Sharing flexibility data provides a measure between privacy protection and data availability.
[0113] Furthermore, logistic regression is performed on each of the behavioral communication data and the corresponding desensitized communication data to obtain a first probability and a second probability. The calculation formula is as follows:
[0114]
[0115] Among them, d0, d1, d2, ..., d n represents the logistic regression parameters, Q1, Q2, ..., Q n represents behavioral communication data, y represents the category predicted to have data value, and P1 represents the first probability.
[0116]
[0117] Among them, d0, d1, d2, ...., d n Represents logistic regression parameters, R1, R2, ...., R n represents the desensitized communication data, y represents the category predicted to have data value, and P2 represents the second probability.
[0118] In detail, the first probability and the second probability are used as a quotient to obtain each group of sharing flexibility data, and the calculation formula is as follows:
[0119]
[0120] Among them, P2 represents the second probability, P1 represents the first probability, and L P Represents shared flexibility data.
[0121] Privacy strength data can ensure that users' sensitive information is effectively protected, reduce the risk of data leakage, and protect users' privacy rights. At the same time, sharing flexibility data allows specific data to be flexibly shared as needed while ensuring privacy security, promoting the effective use and collaborative analysis of data. The combination of the two not only improves the security and privacy protection level of data, but also enhances the efficiency and flexibility of data sharing, which helps to analyze the application effect of the sensitive threshold to be screened on behavioral communication data.
[0122] S5. Analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitive threshold to be screened.
[0123] In an embodiment of the present invention, a desensitized data analysis model of a multi-layer perceptron is used to analyze the privacy strength data and the sharing flexibility data to obtain the application effect value of each sensitive threshold to be screened.
[0124] In detail, the privacy strength data and the sharing flexibility data are analyzed by using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened, including:
[0125] Taking the privacy strength data and the sharing flexibility data as input vectors;
[0126] Obtain a first weight matrix, a first bias term, and an activation function, and construct a first hidden layer using the first weight matrix, the first bias term, and the activation function;
[0127] Sending the input vector to the first hidden layer using a preset desensitized data analysis model to obtain a first hidden result;
[0128] Obtaining a second weight matrix and a second bias item, and constructing a second hidden layer using the second weight matrix and the second bias item;
[0129] The first hidden result is analyzed by using the second hidden layer to obtain an application effect value of the sensitive threshold to be screened.
[0130] In detail, the desensitized data analysis model includes: input layer, first hidden layer, second hidden layer and output layer. The specific data transmission process is as follows:
[0131] Input vector: E(privacy strength data P H , shared flexibility data L P ), the input vector is passed into the neural network through the input layer;
[0132] The first hidden layer: E (privacy strength data P H , shared flexibility data L P) is linearly transformed through the first weight matrix W1 and the first bias term is added to obtain the output I1. The activation function is applied to I1 to obtain the output of the first hidden layer. The calculation formula is as follows:
[0133] I1=W1*E+s1
[0134] a1=σ(I1)
[0135] Among them, W1 represents the first weight matrix, E represents the input vector, s1 represents the first bias term, and σ represents the activation function.
[0136] Second hidden layer: The output of the first hidden layer is passed to the second hidden layer, and the final output is obtained through the second weight matrix W2 and the second bias term s2. The calculation formula is as follows:
[0137] I2=W2*a1+s2
[0138] Wherein, W2 represents the second weight matrix, a1 represents the first hidden result, and s2 represents the second bias term.
[0139] Output layer: The result of the output layer is the application effect value of the sensitive threshold to be screened. The calculation formula is as follows:
[0140] R * =k(W2*σ(W1*E+s1)+s2)
[0141] Among them, W1 represents the first weight matrix, E represents the input vector, s1 represents the first bias term, σ represents the activation function, W2 represents the second weight matrix, s2 represents the second bias term, and k represents the desensitized data analysis model.
[0142] Multilayer perceptrons are effectively used to balance the relationship between privacy protection and data sharing. The desensitized data analysis model outputs the application effect of the sensitive threshold to be screened on behavioral communication data through training, providing data support for the subsequent screening of the optimal sensitive threshold for privacy protection and sharing of user cluster data.
[0143] S6. Perform data analysis using the application effect value and the sensitivity threshold to be screened to obtain an optimized sensitivity threshold, and apply it to user cluster data privacy protection and sharing.
[0144] In an embodiment of the present invention, a line graph is drawn using the application effect value obtained through the desensitized data analysis model and the corresponding sensitive threshold to be screened. The optimized sensitive threshold is screened out from the sensitive threshold to be screened through the line graph and applied to user cluster data privacy protection and sharing.
[0145] In detail, the data analysis is performed using the application effect value and the sensitivity threshold to be screened to obtain an optimized sensitivity threshold, including:
[0146] Each sensitivity threshold to be screened is used as the horizontal axis, and the application effect value corresponding to each sensitivity threshold to be screened is used as the vertical axis;
[0147] Using the abscissa and the ordinate, each sensitivity threshold to be screened and the corresponding application effect value are taken as a set of target data;
[0148] Plotting the target data into a line graph;
[0149] Filter out the highest point in the line graph according to the line graph;
[0150] An optimized sensitivity threshold is generated according to the highest point.
[0151] In detail, a line graph is drawn by using the application effect value obtained by the desensitized data analysis model and the corresponding sensitive threshold to be screened, and the highest point in the line graph is screened out based on the line graph. This shows that when the threshold to be screened corresponding to the highest point is selected, the effect of data desensitization is best, that is, the privacy protection and sharing of user cluster data are optimal, while satisfying the balance between privacy data protection and data sharing.
[0152] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0153] like Figure 2 , which is a functional module diagram of a user cluster data privacy protection and sharing system provided by an embodiment of the present invention.
[0154] In the embodiment of the present disclosure, a user cluster data privacy protection sharing system is provided, and the user cluster data privacy protection sharing system corresponds one-to-one to the user cluster data privacy protection sharing method in the above embodiment. Figure 2 As shown, the user cluster data privacy protection sharing system 100 can be installed in an electronic device. According to the functions to be implemented, the user cluster data privacy protection sharing system 100 includes a sensitivity analysis module 101, an initial threshold screening module 102, a threshold random generation module 103, a data desensitization module 104, a model analysis module 105 and a data analysis module 106. The functional modules are described in detail as follows:
[0155] The sensitivity analysis module 101 is used to obtain the behavior communication data of a plurality of target users, perform sensitivity analysis on the behavior communication data, and obtain a plurality of sensitivity entropy values;
[0156] The initial threshold screening module 102 is used to perform cluster analysis and screening on a plurality of the sensitive entropy values to obtain an initial sensitive threshold;
[0157] A threshold random generation module 103, used to generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor;
[0158] A data desensitization module 104 is used to perform data desensitization on the behavior communication data of each target user using the sensitivity threshold to be screened, and obtain privacy strength data and sharing flexibility data during the data desensitization process;
[0159] A model analysis module 105 is used to analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened;
[0160] The data analysis module 106 is used to perform data analysis using the application effect value and the sensitivity threshold to be screened, obtain an optimized sensitivity threshold, and apply it to user cluster data privacy protection and sharing.
[0161] In one embodiment, the sensitivity analysis module 101 performs sensitivity analysis on the behavior communication data to obtain a plurality of sensitivity entropy values for:
[0162] Acquire several target data features of the behavioral communication data;
[0163] Converting the target data features into binary feature data;
[0164] Counting the occurrence probabilities of the binary feature data one by one;
[0165] The occurrence probability is used to calculate the sensitivity entropy value of the behavior communication data.
[0166] In one embodiment, the initial threshold screening module 102 performs cluster analysis and screening on a plurality of the sensitive entropy values to obtain an initial sensitive threshold for:
[0167] S21, obtaining the number of clusters, and randomly selecting a number of sensitive entropy values corresponding to the number of clusters as initial cluster centers;
[0168] S22, randomly selecting one of the sensitive entropy values as the target entropy value;
[0169] S23, calculating the distance between the target entropy value and each of the initial cluster centers, and selecting the initial cluster center corresponding to the minimum distance;
[0170] S24, assigning the target entropy value to the initial cluster center corresponding to the minimum distance to obtain an updated cluster;
[0171] S25, calculating the update cluster center according to the sensitive entropy value in the update cluster;
[0172] Iterate steps S22-S25, and when the center of each updated cluster no longer changes, several final clusters are obtained;
[0173] Counting the total number of sensitive entropy values in the final clusters, sorting the total number from large to small to select the final cluster with the largest total number;
[0174] The final cluster center among the final clusters with the largest total number is used as an initial sensitive threshold.
[0175] In one embodiment, the threshold random generation module 103 generates a plurality of different random sensitivity thresholds to be screened by using the initial sensitivity threshold and the preset variation factor, for:
[0176] Randomly generate a number of changes according to a preset change factor;
[0177] The initial sensitivity threshold is multiplied by the change amount to obtain a plurality of sensitivity thresholds to be screened.
[0178] In one embodiment, the data desensitization module 104 performs data desensitization on the behavior communication data of each target user using the sensitivity threshold to be screened, for:
[0179] Randomly select a sensitive threshold to be screened and behavioral communication data of a target user;
[0180] Determining whether the sensitive entropy value of the behavior communication data is greater than the sensitivity threshold to be screened;
[0181] If the sensitive entropy value of the behavior communication data is greater than the sensitive threshold to be screened, obtaining a disturbance intensity parameter and a scale parameter;
[0182] generating a noise value using the disturbance intensity parameter and the scale parameter;
[0183] Randomly adding the noise value to the behavior communication data whose sensitive entropy value is greater than the sensitive threshold to be screened, to obtain desensitized communication data;
[0184] If the sensitive entropy value of the behavior communication data is less than or equal to the sensitivity threshold to be screened, dividing the behavior communication data whose sensitive entropy value is less than or equal to the sensitivity threshold to be screened into sensitive attribute data and identification attribute data;
[0185] The identification attribute data is generalized to obtain desensitized communication data.
[0186] In one embodiment, the data desensitization module 104, when performing the acquisition of privacy strength data and sharing flexibility data in the data desensitization process, is used to:
[0187] Counting the number of occurrences of target characteristic data of each of the desensitized communication data;
[0188] Calculate the desensitization entropy value of each desensitized communication data using the number of occurrences;
[0189] Taking the desensitized entropy value and the sensitive entropy value as quotient to obtain an entropy value ratio;
[0190] Generate a number of privacy strength data using the entropy value ratio;
[0191] Performing logistic regression on each of the behavioral communication data and the corresponding desensitized communication data to obtain a first probability and a second probability;
[0192] Each group of shared flexibility data is obtained by taking the quotient of the first probability and the second probability.
[0193] In one embodiment, the model analysis module 105 analyzes the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of the sensitivity threshold to be screened, which is used to:
[0194] Taking the privacy strength data and the sharing flexibility data as input vectors;
[0195] Obtain a first weight matrix, a first bias term, and an activation function, and construct a first hidden layer using the first weight matrix, the first bias term, and the activation function;
[0196] Sending the input vector to the first hidden layer using a preset desensitized data analysis model to obtain a first hidden result;
[0197] Obtaining a second weight matrix and a second bias item, and constructing a second hidden layer using the second weight matrix and the second bias item;
[0198] The first hidden result is analyzed by using the second hidden layer to obtain an application effect value of the sensitive threshold to be screened.
[0199] In the present invention, for a user cluster data privacy protection and sharing method, by obtaining the behavior communication data of several target users, sensitivity analysis is performed on the behavior communication data to obtain several sensitive entropy values, cluster analysis and screening are performed on the several sensitive entropy values to obtain an initial sensitive threshold, the initial sensitive threshold and a preset change factor are used to generate several different random sensitive thresholds to be screened, the sensitive threshold to be screened is used to desensitize the behavior communication data of each target user, and the privacy strength data and sharing flexibility data in the data desensitization process are obtained. The privacy strength data and the sharing flexibility data are analyzed using a preset desensitized data analysis model to obtain the application effect value of the sensitive threshold to be screened, and the application effect value and the sensitive threshold to be screened are used for data analysis to obtain an optimized sensitive threshold, which is applied to the privacy protection and sharing of user cluster data, and a dynamic optimization mechanism for sensitive thresholds is designed to effectively meet the needs of privacy data protection and data sharing at the same time. For the specific definition of a user cluster data privacy protection and sharing system, please refer to the definition of a user cluster data privacy protection and sharing method mentioned above, which will not be repeated here. Each module in the above-mentioned user cluster data privacy protection sharing system can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.
[0200] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, it implements the functions or steps on the service side of a user cluster data privacy protection sharing method.
[0201] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the functions or steps on the client side of a user cluster data privacy protection sharing method.
[0202] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:
[0203] Acquire behavioral communication data of a number of target users, perform sensitivity analysis on the behavioral communication data, and obtain a number of sensitive entropy values;
[0204] Performing cluster analysis and screening on a number of the sensitive entropy values to obtain an initial sensitive threshold;
[0205] Generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor;
[0206] Desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened, and obtaining privacy strength data and sharing flexibility data in the data desensitization process;
[0207] Analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened;
[0208] The application effect value and the sensitivity threshold to be screened are used to perform data analysis to obtain an optimized sensitivity threshold, which is then applied to user cluster data privacy protection and sharing.
[0209] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and systems can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.
[0210] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0211] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.
[0212] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.
[0213] In some implementations of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.
[0214] The readable storage medium of the present invention stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can achieve:
[0215] Acquire behavioral communication data of a number of target users, perform sensitivity analysis on the behavioral communication data, and obtain a number of sensitive entropy values;
[0216] Performing cluster analysis and screening on a number of the sensitive entropy values to obtain an initial sensitive threshold;
[0217] Generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor;
[0218] Desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened, and obtaining privacy strength data and sharing flexibility data in the data desensitization process;
[0219] Analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened;
[0220] The application effect value and the sensitivity threshold to be screened are used to perform data analysis to obtain an optimized sensitivity threshold, which is then applied to user cluster data privacy protection and sharing.
[0221] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.
[0222] The computer-readable storage medium may also store at least one computer executable program / instruction, which may be, for example, a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.
[0223] In addition, the computer device may also include (but not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (eg, keyboard, mouse, speaker, etc.), etc.
[0224] The processor may communicate with external devices via an I / O bus via a wired or wireless network.
[0225] In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and / or method steps in the embodiments described in the present technology.
[0226] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0227] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.
[0228] In the embodiments provided in the present disclosure, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to the multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0229] It should be noted that in the present disclosure, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element limited by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0230] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A user cluster data privacy protection sharing method, characterized in that: The method comprises: Acquire behavioral communication data of a number of target users, perform sensitivity analysis on the behavioral communication data, and obtain a number of sensitive entropy values; Performing cluster analysis and screening on a number of the sensitive entropy values to obtain an initial sensitive threshold; Generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor; Desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened, and obtaining privacy strength data and sharing flexibility data in the data desensitization process; Analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened; The application effect value and the sensitivity threshold to be screened are used to perform data analysis to obtain an optimized sensitivity threshold, which is then applied to user cluster data privacy protection and sharing.
2. A user cluster data privacy protection sharing method as claimed in claim 1, characterized in that: The sensitivity analysis of the behavior communication data is performed to obtain a number of sensitive entropy values, including: Acquire several target data features of the behavioral communication data; Converting the target data features into binary feature data; Counting the occurrence probabilities of the binary feature data one by one; The occurrence probability is used to calculate the sensitivity entropy value of the behavior communication data.
3. A user cluster data privacy protection sharing method as claimed in claim 1, characterized in that: The clustering analysis and screening of the plurality of sensitive entropy values to obtain an initial sensitive threshold value includes: S21, obtaining the number of clusters, and randomly selecting a number of sensitive entropy values corresponding to the number of clusters as initial cluster centers; S22, randomly selecting one of the sensitive entropy values as the target entropy value; S23, calculating the distance between the target entropy value and each of the initial cluster centers, and selecting the initial cluster center corresponding to the minimum distance; S24, assigning the target entropy value to the initial cluster center corresponding to the minimum distance to obtain an updated cluster; S25, calculating the update cluster center according to the sensitive entropy value in the update cluster; Iterate steps S22-S25, and when the center of each updated cluster no longer changes, several final clusters are obtained; Counting the total number of sensitive entropy values in the final clusters, sorting the total number from large to small to select the final cluster with the largest total number; The final cluster center among the final clusters with the largest number is used as an initial sensitive threshold.
4. A user cluster data privacy protection sharing method as claimed in claim 1, characterized in that: The method of generating a plurality of different random sensitivity thresholds to be screened by using the initial sensitivity threshold and the preset variation factor includes: Randomly generate a number of variation amounts according to a preset variation factor; The initial sensitivity threshold is multiplied by the change amount to obtain a plurality of sensitivity thresholds to be screened.
5. A user cluster data privacy protection sharing method as claimed in claim 1, characterized in that: The method of desensitizing the behavior communication data of each target user by using the sensitivity threshold to be screened includes: Randomly select a sensitive threshold to be screened and behavioral communication data of a target user; Determining whether the sensitive entropy value of the behavior communication data is greater than the sensitivity threshold to be screened; If the sensitive entropy value of the behavior communication data is greater than the sensitive threshold to be screened, obtaining a disturbance intensity parameter and a scale parameter; generating a noise value using the disturbance intensity parameter and the scale parameter; Randomly adding the noise value to the behavior communication data whose sensitive entropy value is greater than the sensitive threshold to be screened, to obtain desensitized communication data; If the sensitive entropy value of the behavior communication data is less than or equal to the sensitivity threshold to be screened, dividing the behavior communication data whose sensitive entropy value is less than or equal to the sensitivity threshold to be screened into sensitive attribute data and identification attribute data; The identification attribute data is generalized to obtain desensitized communication data.
6. A user cluster data privacy protection sharing method as claimed in claim 5, characterized in that: The privacy strength data and sharing flexibility data obtained during the data desensitization process include: Counting the number of occurrences of target characteristic data of each of the desensitized communication data; Calculate the desensitization entropy value of each desensitized communication data using the number of occurrences; Taking the desensitized entropy value and the sensitive entropy value as quotient to obtain an entropy value ratio; Generate a number of privacy strength data using the entropy value ratio; Performing logistic regression on each of the behavioral communication data and the corresponding desensitized communication data to obtain a first probability and a second probability; Each group of shared flexibility data is obtained by taking the quotient of the first probability and the second probability.
7. A user cluster data privacy protection sharing method as claimed in claim 1, characterized in that: The method of analyzing the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened includes: Taking the privacy strength data and the sharing flexibility data as input vectors; Obtain a first weight matrix, a first bias term, and an activation function, and construct a first hidden layer using the first weight matrix, the first bias term, and the activation function; Sending the input vector to the first hidden layer using a preset desensitized data analysis model to obtain a first hidden result; Obtaining a second weight matrix and a second bias item, and constructing a second hidden layer using the second weight matrix and the second bias item; The first hidden result is analyzed by using the second hidden layer to obtain an application effect value of the sensitive threshold to be screened.
8. A user cluster data privacy protection sharing system, characterized in that: The system comprises: A sensitivity analysis module is used to obtain the behavior communication data of several target users, perform sensitivity analysis on the behavior communication data, and obtain several sensitivity entropy values; An initial threshold screening module is used to perform cluster analysis and screening on a plurality of the sensitive entropy values to obtain an initial sensitive threshold; A threshold random generation module, used to generate a plurality of different random sensitivity thresholds to be screened using the initial sensitivity threshold and a preset variation factor; A data desensitization module, used to perform data desensitization on the behavior communication data of each target user using the sensitivity threshold to be screened, and obtain privacy strength data and sharing flexibility data during the data desensitization process; A model analysis module, used to analyze the privacy strength data and the sharing flexibility data using a preset desensitized data analysis model to obtain an application effect value of a sensitivity threshold to be screened; The data analysis module is used to perform data analysis using the application effect value and the sensitivity threshold to be screened, obtain an optimized sensitivity threshold, and apply it to user cluster data privacy protection and sharing.
9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute a user cluster data privacy protection sharing method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a user cluster data privacy protection sharing method as described in any one of claims 1 to 7 is implemented.