Financial data security storage method and system

Abstract

The invention relates to the technical field of financial data processing, in particular to a financial data security storage method and system, and the method comprises the steps: obtaining salary data of all employees in at least one year of a current month in financial data; calculating the overall salary core degree of each employee according to the salary data; according to the total salaries of the employees in the salary data, sorting all the employees in the current month according to a preset sorting rule to obtain an employee sorting result; calculating the salary similarity between every two employees in each month based on the salary data, the salary overall core degree and the employee sorting result; according to the salary similarity, carrying out clustering processing on the salary data of all employees in each month to obtain a clustering result; and according to a clustering result, carrying out partitioned secure storage on salary data of each month in the financial data. According to the scheme provided by the invention, the distance between the core employee and the surrounding data points can be shortened, the clustering precision is improved, and the storage reliability and safety of the financial data are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of financial data processing, and particularly relates to a method and system for secure storage of financial data. Background Art

[0002] Financial data usually includes revenue data, expenditure data, asset data, etc. Among them, the salary data of employees every month is one of the core parts of financial data, which is crucial for the company's budget formulation, cost control and financial planning. And salary data is sensitive personal information, including employees' income, tax situation, etc. Enterprises need to properly store employees' salary data to prevent data leakage. Therefore, it is crucial to ensure the secure storage of employees' salary data every month in financial data.

[0003] In the related art, usually the clustering algorithm is directly used to classify the salary data of all employees and used as the basis for block storage. However, due to the large differences in the job responsibilities, positions and experiences of different employees, if the total salary of core employees is quite different from that of other surrounding employees, that is, the density around their salaries is low. In this case, directly using the original distance metric may not effectively capture the potential similarities between employees, and thus the salary data of core employees is likely to be mislabeled as noise points, making it difficult to form clusters. Further, due to the difficulty in guaranteeing the accuracy of the clustering result, the storage reliability and security of financial data are affected. Summary of the Invention

[0004] In order to solve the technical problem that the traditional block security storage scheme has low clustering accuracy, resulting in insufficient storage reliability and security of financial data, the purpose of the present invention is to provide a method and system for secure storage of financial data, and the specific technical solutions adopted are as follows: On the one hand, the present invention provides a method for secure storage of financial data, and the method includes: Obtain the salary data of all employees in at least one year of the current month in the financial data; Calculate the overall salary core degree of each employee according to the salary data; Sort all employees in the current month according to the total salary of employees in the salary data according to the preset sorting rule to obtain the employee sorting result; Calculate the salary similarity between every two employees in each month based on the salary data, the overall salary core degree and the employee sorting result; Cluster the salary data of all employees in each month according to the salary similarity to obtain the clustering result; Perform block secure storage on the salary data of each month in the financial data according to the clustering result.

[0005] A financial data security storage method provided by the present invention calculates the overall salary core degree of each employee based on the salary data, including: Based on each salary component value in the current month of the salary data, determine the salary component core degree and the core contribution degree corresponding to each salary component value of each employee in the current month; Based on the salary component core degree and the core contribution degree, determine the single salary core degree of each employee in the current month; Based on the single salary core degree and the total monthly salary of each employee in the salary data, calculate the overall salary core degree of each employee.

[0006] A financial data security storage method provided by the present invention determines the salary component core degree corresponding to each salary component value of each employee in the current month, including: Mark the same salary component value of all employees in the current month in the salary data on a two-dimensional plane to obtain a two-dimensional data distribution map; Based on the two-dimensional data distribution map, determine the dispersion degree of each salary component value; Based on each salary component value and the dispersion degree of each salary component value, calculate the salary component core degree corresponding to each salary component value of each employee in the current month.

[0007] A financial data security storage method provided by the present invention determines the core contribution degree corresponding to each salary component value of each employee in the current month, including: Respectively calculate the average value of the same salary component value of all employees in the current month to obtain the average value corresponding to each salary component value; Based on the same salary component value of all employees in the current month and the average value corresponding to each salary component value, calculate the overall change amount of each salary component value in the current month; Calculate the sum of the same salary component value of all employees in the current month to obtain the sum value corresponding to each salary component value; Based on the overall change amount of each salary component value in the current month and the sum value corresponding to each salary component value, calculate the core contribution degree corresponding to each salary component value of each employee in the current month.

[0008] A financial data security storage method provided by the present invention determines the single salary core degree of each employee in the current month based on the salary component core degree and the core contribution degree, including: Sum up the core contribution degrees corresponding to all salary component values of each employee in the current month to obtain the sum value of the salary core contributions of each employee; Divide the core contribution degree corresponding to each salary component value of each employee by the sum value of the core contributions of the salary to obtain the core contribution ratio corresponding to each salary component value of each employee; Multiply the core contribution ratio corresponding to each salary component value of each employee by the core degree of the salary component to obtain the sub - contribution degree corresponding to each salary component value of each employee; Sum up the sub - contribution degrees corresponding to all salary component values of each employee to obtain the single - core degree of the salary of each employee in the current month.

[0009] According to a financial data security storage method provided by the present invention, based on the single - core degree of the salary and the total salary of each employee per month in the salary data, calculate the overall core degree of the salary of each employee, including: Average the total salary of each employee for at least one year including the current month to obtain the average total salary of each employee; Calculate the absolute value of the difference between the total salary of each employee per month for at least one year including the current month and the average total salary to obtain the absolute value of the total salary difference of each employee per month; Sum up the absolute values of the total salary differences of each employee for all months including at least one year including the current month to obtain the sum value of the total salary differences of each employee; Input the sum value of the total salary differences of each employee into the maximum - minimum normalization function to obtain the output value of the total salary function of each employee; Multiply the output value of the total salary function of each employee by the single - core degree of the salary to obtain the overall core degree of the salary of each employee.

[0010] According to a financial data security storage method provided by the present invention, based on the salary data, the overall core degree of the salary, and the employee sorting result, calculate the salary similarity between every two employees per month, including: Based on the employee sorting result, determine the number of intermediate employees between every two employees per month; Based on the number of intermediate employees and the overall core degree of the salary, calculate the similarity correction value between every two employees per month; Based on the salary data, determine the total salary of each employee corresponding to every two employees per month; Take the absolute value of the difference between the total salaries of every two employees per month corresponding to each other to obtain the absolute value of the total salary difference; Based on the similarity correction value and the absolute value of the total salary difference, calculate the salary similarity between every two employees per month.

[0011] A financial data secure storage method provided by the present invention, based on the number of intermediate employees and the overall salary core degree, calculates the similarity correction value for every two employees in each month, including: Adding the number of intermediate employees to a preset hyperparameter to obtain a quantity transition value; Summing up the overall salary core degrees of every two employees in each month to obtain an overall core degree summation value; Multiplying the reciprocal of the quantity transition value by the overall core degree summation value to obtain the similarity correction value for every two employees in each month.

[0012] A financial data secure storage method provided by the present invention, based on the salary similarity, performs clustering processing on the salary data of all employees in each month to obtain a clustering result, including: Converting the salary similarity into a corresponding distance value; Based on the distance value, establishing a distance matrix corresponding to the salary data of all employees in each month; According to the distance matrix, performing clustering processing on the salary data of all employees in each month to obtain a clustering result.

[0013] On the other hand, the present invention also provides a financial data secure storage system, which includes: An acquisition module, configured to acquire the salary data of all employees in at least one year including the current month in the financial data; A first calculation module, configured to respectively determine the salary component core degree and the core contribution degree corresponding to each salary component value of each employee in the current month according to each salary component value in the current month of the salary data; determine the single salary core degree of each employee in the current month according to the salary component core degree and the core contribution degree; calculate the overall salary core degree of each employee according to the single salary core degree and the total monthly salary of each employee in the salary data; A sorting module, configured to sort all employees in the current month according to the total monthly salary in the salary data according to a preset sorting rule to obtain an employee sorting result; A second calculation module, configured to calculate the salary similarity between every two employees in each month based on the salary data, the overall salary core degree, and the employee sorting result; A clustering module, configured to perform clustering processing on the salary data of all employees in each month according to the salary similarity to obtain a clustering result; A storage module, configured to perform block-based secure storage on the monthly salary data in the financial data according to the clustering result.

[0014] The present invention has the following beneficial effects: By calculating the overall salary centrality of each employee, based on the total salary of employees in the salary data, sorting all employees in the current month according to the preset sorting rules to obtain the employee sorting result, calculating the salary similarity between every two employees in each month based on the salary data, the overall salary centrality, and the employee sorting result, then clustering the salary data of all employees in each month according to the salary similarity to obtain the clustering result, and finally storing the salary data of each month in the financial data in a segmented and secure manner. Since the salary similarity on which the clustering process is based is comprehensively calculated based on the salary data, the overall salary centrality, and the employee sorting result, it can reduce the distance between core employees and surrounding data points, increase the density around the salaries of core employees, so that the salary data of core employees is more likely to be identified as core points, thereby forming effective clustering, improving the clustering accuracy, and further enhancing the storage reliability and security of financial data. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a flowchart of a method for securely storing financial data provided by an embodiment of the present invention; Figure 2 It is a system structure diagram of a system for securely storing financial data provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of a method and system for securely storing financial data proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following combines the attached Figure 1 and Figure 2 to specifically illustrate the specific solutions of a method and system for securely storing financial data provided by the present invention.

[0020] Please refer to Figure 1 , which shows a method flowchart of a financial data security storage method provided by an embodiment of the present invention. As Figure 1 shown, the above financial data security storage method includes the following steps: Step 110: Obtain the salary data of all employees in the current month for at least one year in the financial data.

[0021] In this embodiment, the salary data can be obtained from the enterprise's financial system. The salary data usually includes multiple components such as basic salary, transportation subsidy, meal subsidy, housing subsidy, high-temperature subsidy, quarterly bonus, performance bonus, overtime pay, commission, year-end bonus, endowment insurance, medical insurance, and provident fund. At the same time, it also includes the total salary of employees calculated based on the above multiple components.

[0022] In practical applications, the salary data can be presented in the form of a salary breakdown sheet. For example, Table 1 below exemplarily shows a salary breakdown sheet of M employees containing N salary component values.

[0023] Table 1 Salary Breakdown Sheet Step 120: Calculate the overall salary core degree of each employee based on the salary data.

[0024] It can be understood that the overall salary core degree can characterize the salary fluctuation situation of employees and the importance of the salary of each employee in the entire salary data.

[0025] Step 130: Sort all employees in the current month according to the total salary of employees in the salary data according to a preset sorting rule to obtain an employee sorting result.

[0026] In this embodiment, the total salary of employees is calculated based on multiple salary component values in the salary data. The preset sorting rule can be to sort the total salary of employees in descending order, or it can also be to sort the total salary of employees in ascending order.

[0027] Step 140: Calculate the salary similarity between every two employees in each month based on the salary data, the overall salary core degree, and the employee sorting result.

[0028] In this embodiment, the salary similarity can represent the closeness of salaries between two employees. The higher the salary similarity, the closer the salaries of the two employees. In this embodiment, the salary similarity between each pair of employees is comprehensively analyzed using salary data, the overall salary core degree, and the employee ranking results. Since the salary similarity considers multiple factors in the determination process, it can more accurately represent the closeness of salaries between two employees, thereby providing an accurate and reliable data basis for subsequent clustering processing.

[0029] Step 150: According to the salary similarity, perform clustering processing on the salary data of all employees in each month to obtain a clustering result.

[0030] In practical applications, a density-based spatial clustering algorithm can be used to perform clustering processing on the salary data of all employees in each month. In this process, the salary similarity can be converted into a distance value, providing an effective data basis for clustering processing through distance measurement, thereby obtaining a clustering result with higher accuracy.

[0031] Step 160: According to the clustering result, perform block-wise secure storage on the monthly salary data in the financial data.

[0032] Since the salary data is divided into multiple clusters after clustering, that is, the monthly salary data is divided into multiple categories, it can provide an effective data basis for subsequent block-wise secure storage. In this embodiment, by improving the determination scheme of salary similarity, more accurate data support can be provided for data clustering, and thus guarantee the secure storage of financial data.

[0033] In one embodiment, according to the salary data, calculate the overall salary core degree of each employee, specifically including: First, according to each salary component value in the current month of the salary data, respectively determine the salary component core degree and core contribution degree corresponding to each salary component value of each employee in the current month.

[0034] It can be understood that for each salary component value, if some employees have a higher value of a certain salary component, it often means that these employees hold higher positions or have more important responsibilities within the enterprise. Therefore, their influence on the operation direction of the enterprise is greater, that is, the core degree is higher. And on the basis of a larger value, if there is a greater salary gap compared with other employees, it can better reflect the different contributions of employees at different levels to the enterprise. Enterprises usually attach more importance to the roles and work behaviors of core employees with large salary gaps. Therefore, the core degree of the salary data of core employees is relatively high.

[0035] In a specific implementation, to determine the salary component core degree corresponding to each salary component value of each employee in the current month, it specifically includes: In the first step, the same salary component values of all employees in the current month in the salary data are marked on a two-dimensional plane to obtain a two-dimensional data distribution map.

[0036] In this embodiment, specifically, the same salary component values of all employees can be plotted on a two-dimensional plane, where both the horizontal and vertical coordinates in the two-dimensional plane are the same salary component values of the employees.

[0037] In the second step, according to the two-dimensional data distribution map, the dispersion degree of each salary component value is determined.

[0038] In practical applications, the LOF (Local Outlier Factor) algorithm can be used to obtain the dispersion degree of the same salary component values of all employees, so as to obtain the dispersion degree of each salary component value. The LOF algorithm is an algorithm for identifying local outliers in a dataset. This algorithm is based on the concept of density and determines whether a point is an outlier by comparing the density of each data point with that of its neighborhood points. According to the above theory, the dispersion degree of each salary component value can be obtained.

[0039] In the third step, based on each salary component value and the dispersion degree of each salary component value, the salary component core degree corresponding to each salary component value of each employee in the current month is calculated.

[0040] In this embodiment, the salary component core degree corresponding to the k-th salary component value of employee Q in the current month can be expressed as follows: (1) Where represents the salary component core degree corresponding to the k-th salary component value of employee Q, represents the dispersion degree of the k-th salary component value of employee Q, represents the maximum-minimum normalization function. The maximum-minimum normalization function is a commonly used data preprocessing method. It can map the data to a specified interval, usually [0,1], so as to eliminate the influence of the dimension between data features and enable data of different features to be compared on the same scale. represents the k-th salary component value of employee Q.

[0041] In a specific implementation, determining the core contribution degree corresponding to each salary component value of each employee in the current month specifically includes: In the first step, the mean values of the same salary component values of all employees in the current month are calculated respectively to obtain the average value corresponding to each salary component value.

[0042] In the second step, based on the same salary component values of all employees in the current month and the average value corresponding to each salary component value, the overall change amount of each salary component value in the current month is calculated.

[0043] In the third step, calculate the sum of the same salary component values of all employees in the current month to obtain the sum value corresponding to each salary component value.

[0044] In the fourth step, based on the overall change amount of each salary component value in the current month and the sum value corresponding to each salary component value, calculate the core contribution degree corresponding to each salary component value of each employee in the current month.

[0045] It can be understood that for all salary component values, the larger the sum of a certain salary component value of all employees, usually it means that the enterprise needs to pay more labor costs for this part of the salary component. This part of the salary component has a decisive impact on the enterprise's salary strategy, resource allocation, and employee stability. Therefore, the core contribution degree should be higher. And on this basis, if the numerical volatility is large, then it will be directly related to the short-term performance of employees and can better mobilize the enthusiasm of employees, enhance the overall work motivation, and thus promote the realization of enterprise goals. Therefore, the core contribution degree is even higher. Accordingly, the core contribution degree corresponding to the k-th salary component value can be expressed as follows: (2) Where, represents the core contribution degree corresponding to the k-th salary component value, represents the maximum-minimum normalization function, M represents the total number of employees, represents the k-th salary component value of the i-th employee, represents the average value of the k-th salary component values of M employees, represents the overall change amount of the k-th salary component values of M employees, represents the sum value of the k-th salary component values of M employees.

[0046] Then, based on the salary component core degree and the core contribution degree, determine the single salary core degree of each employee in the current month.

[0047] In a specific implementation, based on the salary component core degree and the core contribution degree, determine the single salary core degree of each employee in the current month, which specifically includes: In the first step, sum up the core contribution degrees corresponding to all salary component values of each employee in the current month to obtain the salary core contribution sum value of each employee.

[0048] In the second step, divide the core contribution degree corresponding to each salary component value of each employee by the salary core contribution sum value to obtain the core contribution ratio corresponding to each salary component value of each employee.

[0049] In the third step, multiply the core contribution ratio corresponding to each salary component value of each employee by the salary component core degree to obtain the sub - contribution degree corresponding to each salary component value of each employee.

[0050] In the fourth step, sum up the sub - contribution degrees corresponding to all salary component values of each employee to obtain the salary single - core degree of each employee in the current month.

[0051] In this embodiment, the salary single - core degree of employee Q in the current month can be expressed as follows: (3) Where, represents the salary single - core degree of employee Q in the current month, N represents the total number of items of a salary component value, represents the core contribution degree corresponding to the k - th salary component value, represents the salary component core degree corresponding to the k - th salary component value of employee Q, represents the salary component core degree corresponding to the m - th salary component value, represents the sum value of salary core contributions, and both m and k satisfy the value range of [1, N].

[0052] Finally, based on the salary single - core degree and the total monthly salary of each employee in the salary data, calculate the overall salary core degree of each employee.

[0053] In a specific implementation, based on the salary single - core degree and the total monthly salary of each employee in the salary data, calculating the overall salary core degree of each employee specifically includes: In the first step, calculate the average value of the total monthly salary of each employee including at least one year with the current month to obtain the average total salary of each employee.

[0054] In the second step, calculate the absolute value of the difference between the total monthly salary of each employee including at least one year with the current month and the average total salary to obtain the absolute value of the total salary difference of each employee per month.

[0055] In the third step, sum up the absolute values of the total salary differences of each employee for all months including at least one year with the current month to obtain the sum value of the total salary differences of each employee.

[0056] In the fourth step, input the sum value of the total salary differences of each employee into the maximum - minimum normalization function to obtain the output value of the total salary function of each employee.

[0057] In the fifth step, multiply the output value of the total salary function of each employee by the salary single - core degree to obtain the overall salary core degree of each employee.

[0058] It is understandable that, based on the single-core degree of an employee's salary slip, if there are employees with relatively large overall salary fluctuations within one year including the current month in the past, this should attract the attention of the enterprise. Because large salary fluctuations usually reflect problems in aspects such as performance management, incentive mechanisms, and the financial status of the enterprise, and may also bring the risk of employee turnover. The enterprise should evaluate the reasons behind this kind of fluctuation, ensure the fairness and transparency of the salary system, and strengthen the long-term incentives and stability support for employees to maintain an efficient working atmosphere and employee satisfaction. Therefore, the greater the salary fluctuation, the greater the final overall salary core degree based on the single-core degree of the salary slip. Accordingly, the overall salary core degree of employee Q can be expressed as follows: (4) Wherein, represents the overall salary core degree of employee Q, represents the maximum-minimum normalization function, 12 represents a total of 12 months including the current month before, represents the total salary of employee Q in the v-th month including the current month before, represents the average value of the total salary of employee Q within 12 months including the current month before, represents the single-core degree of employee Q's salary in the current month.

[0059] In one embodiment, based on salary data, the overall salary core degree, and the employee ranking results, calculate the salary similarity between every two employees each month, specifically including: In the first step, according to the employee ranking results, determine the number of intermediate employees between every two employees each month.

[0060] In this embodiment, the employee ranking results contain the ranking information of all employees according to their total salaries. For example, the employee ranking results may include the ranking order of all employees sorted in descending order of their total salaries. For any two employees, the number of employees between these two employees, that is, the number of intermediate employees, can be determined according to the ranking order of all employees in the employee ranking results.

[0061] In the second step, based on the number of intermediate employees and the overall salary core degree, calculate the similarity correction value between every two employees each month.

[0062] In a specific implementation, based on the number of intermediate employees and the overall salary core degree, calculate the similarity correction value between every two employees each month, specifically including: First, add the number of intermediate employees to a preset hyperparameter to obtain a quantity transition value.

[0063] Then, sum up the overall salary core degrees of every two employees each month to obtain the overall core degree summation value.

[0064] Finally, multiply the reciprocal of the quantity transition value by the sum value of the overall core degree to obtain the similarity correction value for every two employees per month.

[0065] It can be understood that since this embodiment needs to adjust the measurement method of salary similarity for the salary data of core employees with relatively low surrounding density and difficult to form clusters, so as to narrow the distance between core employees and surrounding data points, increase the surrounding density of the salaries of these core employees, so that they are more easily identified as core data points, and then form effective clustering.

[0066] Therefore, when correcting the salary similarity between two employees, the higher the degree to which both of them are core, and the lower the density of the surrounding data points, the higher the salary similarity should be further increased. Accordingly, the similarity correction value corresponding to the salary data of employee A and employee B can be expressed as follows: (5) Wherein, represents the similarity correction value corresponding to the salary data of employee A and employee B, represents the number of intermediate employees between employee A and employee B determined according to the employee sorting result, represents a non-zero hyperparameter, represents the overall core degree of employee A's salary, represents the overall core degree of employee B's salary. In this embodiment the value of is 0.01. In specific applications, the implementer can set it according to the specific situation.

[0067] The third step is to determine the total salary of each employee corresponding to every two employees per month based on the salary data.

[0068] The fourth step is to take the difference between the total salaries of each employee corresponding to every two employees per month and take the absolute value to obtain the absolute value of the total salary difference.

[0069] The fifth step is to calculate the salary similarity between every two employees per month based on the similarity correction value and the absolute value of the total salary difference.

[0070] It can be understood that for the salary data of employee A and employee B, the initial similarity is the similarity of the total salary of the employees in the current month. From the above description, the higher the similarity correction value, the higher the initial similarity should be further increased. Accordingly, the salary similarity between employee A and employee B can be expressed as follows: (6) Wherein, represents the salary similarity between employee A and employee B, Represents the similarity correction value of the salary data of Employee A and Employee B. Represents a hyperparameter. Represents the total salary of Employee A. Represents the total salary of Employee B. Represents a non-zero hyperparameter. Represents the initial similarity of the salary data of Employee A and Employee B, and norm( ) represents the normalization function. In this embodiment, The value of is 0.1. The value of is 0.01. In specific applications, the implementer can set it according to specific circumstances.

[0071] In some embodiments, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm can be used to implement the clustering operation. The DBSCAN clustering algorithm can classify data points into core points, boundary points, and noise points, can discover clusters of any shape, and can identify noise points in the dataset.

[0072] In one embodiment, according to the salary similarity, the salary data of all employees in each month is clustered to obtain a clustering result, which specifically includes: First, convert the salary similarity into a corresponding distance value.

[0073] It can be understood that since the DBSCAN clustering algorithm depends on the distance metric, and a high salary similarity means that the salaries of two employees are closer, that is, the distance between two data points is smaller. Therefore, the distance value between the data points corresponding to Employee A and Employee B can be expressed as follows: (7) Where Represents the distance value between the data points corresponding to Employee A and Employee B. Represents the salary similarity between Employee A and Employee B.

[0074] Then, based on the distance value, establish a distance matrix corresponding to the salary data of all employees in each month.

[0075] Finally, according to the distance matrix, cluster the salary data of all employees in each month to obtain a clustering result.

[0076] In practical applications, before the clustering process, the key parameters of the DBSCAN clustering algorithm need to be defined first. The two key parameters of the DBSCAN clustering algorithm include the neighborhood radius and the minimum number of neighbors required for a core point. In this embodiment, the minimum number of neighbors required for a core point is defined as 3, and then the K-Distance graph is used to determine the neighborhood radius.

[0077] Subsequently, the distance matrix can be imported into the DBSCAN clustering algorithm and the clustering process is executed to obtain the clustering result. Then, the clustering result is processed using a hash function, and the monthly salary data in the financial data can be securely stored in blocks.

[0078] In some embodiments, the process of processing the clustering result using a hash function is as follows: First, a hash calculation is performed on each class of salary data in the clustering result using a hash function to generate a hash value of a fixed length.

[0079] Then, each class of salary data in the clustering result is encrypted to obtain the encrypted salary data. In practical applications, the AES (Advanced Encryption Standard) encryption algorithm can be specifically used to encrypt the salary data.

[0080] Finally, the generated hash value and the encrypted salary data are stored in a distributed storage system (such as a distributed file system or a cloud storage system, etc.) to improve the storage security and reliability of the salary data in the financial data.

[0081] Thus, through the hash function, the salary data in the classified financial data can be securely stored in blocks, better ensuring the integrity and security of the salary data of all employees every month in the financial data, and reducing the risk of data leakage or tampering.

[0082] In summary, for the financial data security storage method provided in this embodiment, when directly using the DBSCAN clustering algorithm for clustering, due to the low density around the salaries of core employees and the difficulty in forming clusters, by adjusting the method for measuring salary similarity, the distance between core employees and surrounding data points is reduced, and the density around the salaries of these core employees is increased, so that the salary data of core employees is more easily identified as core points, and then effective clustering is formed, improving the clustering accuracy, and further enhancing the storage security and reliability of financial data.

[0083] Based on the same general inventive concept, the present invention also protects a financial data security storage system. The financial data security storage system provided by the present invention is described below, and the financial data security storage system described below can be mutually corresponding and referred to the financial data security storage method described above.

[0084] Please refer to Figure 2 , which shows the system structure diagram of a financial data security storage system provided by an embodiment of the present invention. As Figure 2 shown, the above system specifically includes: An acquisition module 210, configured to acquire salary data of all employees in at least one year including the current month in financial data.

[0085] A first calculation module 220, configured to respectively determine, according to each salary component value in the current month in the salary data, a salary component core degree and a core contribution degree corresponding to each salary component value of each employee in the current month; determine a single salary core degree of each employee in the current month according to the salary component core degree and the core contribution degree; and calculate an overall salary core degree of each employee according to the single salary core degree and the total monthly salary of each employee in the salary data.

[0086] A sorting module 230, configured to sort all employees in the current month according to the total monthly salary in the salary data according to a preset sorting rule to obtain an employee sorting result.

[0087] A second calculation module 240, configured to calculate a salary similarity between every two employees in each month based on the salary data, the overall salary core degree, and the employee sorting result.

[0088] A clustering module 250, configured to perform clustering processing on the salary data of all employees in each month according to the salary similarity to obtain a clustering result.

[0089] A storage module 260, configured to perform block-based secure storage on the monthly salary data in the financial data according to the clustering result.

[0090] It can be seen that for the financial data secure storage system provided in this embodiment, since the salary similarity on which the clustering processing is based is comprehensively calculated based on the salary data, the overall salary core degree, and the employee sorting result, the distance between core employees and surrounding data points can be reduced, and the density around the salaries of core employees can be increased, so that the salary data of core employees is more likely to be recognized as core points, thereby forming effective clustering, improving the clustering accuracy, and further enhancing the storage reliability and security of financial data.

[0091] Regarding the system in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated herein again.

[0092] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized.

Claims

1. A method for secure storage of financial data, characterized in that: The method comprises: Get the salary data of all employees in the current month within at least one year in the financial data; According to each wage component value of the current month in the wage data, the wage component coreness and core contribution corresponding to each wage component value of each employee in the current month are determined respectively; according to the wage component coreness and the core contribution, the single wage coreness of each employee in the current month is determined; according to the single wage coreness and the monthly total wage of each employee in the wage data, the overall wage coreness of each employee is calculated; According to the total salary of employees in the salary data, all employees in the current month are sorted according to the preset sorting rules to obtain the employee sorting result; Based on the salary data, the overall salary coreness and the employee ranking results, calculate the salary similarity between every two employees in each month; According to the salary similarity, the salary data of all employees in each month are clustered to obtain a clustering result; Based on the clustering results, the monthly salary data in the financial data is divided into blocks and securely stored.

2. A method for secure storage of financial data according to claim 1, characterized in that: Determine the salary component coreness corresponding to each salary component value of each employee in the current month, including: Mark the same wage component value of all employees in the current month in the wage data on a two-dimensional plane to obtain a two-dimensional data distribution graph; Determine the dispersion of each wage component value according to the two-dimensional data distribution graph; Based on each wage component value and the dispersion of each wage component value, the wage component coreness corresponding to each wage component value of each employee in the current month is calculated.

3. A method for securely storing financial data according to claim 1, characterized in that: Determine the core contribution of each employee's salary component value for the current month, including: Calculate the average of the same salary component values ​​of all employees in the current month to obtain the average value corresponding to each salary component value; Based on the same wage component value of all employees in the current month and the average value corresponding to each wage component value, calculate the overall change of each wage component value in the current month; Calculate the sum of the same salary component values ​​of all employees in the current month, and obtain the sum value corresponding to each salary component value; Based on the overall change in the value of each wage component in the current month and the summed value corresponding to each wage component value, the core contribution corresponding to each wage component value of each employee in the current month is calculated.

4. A method for securely storing financial data according to claim 1, characterized in that: Based on the salary component coreness and the core contribution, determine the single salary coreness of each employee in the current month, including: Sum up the core contribution of all salary components of each employee in the current month to get the sum of the core contribution of each employee's salary; The core contribution degree corresponding to each wage component value of each employee is divided by the sum of the core contributions of the wages to obtain the core contribution ratio corresponding to each wage component value of each employee; Multiplying the core contribution ratio corresponding to each wage component value of each employee by the core degree of the wage component to obtain the sub-item contribution degree corresponding to each wage component value of each employee; Sum up the contribution of each employee's individual items corresponding to all salary components to obtain the single core degree of each employee's salary in the current month.

5. A method for securely storing financial data according to claim 1, characterized in that: The overall coreness of the salary of each employee is calculated based on the single coreness of the salary and the total monthly salary of each employee in the salary data, including: Average the total salary of each employee in at least one year including the current month to get the mean total salary of each employee; Calculate the absolute value of the difference between the total monthly salary of each employee in at least one year including the current month and the average of the total salary to obtain the absolute value of the total monthly salary difference of each employee; Sum the absolute values ​​of the total salary differences of each employee in all months within at least one year including the current month to obtain the sum of the total salary differences of each employee; Input the sum of the total salary difference of each employee into the maximum and minimum normalization function to obtain the total salary function output value of each employee; The total salary function output value of each employee is multiplied by the single salary coreness to obtain the overall salary coreness of each employee.

6. A method for securely storing financial data according to claim 1, characterized in that: Based on the salary data, the overall salary coreness and the employee ranking result, the salary similarity between every two employees in each month is calculated, including: According to the employee ranking result, determine the number of intermediate employees between every two employees in each month; Based on the number of intermediate employees and the overall coreness of the salary, a similarity correction value between every two employees in each month is calculated; Based on the salary data, determine the total salary of every two employees in each month; Difference the total wages of every two employees in each month and find the absolute value to get the absolute value of the total wage difference; Based on the similarity correction value and the absolute value of the total salary difference, the salary similarity between every two employees in each month is calculated.

7. A method for securely storing financial data according to claim 6, characterized in that: Based on the number of intermediate employees and the overall coreness of the salary, the similarity correction value between every two employees in each month is calculated, including: Adding the intermediate number of employees to a preset hyperparameter to obtain a quantity transition value; Sum the overall coreness of the wages of every two employees in each month to get the overall coreness sum value; The reciprocal of the quantity transition value is multiplied by the overall coreness sum value to obtain a similarity correction value for every two employees in each month.

8. A method for securely storing financial data according to claim 1, characterized in that: Based on the salary similarity, the salary data of all employees in each month are clustered to obtain clustering results, including: Converting the salary similarity into a corresponding distance value; Based on the distance values, a distance matrix corresponding to the salary data of all employees in each month is established; Based on the distance matrix, the salary data of all employees in each month are clustered to obtain a clustering result.

9. A financial data security storage system, characterized in that: The system comprises: The acquisition module is used to obtain the salary data of all employees in the current month within at least one year from the financial data; The first calculation module is used to determine the salary component coreness and core contribution corresponding to each salary component value of each employee in the current month according to each salary component value of the current month in the salary data; determine the single salary coreness of each employee in the current month according to the salary component coreness and the core contribution; calculate the overall salary coreness of each employee according to the single salary coreness and the monthly total salary of each employee in the salary data; A sorting module is used to sort all employees in the current month according to the total salary of employees in the salary data according to the preset sorting rules to obtain the employee sorting result; A second calculation module is used to calculate the salary similarity between every two employees in each month based on the salary data, the overall salary coreness and the employee ranking result; A clustering module, used to perform clustering processing on the salary data of all employees in each month according to the salary similarity, and obtain a clustering result; The storage module is used to safely store the monthly salary data in the financial data in blocks according to the clustering results.

Images