Business data management method and system based on data analysis

Abstract

The invention relates to the field of business management, in particular to a business data management method and system based on data analysis. The method comprises the following steps: collecting business data in a business database, and dividing each field in a customer data table of each customer; calculating an information chaos coefficient, and calculating a uniqueness ratio; obtaining a sensitivity original score; constructing a field sensitivity score based on the sensitivity original score, calculating a field correlation factor, and obtaining a field coupling coefficient; dividing the encryption strength grade of each field of each user on the basis of the field coupling coefficient, obtaining the key length and the update round number of each field of each customer in the encryption process by combining the field coupling coefficient, encrypting the service data of each field on the basis of the key length and the update round number, and managing the service data; and the refinement, intelligence and security level of business data management is improved.

Description

Technical Field

[0001] The present application relates to the field of business management, and in particular to a business data management method and system based on data analysis. Background Art

[0002] With the development of information technology and the digital economy, the amount of business data generated by daily enterprise operations has exploded. Data types encompass customer identity information, transaction records, financial data, consumer behavior, and other dimensions. This data is not only a core resource for internal decision-making and business optimization, but also carries customer privacy and corporate trade secrets. Its security and reliability directly impact a company's sustainable development and social trust. However, business data is vulnerable to various risks during storage, transmission, and sharing. On the one hand, external attack methods are becoming increasingly sophisticated, such as frequency analysis, feature association inference, dictionary attacks, and side-channel attacks, all of which can lead to the leakage of sensitive data. On the other hand, internal risks are equally important. Internal abuse of privileges, improper operation, and even malicious leaks can undermine data security. A data breach can not only cause direct financial losses to an enterprise, but also lead to serious consequences such as regulatory penalties and damage to brand reputation.

[0003] Existing business data encryption and management methods rely heavily on traditional symmetric or asymmetric encryption mechanisms. In practice, a uniform key length and fixed number of encryption rounds are commonly used to encrypt all data fields, such as the AES algorithm. This situation hinders differentiated protection for fields of varying sensitivity. For highly sensitive or highly coupled fields, fixed-parameter encryption often proves insufficient to protect against targeted inference and frequency attacks, posing a security risk. For less sensitive fields, employing overly strong encryption strategies wastes computing resources and increases system response latency, impacting overall business efficiency. Summary of the Invention

[0004] In order to solve the problem that traditional business data encryption management methods are difficult to strike a balance between security and efficiency, the present application provides a business data management method and system based on data analysis.

[0005] In a first aspect, the present application provides a business data management method based on data analysis, which adopts the following technical solutions: Business data from the business database is collected and divided into various fields in each customer's customer data table. The information confusion coefficient is calculated based on the frequency of occurrence of business data in the field, and the uniqueness ratio is calculated by analyzing the proportion of each field. The sum of the absolute value of the difference between the uniqueness ratio and the preset value and the field label value is used as the original sensitivity score of each field, reflecting the original sensitivity of the business data in the field. The field sensitivity score is constructed based on the original sensitivity score. The correlation between any two fields of each customer is analyzed to calculate the field correlation factor, and the field coupling coefficient is obtained by combining the field sensitivity score. Based on the field coupling coefficient, the encryption strength level of each field of each user is divided. Combined with the field coupling coefficient, the key length and update round number of each field of each customer during the encryption process are obtained. The update round number is the sum of the floor value of the product of the preset maximum increase round number and the field coupling coefficient and the preset basic round number. Based on the key length and update round number, the business data of each field is encrypted and the business data is managed.

[0006] Beneficial Effects: By gradually developing an analytical approach based on the frequency distribution, uniqueness ratio, raw sensitivity score, sensitivity score, field correlation factor, and coupling coefficient of business data, the key length and number of rounds in the encryption parameters can be adaptively adjusted based on the sensitivity and coupling of different fields, thus avoiding the resource waste and security deficiencies of the "one-size-fits-all" approach of traditional AES encryption. This solution not only enhances data security and better protects against attacks involving frequency analysis and field correlation inference, but also reduces computational overhead on less sensitive data, improving overall system efficiency.

[0007] Furthermore, the method for obtaining the information confusion coefficient is: For each field of each customer, the frequency of the business data in each field appearing in the corresponding fields of all customers is counted as the business frequency of each field. Based on the business frequency of each field, the Shannon entropy of each field of each customer is calculated as the information confusion coefficient of each field of each customer.

[0008] Beneficial Effects: Shannon entropy characterizes the uncertainty of field value distribution, quantitatively reflecting the complexity of field information. Fields with uniform frequency distribution have a higher information chaos coefficient, making them more difficult to crack using statistical methods after encryption, thus enhancing defense against attacks.

[0009] Furthermore, the uniqueness ratio is obtained by taking the total number of customers as the total number of business data records, and taking the set of business data of all customers in each field as the field value set of each field; For each field, the ratio between the total number of elements in the field value set of each field and the total number of business data records is calculated as the uniqueness ratio of each field.

[0010] Benefits: The uniqueness ratio measures the distinguishability of field values. A ratio close to 1 indicates strong identifiability, such as an ID number field. A low ratio indicates a field at risk of frequency attacks. By incorporating the uniqueness ratio into the sensitivity scoring system, encryption strategies can address both strong identifiability and weak distinction risks, achieving more precise protection.

[0011] Furthermore, the tag value of the field includes: when the customer's field is a sensitive field, the tag value of the customer's field is assigned a value of 1; when the customer's field is a non-sensitive field, the tag value of the customer's field is assigned a value of 0.

[0012] Furthermore, the calculation formula for the field sensitivity score is: Where, Indicates the field sensitivity score of each field for each customer; is the normalization function, is the preset parameter adjustment factor; is the information confusion coefficient of each field, The original sensitivity score of each field for each customer.

[0013] Furthermore, the method for obtaining the field correlation factor is: for each customer, select any two fields, and calculate the probability of the two fields appearing in all fields respectively, as the independent probabilities of the two fields; calculate the probability of the two fields appearing simultaneously in all fields as the joint probability of the two fields; for the any two fields, calculate the product of the independent probabilities of the two fields as the first product, and take the ratio between the joint probability of the two fields and the first product as the field correlation factor between the two fields of each customer.

[0014] Benefits: Field correlation factors reveal the potential dependencies between two fields through probabilistic relationships, enabling the detection of combinatorial sensitivity issues that are difficult to detect with a single field. For example, the correlation between birthday fields and ID numbers can lead to potential data leaks. This approach effectively prevents inference attacks caused by insufficient protection of a single field.

[0015] Furthermore, the method for obtaining the field coupling coefficient is: calculating the absolute value of the difference between the field correlation factor and the numerical value 1 as the first difference absolute value, normalizing all the first difference absolute values, and using them as the field correlation coefficient; for each field of each customer, calculating the product of the field correlation coefficient between the field and any other fields and the field sensitivity score of the other arbitrary fields as the second product between each field and any other fields, calculating the sum of the second products between the field and all other arbitrary fields, and the ratio of the sum of the field sensitivity scores of all other arbitrary fields as the field coupling coefficient of each field of each customer.

[0016] Furthermore, the encryption strength level of each field of each user is divided based on the field coupling coefficient, including: dividing the interval [0,1] into three intervals evenly, and defining them as the first interval, the second interval and the third interval from small to large; when the field coupling coefficient of the customer's field is in the first interval, the encryption strength level of the field is defined as low strength; when the field coupling coefficient of the customer's field is in the second interval, the encryption strength level of the field is defined as medium strength; when the field coupling coefficient of the customer's field is in the third interval, the encryption strength level of the field is defined as high strength.

[0017] Beneficial Effects: The field coupling coefficient combines the sensitivity of a field and its interactivity with other fields. By setting encryption strength in a tiered manner, it avoids under-encryption of high-risk fields and over-encryption of low-risk fields. This ensures strong protection for highly sensitive data while reducing system performance overhead, achieving differentiated and refined data management.

[0018] Furthermore, the key length of each field of each customer during the encryption process is obtained, including: when the encryption strength level of the customer's field is low, medium and high, the corresponding key lengths are 128, 192 and 256.

[0019] In a second aspect, the present application provides a business data management system based on data analysis, which adopts the following technical solutions: A business data management system based on data analysis includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the business data management method based on data analysis is implemented.

[0020] The above-mentioned business data management method based on data analysis is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0021] This application has the following technical effects: This application addresses the shortcomings of traditional encryption algorithms that use only fixed key lengths and round numbers for global, unified encryption by introducing multi-dimensional sensitivity and coupling quantitative indicators into the business data management process. A quantitative mechanism based on the information chaos coefficient and uniqueness ratio is proposed, which can accurately characterize the complexity and distinguishability of data distribution in each field, avoiding the one-sidedness of judging sensitivity based solely on field labels, thereby enabling fine-grained identification of potentially high-risk fields in the early stages of data management. Field sensitivity scores and field correlation factors are constructed, which not only reflect the sensitivity of the field itself but also consider cross-field correlations, capturing combinatorial inference risks that a single field cannot expose. For example, cross-field inference issues such as ID card numbers and dates of birth can be effectively identified through this mechanism, thereby expanding the coverage of encryption protection. The introduction of a field coupling coefficient and the tiered control of encryption strength based on its value enable differentiated allocation of encryption resources: high-risk fields are protected with longer keys and more rounds, while low-risk fields use lightweight encryption strategies to avoid unnecessary performance consumption. Compared with the traditional one-size-fits-all AES encryption method, this application effectively balances security and computational efficiency.

[0022] During the encryption execution phase, the AES key length and update rounds are dynamically adjusted by combining field coupling coefficients with preset parameters, enabling the encryption strength of each field to adapt to changes in data sensitivity and relevance. This dynamic adjustment mechanism significantly enhances the system's resilience to frequency analysis attacks, correlation attacks, and inference attacks, while ensuring that overall operational efficiency is not compromised by excessive encryption. This solves the difficulty of balancing security and efficiency in traditional business data encryption management methods, enhancing the sophistication, intelligence, and security of business data management. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a method flow chart of a business data management method based on data analysis in this application. DETAILED DESCRIPTION

[0024] An embodiment of the present application discloses a business data management method based on data analysis, which collects business data from a business database and divides the fields in the customer data table of each customer; calculates the information confusion coefficient and the uniqueness ratio; obtains the original sensitivity score; constructs the field sensitivity score based on the original sensitivity score, calculates the field correlation factor, and obtains the field coupling coefficient; divides the encryption strength level of each field of each user based on the field coupling coefficient, obtains the key length and the number of update rounds of each field of each customer during the encryption process in combination with the field coupling coefficient, encrypts the business data of each field based on the key length and the number of update rounds, and manages the business data; thereby improving the refinement, intelligence, and security level of business data management.

[0025] Reference Figure 1 , a business data management method based on data analysis includes steps S1 to S4.

[0026] Step S1: Collect business data from the business database and divide the fields in the customer data table of each customer.

[0027] Collect the data summarized in the business database and pre-process the data in the business database; in one embodiment of the present application, clean the data in the business database to process missing values, duplicate data, and data that violates data definition and business logic. The data cleaning method is a well-known technology and will not be described in detail in this application; at this point, the pre-processed business database is obtained.

[0028] For all business data, build a customer data table for all customers. For example, for each customer, the customer number, name, gender, ID number, telephone number, address, and purchase category are used as the customer data table for each customer. The implementer can select other business data as the customer data table for each customer based on actual conditions. Each of the above categories is a field in the customer's business data. For example, if the customer's name is A and the number is 001, then A is a field and 001 is a field.

[0029] So far, the fields of the customer data table have been divided for each customer.

[0030] Step S2: Calculate the information confusion coefficient based on the frequency of occurrence of business data in the field, analyze the proportion of each field to calculate the uniqueness ratio; take the sum of the absolute value of the difference between the uniqueness ratio and the preset value and the label value of the field as the original sensitivity score of each field, reflecting the original sensitivity of the business data in the field; construct a field sensitivity score based on the original sensitivity score, analyze the correlation between any two fields of each customer to calculate the field correlation factor, and obtain the field coupling coefficient in combination with the field sensitivity score.

[0031] To manage the data quality of business data, due to the confidential nature of the business database, encryption algorithms need to be used to encrypt business data to ensure the security of business data. In the process of encrypting business data, the traditional AES encryption algorithm can encrypt and manage business data, and prevent business data from being leaked to a certain extent. However, due to the strong correlation between business data and the different degrees of susceptibility to cracking of business data itself, data can still be inferred after encryption.

[0032] Therefore, based on the above analysis, the total number of customers is taken as the total number of business data records, and the set consisting of the business data of all customers in each field is taken as the field value set of each field; it should be noted that all elements in the field value set are unique. For example, taking the gender field of all customers as an example, the field value set of the gender field only includes {male, female}, and only one copy of the repeated value is retained.

[0033] For each field of each customer, the frequency of the business data in each field appearing in the corresponding fields of all customers is counted as the business frequency of each field. Based on the business frequency of each field, the Shannon entropy of each field of each customer is calculated as the information confusion coefficient of each field of each customer; among which, the calculation method of Shannon entropy is a well-known technology and will not be elaborated in this application.

[0034] Furthermore, for each field, the ratio between the total number of elements in the field value set of each field and the total number of business data records is calculated as the uniqueness ratio of each field; it should be noted that when the value of the uniqueness ratio is closer to the value 1, it means that in this field, the business data of all customers are different, such as the ID card number. At this time, the identifiability of the field is higher and the risk is also higher; on the contrary, when the value of the uniqueness ratio is less than the value 1, it means that the discrimination of the field is lower, it is easy to be frequency analyzed, and the risk is higher; therefore, the closer the uniqueness ratio is to the middle value, the stronger the analysis resistance and the higher the security.

[0035] At the same time, each field of each customer is marked with a label value. When the customer's field is a sensitive field, the label value of the customer's field is assigned to 1; otherwise, the customer's field is assigned to 0. In this application, the customer number and ID number in the customer data table are regarded as sensitive fields.

[0036] Based on the above analysis, the original sensitivity score of each field is constructed, and the calculation formula is: Where, The original sensitivity score of each field for each customer; is the uniqueness ratio of each field, The label value of each field.

[0037] It should be noted that the original sensitivity score of each field reflects the sensitivity of the field to being cracked. When the customer's field is a sensitive field, The value of is 1, and The closer to the value 1 or the smaller it is, the greater the risk of being cracked. The larger the value of , the larger the value of the original sensitivity score obtained; conversely, the smaller the sensitivity of the field that is easy to be cracked is, the smaller the value of the original sensitivity score obtained is.

[0038] Furthermore, in order to analyze the business sensitivity of each field for each customer, the degree of confusion of the information contained in the field content, and the identifiability, a field sensitivity score for each field for each customer is constructed. The calculation formula is: Where, Indicates the field sensitivity score of each field for each customer; is the normalization function, To preset the parameter adjustment factor and prevent the denominator from being 0, in one embodiment of the present application, The value of is 0.1, and implementers can choose other values ​​based on actual conditions; is the information confusion coefficient of each field, The original sensitivity score of each field for each customer.

[0039] It should be noted that the field sensitivity score comprehensively reflects the sensitivity of the field based on the amount of information in the business data in the field and the sensitivity and uniqueness of the field. When the amount of information in the business data in the field is large, the value of the obtained information confusion coefficient is larger, and the difficulty of cracking after encryption increases. At the same time, when the value of the original sensitivity score is smaller, the possibility that the customer's field is a sensitive field is smaller, and the value of the obtained field sensitivity score is smaller; conversely, the value of the obtained field sensitivity score is larger.

[0040] Furthermore, in business data, each field of each customer does not exist in isolation. There is often a strong or weak correlation between the fields. For example, there is a correlation between the customer's ID number field and the customer's date of birth field. The 6th to 14th digits of the ID number contain birthday information. Based on this analysis, even if a field itself does not appear sensitive, once combined with other fields, sensitive information can be inferred. The sensitivity of a single field can only reflect the risk of the single field itself and cannot capture the risk of cross-field inference.

[0041] Therefore, a field correlation factor between any two fields is constructed based on the field sensitivity score. Specifically, for each customer, any two fields are selected, and the probabilities of the two fields appearing in all fields are calculated respectively, which are used as the independent probabilities of the two fields; the probability of the two fields appearing simultaneously in all fields is calculated as the joint probability of the two fields; for the any two fields, the product of the independent probabilities of the two fields is calculated as the first product, and the ratio between the joint probability of the two fields and the first product is used as the field correlation factor between the two fields for each customer.

[0042] It should be noted that, when the field correlation factor is greater than the value 1, it means that the two fields appear together more often than the two fields appear independently, and the two fields show a more positive correlation; on the contrary, when the field correlation factor is less than the value 1, it means that the two fields appear together less often than the two fields appear independently, and the two fields show a more negative correlation; that is, the more the value of the field correlation factor deviates from the value 1, the stronger the correlation between the two fields; conversely, the weaker the correlation between the two fields.

[0043] The field correlation factor can identify fields that have high entropy but are highly coupled with other sensitive fields. This can avoid focusing only on surface sensitive fields and ignoring the problem of combined sensitive fields, thereby improving the coverage of security protection.

[0044] Furthermore, the absolute value of the difference between the field correlation factor and the value 1 is calculated as the first absolute value of the difference, and all the first absolute values ​​of the difference are normalized to form the field correlation coefficient. For each field of each customer, the product of the field correlation coefficient between the field and any other fields and the field sensitivity score of the other arbitrary fields is calculated as the second product between each field and any other fields, and the ratio of the sum of the second products between the field and all other arbitrary fields to the sum of the field sensitivity scores of all other arbitrary fields is calculated as the field coupling coefficient of each field of each customer.

[0045] It should be noted that the field coupling coefficient combines the sensitivity of the field itself and the high degree of coupling between the customer's fields. When the field coupling coefficient is high, it means that the field is more sensitive and the degree of coupling with other fields is also higher, and the probability of being cracked after encryption is greater; conversely, when the field coupling coefficient is low, the probability of being cracked after encryption is smaller.

[0046] Step S3: Divide the encryption strength level of each field of each user based on the field coupling coefficient.

[0047] At this point, the field coupling coefficient of each field of each customer is obtained, and the encryption strength level of each field of each customer is constructed based on the field coupling coefficient. Since the value range of the field coupling coefficient is [0,1], [0,1] is evenly divided into three intervals, which are defined as the first interval, the second interval and the third interval from small to large. When the field coupling coefficient of the customer's field is in the first interval, the encryption strength level of the field is defined as low strength; when the field coupling coefficient of the customer's field is in the second interval, the encryption strength level of the field is defined as medium strength; when the field coupling coefficient of the customer's field is in the third interval, the encryption strength level of the field is defined as high strength.

[0048] Step S4: Combined with the field coupling coefficient, the key length and update round number of each field of each customer during the encryption process are obtained. The update round number is the sum of the floor value of the product of the preset maximum increase round number and the field coupling coefficient and the preset basic round number. The business data of each field is encrypted based on the key length and update round number, and the business data is managed.

[0049] When using AES to encrypt customer fields, the parameters involved include key length and number of rounds. Traditional applications usually use the same AES parameters for all fields, such as AES-128 and a fixed number of rounds. This has insufficient security costs when facing highly sensitive or highly coupled fields, and causes waste of resources for low-sensitivity fields. Therefore, this application drives the hierarchical adjustment of AES core parameters based on the encryption strength level, and performs on-demand encryption and hierarchical reinforcement.

[0050] Specifically, in this application, when the encryption strength levels of the customer's fields are low, medium, and high, respectively, during the encryption process using AES, the corresponding key lengths are selected as 128, 192, and 256, and the number of rounds is calculated based on the field coupling coefficient. The calculation formula is: Where, The number of rounds updated when using AES encryption for each field of each customer; is the preset basic round number corresponding to the AES key length. In one embodiment of the present application, when the key length corresponding to the customer's field is 128, the preset basic round number is 10; when the key length corresponding to the customer's field is 192, the preset basic round number is 12; when the key length corresponding to the customer's field is 256, the preset basic round number is 14. To preset the maximum number of increase rounds, in one embodiment of the present application, The value of is 8, is the field coupling coefficient of each field for each customer; Represents a floor function; for the preset values ​​in the formula, implementers can select other values ​​based on actual conditions.

[0051] It should be noted that the number of update rounds can be based on the difficulty of cracking the business data. Different update rounds can be automatically obtained for business data in different fields. For business data that is easy to crack, more update rounds are obtained; conversely, for business data that is more difficult to crack, fewer rounds are used for encryption.

[0052] So far, in the process of using the AES algorithm to encrypt each field of each customer, different key lengths and round numbers are adopted. When the customer's field is easy to crack, it is necessary to increase the key length and number of rounds to strengthen the encryption effect of the business data in the field, prevent business data leakage, and enhance the management effect of business data; when the customer's field is difficult to crack, choose a shorter key length and reduce the number of rounds, so as to avoid performance waste caused by blindly strengthening business data equally.

[0053] An embodiment of the present application also discloses a business data management system based on data analysis, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a business data management method based on data analysis according to the present application is implemented.

[0054] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0055] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A business data management method based on data analysis, characterized in that: The method comprises the steps of: collecting business data from a business database and dividing each field in a customer data table of each customer; The information confusion coefficient is calculated based on the frequency of occurrence of business data in the field, and the uniqueness ratio is calculated by analyzing the proportion of each field. The sum of the absolute value of the difference between the uniqueness ratio and the preset value and the field label value is used as the original sensitivity score of each field, reflecting the original sensitivity of the business data in the field. Build a field sensitivity score based on the original sensitivity score, analyze the correlation between any two fields of each customer, calculate the field correlation factor, and combine the field sensitivity score to obtain the field coupling coefficient; Based on the field coupling coefficient, the encryption strength level of each field of each user is divided. Combined with the field coupling coefficient, the key length and update round number of each field of each customer during the encryption process are obtained. The update round number is the sum of the floor value of the product of the preset maximum increase round number and the field coupling coefficient and the preset basic round number. Based on the key length and update round number, the business data of each field is encrypted and the business data is managed.

2. A business data management method based on data analysis according to claim 1, characterized in that: The method for obtaining the information confusion coefficient is: For each field of each customer, the frequency of the business data in each field appearing in the corresponding fields of all customers is counted as the business frequency of each field. Based on the business frequency of each field, the Shannon entropy of each field of each customer is calculated as the information confusion coefficient of each field of each customer.

3. A business data management method based on data analysis according to claim 1, characterized in that: The method for obtaining the uniqueness ratio is as follows: taking the total number of customers as the total number of business data records, and taking the set of business data of all customers in each field as the field value set of each field; For each field, the ratio between the total number of elements in the field value set of each field and the total number of business data records is calculated as the uniqueness ratio of each field.

4. The business data management method based on data analysis according to claim 1, characterized in that: The tag value of the field includes: when the customer's field is a sensitive field, the tag value of the customer's field is assigned a value of 1; when the customer's field is a non-sensitive field, the tag value of the customer's field is assigned a value of 0.

5. The business data management method based on data analysis according to claim 1, characterized in that: The calculation formula for the field sensitivity score is: Where, Indicates the field sensitivity score of each field for each customer; is the normalization function, is the preset parameter adjustment factor; is the information confusion coefficient of each field, The original sensitivity score of each field for each customer.

6. A business data management method based on data analysis according to claim 1, characterized in that: The method for obtaining the field correlation factor is as follows: for each customer, select any two fields, calculate the probability of the two fields appearing in all fields respectively, and use them as the independent probabilities of the two fields respectively; calculate the probability of the two fields appearing simultaneously in all fields as the joint probability of the two fields; for the any two fields, calculate the product of the independent probabilities of the two fields as the first product, and use the ratio between the joint probability of the two fields and the first product as the field correlation factor between the two fields of each customer.

7. A business data management method based on data analysis according to claim 1, characterized in that: The method for obtaining the field coupling coefficient is as follows: calculating the absolute value of the difference between the field correlation factor and the value 1 as the first absolute value of the difference, and normalizing all the first absolute values ​​of the difference to obtain the field correlation coefficient; For each field of each customer, the product of the field correlation coefficient between the calculated field and any other fields and the field sensitivity score of the other arbitrary fields is used as the second product between each field and any other fields. The sum of the second products between the calculated field and all other arbitrary fields and the ratio of the sum of the field sensitivity scores of all other arbitrary fields are used as the field coupling coefficient of each field of each customer.

8. The business data management method based on data analysis according to claim 1, characterized in that: The method of dividing the encryption strength level of each field of each user based on the field coupling coefficient includes: dividing the interval [0, 1] into three intervals evenly, which are defined as the first interval, the second interval, and the third interval from small to large, respectively; when the field coupling coefficient of the customer's field is in the first interval, the encryption strength level of the field is defined as low strength; when the field coupling coefficient of the customer's field is in the second interval, the encryption strength level of the field is defined as medium strength; when the field coupling coefficient of the customer's field is in the third interval, the encryption strength level of the field is defined as high strength.

9. The business data management method based on data analysis according to claim 1, characterized in that: Obtain the key length for each field of each customer during the encryption process, including: when the encryption strength level of the customer's field is low, medium, and high, the corresponding key lengths are 128, 192, and 256.

10. A business data management system based on data analysis, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a business data management method based on data analysis according to any one of claims 1 to 9 is implemented.

Images