Payment transaction record management method and system

Abstract

The invention discloses a payment transaction record management method and system, and relates to the technical field of payment transaction management, and the method comprises the steps: collecting payment transaction record data, carrying out the preprocessing, and constructing a data classification model for data classification; carrying out segmented encryption on the classified payment transaction record data and transmitting the payment transaction record data to a central database, and decrypting the encrypted payment transaction record data and storing the encrypted payment transaction record data in the central database; and setting access control authority management storage data, and constructing a visual interface to display a detected data classification result and an access record in real time. According to the method, payment transaction record data is collected and preprocessed, and a data classification model is constructed for data classification; carrying out segmented encryption on the classified payment transaction record data and transmitting the payment transaction record data to a central database, and decrypting the encrypted payment transaction record data and storing the encrypted payment transaction record data in the central database; the security and management efficiency of the data are remarkably improved, and the overall transmission efficiency of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of payment transaction management, and in particular to a payment transaction record management method and system. Background Art

[0002] With the rapid development of e-commerce, payment transactions have gradually shifted from traditional cash transactions to cashless payments. Internet-based payment methods such as mobile payments and online payments are becoming increasingly popular. The complexity and security of payment systems have also increased. The management of payment transaction records not only plays an important role in financial accounting and user experience, but also involves many technical challenges such as data security and privacy protection.

[0003] The existing technologies have the following deficiencies when processing payment transaction data. The existing technologies are often unable to perform hierarchical processing according to the sensitivity of the data, resulting in highly sensitive data and less sensitive data being treated equally, thereby increasing the risk of data leakage. In the process of data encryption and transmission, most of the existing technical solutions use a general encryption algorithm and fail to perform hierarchical encryption according to the sensitivity level of the data, resulting in insufficient security. Summary of the invention

[0004] In view of the problems existing in the above-mentioned existing payment transaction record management method and system, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the existing technology is often unable to perform hierarchical processing according to the sensitivity of the data, resulting in high-sensitivity data and low-sensitivity data being treated equally, thereby increasing the risk of data leakage. In the process of data encryption and transmission, most of the existing technical solutions adopt a general encryption algorithm and fail to perform hierarchical encryption according to the sensitivity level of the data, resulting in insufficient security.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a payment transaction record management method, which includes collecting payment transaction record data and performing preprocessing, building a data classification model to classify the data; encrypting the classified payment transaction record data in segments and transmitting them to a central database, decrypting the encrypted payment transaction record data and storing them in the central database; setting access control permissions to manage stored data, and building a visual interface to display the detected data classification results and access records in real time.

[0007] As a preferred solution of the payment transaction record management method of the present invention, wherein: the collection of payment transaction record data and pre-processing refers to collecting the user's payment transaction record data in real time through an API interface;

[0008] The payment transaction record data includes transaction time, transaction amount, transaction number and user ID;

[0009] Use one-hot encoding to convert user IDs in payment transaction record data into sparse vectors, use periodic encoding to extract transaction time features in payment transaction record data, and use sliding window technology to convert payment transaction record data into time series data;

[0010] The interpolation method is used to fill the missing values ​​of the time series data, the filled time series data is standardized, and the sliding window method is used to segment the standardized time series data.

[0011] As a preferred solution of the payment transaction record management method of the present invention, wherein: the construction of a data classification model for data classification refers to extracting training data from the UCI machine learning library to generate a classification training set;

[0012] Use time integral to analyze the transaction time period of payment transaction record data. The formula is:

[0013] ,

[0014] in is the integral analysis result of the time dimension, and are the upper and lower limits of time integral respectively, n is the total number of payment transaction record data, is the Lagrange multiplier of the i-th training data, is the regularization parameter, is the sensitivity weight of the i-th payment transaction record data, is the RBF kernel function, To pay for the Euclidean distance between transaction record data, is the label of the i-th payment transaction record data, and t is the variable of the time dimension;

[0015] Integral analysis results based on time dimension Construct a data classification model, the formula is:

[0016] ,

[0017] Among them, F Output low sensitivity, medium sensitivity, and high sensitivity labels for the model. sign is the symbol function. is the normalization factor, x is the input payment transaction record data, To dynamically adjust parameters;

[0018] Use the classification training set to train the data classification model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0019] The segmented time series data is input into the trained data classification model to obtain the sensitivity level of each time series data segment, and the segmented payment transaction record data is classified according to the sensitivity level.

[0020] As a preferred solution of the payment transaction record management method of the present invention, the step of encrypting the classified payment transaction record data in segments and transmitting the segmented data to the central database includes:

[0021] Generate a key using a random number generator and store it in a keystore;

[0022] Classify and encrypt data based on the classification results of data sensitivity level;

[0023] Use TLS protocol to transfer encrypted payment transaction record data to the central database;

[0024] The encryption includes using a hash function to hash the highly sensitive payment transaction record data, the formula is:

[0025] ,

[0026] in is the hash value, W is the total dimension of the hash function, The payment transaction record data of the i-th segment with high sensitivity level, is the wth dimension parameter that controls the hash distribution;

[0027] Use an asymmetric encryption algorithm to encrypt the hash value. The formula is:

[0028] ,

[0029] in Encrypting the payment transaction record data for the highly sensitive payment transaction record data, J is the encryption key, k and 0 are the upper and lower limits of the abnormal integral respectively, and p is the mathematical variable of the abnormal integral;

[0030] Encrypting the medium-sensitive level payment transaction record data using a symmetric encryption algorithm to obtain medium-sensitive level payment transaction record data;

[0031] Directly transmit low-sensitivity payment transaction record data.

[0032] As a preferred solution of the payment transaction record management method of the present invention, wherein: the decrypting of the encrypted payment transaction record data and storing it in the central database means that the central database extracts the same key as that used for encryption from the key library;

[0033] The central database decrypts the encrypted payment transaction record data to obtain decrypted data;

[0034] Sort the decrypted data in time series and store them in a central database;

[0035] The decryption includes using an asymmetric encryption algorithm to perform an inverse operation to restore the hash value, and the formula is:

[0036] ,

[0037] Use the same hash function as in the encryption process to reverse the hash value and recover the highly sensitive payment transaction record data;

[0038] Use the same symmetric encryption algorithm as used in the encryption process to perform inverse decryption and recover payment transaction record data of medium sensitivity level;

[0039] Receive payment transaction record data with low sensitivity level.

[0040] As a preferred solution of the payment transaction record management method described in the present invention, wherein: the setting of access control permission management storage data refers to applying access permissions and access passwords to the data after the central database stores the data, defining user roles, and assigning access permission levels to each user role. When the user accesses the data, the central database verifies the user's access permission and access password and monitors the data access process throughout, and verifies the integrity of the access data after the access is completed.

[0041] As a preferred solution of the payment transaction record management method of the present invention, wherein: the construction of a visual interface to display the detected data classification results and access records in real time refers to using React.js as a front-end development framework to construct a visual interface to display access records in real time, using charts to display real-time data classification results, and using ECharts to draw a real-time updated abnormal transaction quantity change trend;

[0042] Add a transaction pop-up window in the visual interface to display payment transaction record data.

[0043] Another object of the present invention is to provide a payment transaction record management system, which comprises:

[0044] The collection and classification module is used to collect payment transaction record data and perform preprocessing, and build a data classification model to classify the data;

[0045] An encryption and decryption module, used to encrypt the classified payment transaction record data in segments and transmit them to the central database, decrypt the encrypted payment transaction record data and store them in the central database;

[0046] The storage visualization module is used to set access control permissions to manage storage data and build a visualization interface to display the detection data classification results and access records in real time.

[0047] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the payment transaction record management method when executing the computer program.

[0048] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the payment transaction record management method.

[0049] The beneficial effects of the present invention are as follows: the present invention collects payment transaction record data and performs preprocessing, constructs a data classification model to classify the data; encrypts the classified payment transaction record data in segments and transmits it to a central database, decrypts the encrypted payment transaction record data and stores it in the central database; significantly improves data security and management efficiency, and improves the overall transmission efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0051] Figure 1 A flowchart of a payment transaction record management method is provided.

[0052] Figure 2 This is a schematic diagram of the structure of the payment transaction record management system. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0056] Example 1, reference Figure 1, which is the first embodiment of the present invention, and provides a payment transaction record management method, the payment transaction record management method comprising:

[0057] S1. Collect payment transaction record data and perform preprocessing, and build a data classification model to classify the data;

[0058] Specifically, collecting payment transaction record data and preprocessing it means collecting the user's payment transaction record data in real time through an API interface;

[0059] The payment transaction record data includes transaction time, transaction amount, transaction number and user ID;

[0060] Use one-hot encoding to convert user IDs in payment transaction record data into sparse vectors, use periodic encoding to extract transaction time features in payment transaction record data, and use sliding window technology to convert payment transaction record data into time series data;

[0061] The interpolation method is used to fill the missing values ​​of the time series data, the filled time series data is standardized, and the sliding window method is used to segment the standardized time series data.

[0062] The real-time collection method can ensure that the system can obtain the latest transaction data in the first place, especially in large-scale online payment scenarios, which can realize real-time monitoring and response to transaction data and avoid the risks caused by data lag. The real-time collected data provides timely and reliable input for subsequent preprocessing, classification and encrypted transmission. The use of periodic coding to extract transaction time features can effectively capture the time pattern in payment transactions, which is particularly important for discovering abnormal transaction behavior or seasonal changes within a specific time period. The user ID is converted into a sparse vector through one-hot encoding, so that each user's identity information is represented by a unique and clear numerical value, which is convenient for system classification and identification. The combined use of methods can improve the system's ability to identify and analyze payment transaction data, especially when facing complex and changeable transaction scenarios, its advantages are particularly obvious. The interpolation method is used to fill in missing values ​​to ensure the continuity and integrity of the data, which not only improves the processing accuracy of time series data, but also lays a solid foundation for subsequent data classification and encrypted transmission. Data segmentation is performed on standardized data, and the continuous time series data is further divided into multiple sub-segments. Each sub-segment can represent a transaction behavior within a period of time. This segmentation processing provides a basis for subsequent sensitivity classification and encryption, enabling the system to flexibly select different encryption and processing strategies according to data characteristics.

[0063] Furthermore, building a data classification model to perform data classification refers to extracting training data from the UCI machine learning library and generating a classification training set;

[0064] The basic classification model is constructed based on the SVM model of the existing RBF kernel. The formula is:

[0065] ,

[0066] in Output labels for the underlying classification model, is the label of the i-th training data, is the Lagrange multiplier of the i-th training data, is the bias term, Total number of payment transaction records;

[0067] RBF kernel function, the formula is:

[0068] ,

[0069] in is the kernel function parameter, and Input the i-th and j-th payment transaction record data;

[0070] Use time integral to analyze the transaction time period of payment transaction record data. The formula is:

[0071] ,

[0072] in is the integral analysis result of the time dimension, and are the upper and lower limits of the time integral, indicating the transaction time period to be analyzed, and n is the total number of payment transaction record data. is the Lagrange multiplier of the i-th training data, is a regularization parameter to prevent the denominator from being 0. is the sensitivity weight of the i-th payment transaction record data, To pay for the Euclidean distance between transaction record data, is the label of the i-th payment transaction record data, and t is the variable of the time dimension;

[0073] The integral method captures the subtle changes in trading behavior over time through cumulative effects, ensuring that the trading behavior of the entire time period can be comprehensively evaluated, rather than just considering the trading behavior at a single moment. Traditional methods, such as static timestamps or simple time weights, cannot reflect the complex time dynamic changes of trading behavior. Simple fixed weight adjustments or timestamp weighted processing of transactions can only time-mark a single trading behavior, and cannot reflect the risk accumulation of multiple trading behaviors in a time interval. The integral method can integrate trading behaviors at different moments into the analysis and provide a global time dynamic evaluation. The integral method has unique advantages in scenarios with nonlinear changes. It can calculate the accumulation of trading risks under complex time patterns, while other methods such as Simple time window division or fixed time characteristics cannot accurately express the nonlinear changes in trading risks. The integral method can flexibly handle different time intervals and analyze the behavior of multiple trading periods. Other methods such as simple time series cannot dynamically adapt to trading changes in different time periods. In traditional classification models, static time tags or simple time weights are usually used, which cannot reflect the cumulative risk effect of transactions in a certain time period. The integral method can accumulate all trading behaviors in a certain time interval, so that the model can better reflect the risk changes in the entire time interval, rather than focusing on a single moment. The integral method can handle complex nonlinear time dynamics and allow trading behaviors to exhibit different risk patterns in different time periods.

[0074] Integral analysis results based on time dimension Construct a data classification model, the formula is:

[0075] ,

[0076] where F Output low sensitivity, medium sensitivity, and high sensitivity labels for the model. sign is the symbol function. is the normalization factor, x is the input payment transaction record data, To dynamically adjust parameters;

[0077] Traditional fixed weights or manually set weighting methods cannot dynamically adapt to changes in each transaction in practical applications, especially in financial payment scenarios, where changes in transaction amounts and frequencies are the norm, and fixed weights cannot capture these changes. If fixed weights are used, the system can only classify based on preset rules and cannot adjust strategies in real time according to transaction characteristics. Other methods such as fixed weights or manual threshold settings lack flexibility and real-time performance and cannot meet the needs of complex transaction scenarios. Real-time adjustment of model weights can more accurately identify large-value, high-risk transactions, greatly improving the flexibility of the model and the accuracy of classification. If normalization is not performed, the model is prone to being overly sensitive to larger transaction amounts and ignoring other times and frequencies. Normalization ensures the stability of the model on multidimensional data and avoids imbalance problems between features. Normalization enables the classification model to handle complex multidimensional data. The payment transaction record data is highly nonlinear, and the linear kernel function is only applicable to simple linear classification problems. The complexity of payment transaction data requires the model to be able to handle nonlinear relationships. The linear kernel function cannot handle nonlinear transaction data and cannot solve the complex pattern classification problems in payment transactions. By introducing the RBF kernel function, the model can capture the nonlinear relationships in transaction data, thereby effectively distinguishing transactions of different sensitivity levels. The risks of payment transactions at night and during the day are different. Dynamic adjustment of the bias term can reflect the impact of time on transaction risks. Fixed bias cannot capture changes in time characteristics. The risk of transaction behavior changes significantly and needs to be handled by dynamically adjusting the bias term. This improves the classification accuracy of the model in different time periods. Compared with the existing technology, this method significantly improves the model's ability to handle time-sensitive transactions.

[0078] Calculate sensitivity weights using an adaptive weight formula based on transaction amount , the formula is:

[0079] ,

[0080] in is the weight adjustment parameter to control the sensitivity of the weight to the difference in transaction amount. is the i-th transaction amount, is the average amount of training transactions;

[0081] Introducing normalization factor , the formula is:

[0082] ,

[0083] in Adjust the parameters for normalization and control the degree of normalization;

[0084] Use the classification training set to train the data classification model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0085] The segmented time series data is input into the trained data classification model to obtain the sensitivity level of each time series data segment, and the segmented payment transaction record data is classified according to the sensitivity level.

[0086] By extracting relevant payment transaction record training data from the UCI machine learning library and generating a classification training set for model training, it is ensured that the data used is widely representative and rich, so that a variety of possible transaction scenarios can be covered during the model training process, and the robustness and accuracy of the model are improved. The UCI dataset provides high-quality data that has been widely verified, reduces the complexity of data processing, and provides a reliable training basis for the model. Time integral analysis can identify these abnormal behaviors and help the system assign a higher sensitivity level to these transaction data. It can not only accurately classify normal transactions and abnormal transactions, but also dynamically adjust the data processing method according to the characteristics of the time period when the transaction occurs, effectively improving the security of the system, and the classification method based on sensitivity weights. The security of the system has been significantly improved, ensuring that highly sensitive transaction data can be more strictly protected, while low-sensitivity data can be processed and transmitted more efficiently. By using the loss function and Adam optimizer to iteratively optimize the parameters, the model can continuously adjust its parameters during the training process to improve the accuracy and stability of classification. As an adaptive optimization algorithm, Adam optimizer can converge quickly and reduce model training time while maintaining a high classification effect. Segmented processing helps to improve the efficiency of data processing. At the same time, different encryption processing is performed on data according to the sensitivity level to ensure the rational use of system resources. The hierarchical processing method not only improves the overall security of the system, but also achieves a balance in resource consumption. It is especially suitable for the processing of large-scale payment transaction data.

[0087] S2, encrypting the classified payment transaction record data in segments and transmitting them to a central database, decrypting the encrypted payment transaction record data and storing them in the central database;

[0088] Specifically, the classified payment transaction record data is encrypted in segments and transmitted to the central database, including:

[0089] Generate a key using a random number generator and store it in a keystore;

[0090] Perform segmented encryption based on data sensitivity level classification results;

[0091] Use TLS protocol to transfer encrypted payment transaction record data to the central database;

[0092] Existing hash functions , the formula is:

[0093] ,

[0094] in is a binary bit string, which is the fixed-length output after processing;

[0095] The encryption includes using a hash function to hash the highly sensitive payment transaction record data, the formula is:

[0096] ,

[0097] in is the hash value, W is the total dimension of the hash function, The payment transaction record data of the i-th segment with high sensitivity level, is the wth dimension parameter that controls the hash distribution;

[0098] The core of the hash function lies in its irreversibility, that is, once the data is processed by the hash function, it cannot be easily restored to the original data through reverse engineering. In the payment transaction system, it is extremely important to ensure the irreversibility of the data. If a simple linear function is used or the exponential function is not introduced, the generated hash value can be easily cracked by attackers through common hash collisions or other methods. The hash function using this formula controls the distribution of hash values ​​by introducing dimensions and parameters, and combines the exponential function to perform weighted summation on the data, so that the hash value of each transaction data has a high uniqueness, avoiding the risk of duplication or collision of hash values. In security-sensitive payment transaction systems, collision attacks are a common threat. If two different transactions generate the same hash value, attackers can use this vulnerability to commit fraud, so that the generated hash value is strongly correlated with the characteristics of each transaction data, ensuring that the characteristics of each transaction data are fully reflected. Traditional hash algorithms are more efficient in processing simple data, but are prone to hash conflicts and data loss when facing complex and high-dimensional data. The formula can ensure that high-dimensional data maintains its uniqueness during mapping, avoids information loss, and thus better supports the encryption needs of multi-dimensional data.

[0099] The existing asymmetric encryption algorithm has the following formula:

[0100] ,

[0101] in Function for public key encryption;

[0102] Use an asymmetric encryption algorithm to encrypt the hash value. The formula is:

[0103] ,

[0104] in Encrypting the payment transaction record data for the highly sensitive payment transaction record data, J is the encryption key, k and 0 are the upper and lower limits of the abnormal integral respectively, and p is the mathematical variable of the abnormal integral;

[0105] Although the existing asymmetric encryption algorithms have high security, their structures are fixed and the encryption process is predictable. By introducing integral operations, the encryption process is not only a fixed function processing, but also a cumulative nonlinear function, which increases the difficulty for attackers to crack through pattern recognition or derivation formulas. The complexity and unpredictability of integrals make the ciphertext generated each time encrypted more random. When traditional encryption algorithms process high-dimensional data, it is often difficult to capture the complex relationship between multidimensional data. Payment transaction data contains multiple dimensions such as transaction time, amount, user ID, etc. The correlation of these data is difficult to be effectively protected in existing technologies. The integral encryption formula combines multidimensional data with hash values ​​to ensure that multidimensional data is integral. The process can be effectively encrypted to prevent attackers from cracking the data by analyzing the relationship between data dimensions. The cumulative effect of the integral method can comprehensively process multi-dimensional data, making the encryption process more secure and stable. Traditional encryption algorithms usually use the same encryption strength for all data and cannot adjust the encryption complexity and resource consumption according to the needs of the scenario. The integral encryption formula realizes dynamic adjustment of encryption strength by adjusting the integral upper limit. The integral encryption formula ensures that the ciphertext generated by each encryption is unique. Even similar input data will generate significantly different encryption results after integral calculation. The integral accumulates subtle changes in the data, ensures the irreversibility of the encryption process, and prevents similar data from generating similar ciphertexts.

[0106] Encrypting the medium-sensitive level payment transaction record data using a symmetric encryption algorithm to obtain medium-sensitive level payment transaction record data;

[0107] Directly transmit low-sensitivity payment transaction record data.

[0108] By combining different encryption methods for high sensitivity, medium sensitivity and low sensitivity, the system can flexibly process data of different levels. For high sensitivity data, a combination of hash hashing and asymmetric encryption is used to ensure that the data is highly secure even during transmission and storage. For medium sensitivity data, a symmetric encryption algorithm is used for encryption to ensure encryption speed and efficiency. For low sensitivity data, the system chooses direct transmission to reduce unnecessary computing burden. The hierarchical encryption processing method not only improves the overall security, but also greatly improves the utilization of system resources, allowing the system to maintain high efficiency when processing large-scale payment transaction data. The irreversibility of the hash function ensures that even if the hash value is leaked, the attacker cannot restore the original data. The public key-private key mechanism of asymmetric encryption allows only the recipient holding the private key to decrypt the data, further enhancing the security of data transmission. The protection of highly sensitive data has reached the highest level, effectively preventing potential risks such as data leakage and tampering. The TLS protocol prevents security threats such as man-in-the-middle attacks and data theft through the encryption mechanism of the transport layer. No matter how many nodes the data passes through in the transmission path, the TLS protocol can ensure the confidentiality and integrity of the data.

[0109] Further, decrypting the encrypted payment transaction record data and storing it in the central database means that the central database extracts the same key as that used for encryption from the key library;

[0110] The central database decrypts the encrypted ciphertext to obtain decrypted data;

[0111] Sort the decrypted data in time series and store them in a central database;

[0112] The existing asymmetric decryption algorithm is as follows:

[0113] ,

[0114] in Function to decrypt the private key;

[0115] The decryption includes using an asymmetric encryption algorithm to perform an inverse operation to restore the hash value, and the formula is:

[0116] ,

[0117] Use the same hash function as in the encryption process to reverse the hash value and recover the highly sensitive payment transaction record data;

[0118] Integral inverse solution is a key step to keep the encryption process reversible. Ordinary linear decryption methods or simple algebraic solutions cannot cope with the complex integral encryption process, especially the encryption process involving nonlinearity and cumulative effects. Only through the integral inverse operation can the decryption process accurately restore the original data. Nonlinearity not only enhances the unpredictability of encryption, but also makes the encryption result dependent on the changes in the entire integral interval. Integral inverse solution gradually eliminates the cumulative changes in the integral process and restores the hash value. Integral inverse solution ensures the high complexity of the decryption process by processing complex integral and nonlinear cumulative processes. Compared with conventional inverse algebraic operations, integral inverse solution can better handle the nonlinear changes introduced in the encryption process and enhance the security and anti-attack ability of the encryption algorithm. Integral encryption introduces highly complex nonlinear changes, making the encrypted data pattern difficult to be recognized by attackers. The decryption process is essentially a complex nonlinear inverse operation. Attackers cannot easily attack by analyzing data patterns, which increases the difficulty of cracking and significantly improves the system's anti-attack ability.

[0119] Use the same hash function as in the encryption process to reverse the hash value and recover the highly sensitive payment transaction record data;

[0120] Use the same symmetric encryption algorithm as used in the encryption process to perform inverse decryption and recover payment transaction record data of medium sensitivity level;

[0121] Receive payment transaction record data with low sensitivity level.

[0122] The key library not only ensures the secure storage of keys, but also prevents keys from being exposed in the system or maliciously acquired. Through this key management mechanism, the present invention can effectively prevent the risk of encryption key leakage and improve the security of the entire system. Through this key management mechanism, the present invention can effectively prevent the risk of encryption key leakage and improve the security of the entire system. Storing data in time series can enable the system to have higher query efficiency when processing historical data and improve overall processing performance. The double encryption mechanism greatly improves the security of highly sensitive data, ensuring that even if the data is intercepted or tampered with during transmission, attackers cannot easily obtain or modify the data, so that the system can effectively protect the confidentiality and integrity of the data when facing high security requirements. Since the security of the symmetric encryption algorithm is sufficient to meet the risk requirements of medium-sensitivity data, this method optimizes the use of system resources while ensuring data security, and selects appropriate encryption and processing strategies according to the sensitivity of the data to balance security and processing efficiency.

[0123] S3, set access control permissions to manage storage data, and build a visual interface to display the detected data classification results and access records in real time;

[0124] Specifically, setting access control permissions to manage stored data means that after the central database stores the data, it applies access permissions and access passwords to the data, defines user roles, and assigns access permission levels to each user role. When a user accesses the data, the central database verifies the user's access permissions and access passwords and monitors the entire data access process, and verifies the integrity of the accessed data after the access is completed.

[0125] Through this division of permissions, the system can minimize the exposure of data and ensure that only necessary users can access sensitive information. The hierarchical management mechanism improves the security and flexibility of the system, which is particularly important when processing a large number of payment transaction records involving personal privacy or financial information. The double verification mechanism ensures that only authorized users can enter the system to access data, preventing possible identity theft or malicious attacks. This mechanism can significantly improve the security protection level of data. The monitoring mechanism not only helps to ensure the legality of data access, but also provides a complete access record for subsequent audits and abnormal behavior analysis. For payment transaction record data, any tampering or damage may have a significant impact on financial results. Therefore, data integrity verification provides additional security to ensure the authenticity and accuracy of transaction data. Security measures at different levels complement each other to ensure that data can still be effectively protected even in the most complex security environment. The multi-level security system is not only suitable for the processing of payment transaction record data, but can also be extended to the management of other sensitive data, such as medical data, financial data and other high-security scenarios.

[0126] Further, building a visual interface to display the detected data classification results and access records in real time means using React.js as a front-end development framework to build a visual interface to display the access records in real time, using charts to display the real-time data classification results, and using ECharts to draw a real-time updated abnormal transaction quantity change trend;

[0127] Add a transaction pop-up window in the visual interface to display payment transaction record data.

[0128] Through the componentized design of React.js, the system can modularly process complex information such as data classification results and access records. Each component can be updated independently, ensuring the real-time and responsiveness of the interface. When the payment transaction data is updated, React.js can only re-render the changed part without redrawing the entire interface, thereby improving system performance. This technical solution greatly improves the user experience and makes complex payment data processing and display efficient and intuitive. By using ECharts to draw a trend chart of the number of abnormal transactions, the present invention can track and display abnormal transactions occurring in the system in real time. When an abnormal transaction occurs, ECharts The chart can automatically update and dynamically display the number and time distribution of abnormal transactions. The display of real-time charts can help managers quickly identify potential security threats or abnormal transaction behaviors, and take corresponding countermeasures in time, which improves the flexibility and depth of data analysis. The transaction pop-up box greatly simplifies the operation process, which not only improves the user's operation efficiency, but also reduces the trouble caused by frequent interface switching. By displaying the classified data results in the form of charts in real time, users can intuitively understand the distribution of each type of payment transaction data. The chart display method can not only help users quickly obtain a global data overview, but also facilitate users to identify and analyze the changing trends of transaction data through real-time updates of classification results. The display module for user access records is embedded in the visual interface, which can track the access behavior of each user in the system in real time. The real-time display function of access records can effectively prevent unauthorized access and provide reliable basic data support for the system's security audit.

[0129] Example 2, reference Figure 2 , which is the second embodiment of the present invention, and which is different from the previous embodiment, provides a payment transaction record management method system, including:

[0130] The collection and classification module is used to collect payment transaction record data and perform preprocessing, and build a data classification model to classify the data;

[0131] An encryption and decryption module, used to encrypt the classified payment transaction record data in segments and transmit them to the central database, decrypt the encrypted payment transaction record data and store them in the central database;

[0132] The storage visualization module is used to set access control permissions to manage storage data and build a visualization interface to display the detection data classification results and access records in real time.

[0133] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0135] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0136] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A payment transaction record management method, characterized in that: include, Collect payment transaction record data and perform preprocessing, and build a data classification model to classify the data; The classified payment transaction record data is encrypted in segments and transmitted to a central database, and the encrypted payment transaction record data is decrypted and stored in the central database; Set access control permissions to manage stored data and build a visual interface to display the detected data classification results and access records in real time.

2. The payment transaction record management method according to claim 1, characterized in that: The collection of payment transaction record data and pre-processing refers to collecting the user's payment transaction record data in real time through the API interface; The payment transaction record data includes transaction time, transaction amount, transaction number and user ID; Use one-hot encoding to convert user IDs in payment transaction record data into sparse vectors, use periodic encoding to extract transaction time features in payment transaction record data, and use sliding window technology to convert payment transaction record data into time series data; The interpolation method is used to fill the missing values ​​of the time series data, the filled time series data is standardized, and the sliding window method is used to segment the standardized time series data.

3. The payment transaction record management method according to claim 2, characterized in that: The constructing of a data classification model to perform data classification refers to extracting training data from the UCI machine learning library to generate a classification training set; Use time integral to analyze the transaction time period of payment transaction record data. The formula is: ， in is the integral analysis result of the time dimension, and are the upper and lower limits of time integral respectively, n is the total number of payment transaction record data, is the Lagrange multiplier of the i-th training data, is the regularization parameter, is the sensitivity weight of the i-th payment transaction record data, is the RBF kernel function, To pay for the Euclidean distance between transaction record data, is the label of the i-th payment transaction record data, and t is the variable of the time dimension; Integral analysis results based on time dimension Construct a data classification model, the formula is: ， Among them, F Output low sensitivity, medium sensitivity, and high sensitivity labels for the model. sign is the symbol function. is the normalization factor, x is the input payment transaction record data, To dynamically adjust parameters; Use the classification training set to train the data classification model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The segmented time series data is input into the trained data classification model to obtain the sensitivity level of each time series data segment, and the segmented payment transaction record data is classified according to the sensitivity level.

4. The payment transaction record management method according to claim 3, characterized in that: The encrypting the classified payment transaction record data in segments and transmitting them to the central database includes generating a key using a random number generator and storing the key in a key library; Classify and encrypt data based on the classification results of data sensitivity level; Use TLS protocol to transfer encrypted payment transaction record data to the central database; The encryption includes using a hash function to hash the highly sensitive payment transaction record data, the formula is: ， in is the hash value, W is the total dimension of the hash function, The payment transaction record data of the i-th segment with high sensitivity level, is the wth dimension parameter that controls the hash distribution; Use an asymmetric encryption algorithm to encrypt the hash value. The formula is: ， in Encrypting the payment transaction record data for the highly sensitive payment transaction record data, J is the encryption key, k and 0 are the upper and lower limits of the abnormal integral respectively, and p is the mathematical variable of the abnormal integral; Encrypting the medium-sensitive level payment transaction record data using a symmetric encryption algorithm to obtain medium-sensitive level payment transaction record data; Directly transmit low-sensitivity payment transaction record data.

5. The payment transaction record management method according to claim 4, characterized in that: Decrypting the encrypted payment transaction record data and storing it in the central database means that the central database extracts the same key as that used for encryption from the key library; The central database decrypts the encrypted payment transaction record data to obtain decrypted data; Sort the decrypted data in time series and store them in a central database; The decryption includes using an asymmetric encryption algorithm to perform an inverse operation to restore the hash value, and the formula is: ， Use the same hash function as in the encryption process to reverse the hash value and recover the highly sensitive payment transaction record data; Use the same symmetric encryption algorithm as used in the encryption process to perform inverse decryption and recover payment transaction record data of medium sensitivity level; Receive payment transaction record data with low sensitivity level.

6. The payment transaction record management method according to claim 5, characterized in that: The setting of access control rights to manage stored data refers to applying access rights and access passwords to the data after the central database stores the data, defining user roles, and assigning access rights levels to each user role. When the user accesses the data, the central database verifies the user's access rights and access passwords and monitors the entire data access process, and verifies the integrity of the accessed data after the access is completed.

7. The payment transaction record management method according to claim 6, characterized in that: The said construction of a visual interface to display the detected data classification results and access records in real time refers to using React.js as a front-end development framework to construct a visual interface to display access records in real time, using charts to display real-time data classification results, and using ECharts to draw a real-time updated abnormal transaction quantity change trend; Add a transaction pop-up window in the visual interface to display payment transaction record data.

8. A payment transaction record management system according to any one of claims 1 to 7, characterized in that: include, The collection and classification module is used to collect payment transaction record data and perform preprocessing, and build a data classification model to classify the data; An encryption and decryption module, used to encrypt the classified payment transaction record data in segments and transmit them to the central database, decrypt the encrypted payment transaction record data and store them in the central database; The storage visualization module is used to set access control permissions to manage storage data and build a visualization interface to display the detection data classification results and access records in real time.

9. A computer device comprising: A memory and a processor; the memory stores a computer program, wherein the processor implements the steps of the payment transaction record management method described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the payment transaction record management method described in any one of claims 1 to 7 are implemented.

Images