Mobile payment information management system based on aggregation transaction

Abstract

The invention relates to the technical field of transaction management, and discloses a mobile payment information management system based on aggregated transaction, which comprises the steps of determining a fund settlement chain according to fund settlement information, judging whether an aggregated payment platform is compliant based on business permission information and the fund settlement chain, generating a violation number according to static region information, and sending the violation number to a server. And generating sequence similarity according to the dynamic region information, inputting the violation number and the sequence similarity into a pre-constructed risk classification model to obtain a payment risk result, completing fund settlement when the payment risk result is low risk, and interrupting fund settlement when the payment risk result is high risk. According to the invention, through a multi-level transaction information management mode, specific requirements of an aggregation payment platform are effectively met, and through carrying out layer-by-layer analysis and management on transaction information of multiple dimensions, the system can monitor and judge the risk level of payment operation in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction management, and more specifically, to a mobile payment information management system based on aggregated transactions. Background Art

[0002] Aggregate payment relies on the payment interface of banks or non-bank payment institutions and utilizes the technical integration capabilities of the aggregate payment service provider itself to integrate the payment methods of multiple banks and non-bank payment institutions into a unified payment portal, providing merchants with one-stop payment collection services. Compared with traditional third-party payment services, aggregate payment has obvious advantages and characteristics, such as low thresholds and convenient payment, but there are also certain challenges in terms of security supervision. Due to the late rise of aggregate payment, the current regulatory technology is mainly aimed at third-party payment, and the supervision of aggregate payment is still insufficient.

[0003] For example, the Chinese patent with publication number CN117911031A provides a risk control system for online transaction payments, which introduces risk identification links in the whole process before, during and after the transaction, and verifies whether the user information, payment environment and other information before and after the transaction have changed, thereby verifying whether the transaction links are initiated by real users throughout the process and whether the transaction is credible. The Chinese patent with publication number CN107895269A provides an electronic payment control method and an application server. The provided electronic payment control method and application server enable the third-party payment platform to automatically and efficiently assess customer risks during a large number of electronic payment transactions.

[0004] Although the above patents provide supervision for online transactions, they are mainly aimed at traditional third-party payment links. However, aggregated payment is significantly different from third-party payment. Since aggregated payment emerged relatively late, existing regulatory technologies have not been fully developed to cope with its characteristics and lack multi-level transaction information management methods. In addition, there are few effective management solutions in existing technologies that meet the unique needs of aggregated payment.

[0005] In view of this, the present invention proposes a mobile payment information management system based on aggregated transactions to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a mobile payment information management system based on aggregated transactions.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A mobile payment information management system based on aggregated transactions, comprising:

[0009] Compliance detection module: used to obtain the business license information and fund settlement information of the aggregate payment platform, determine the fund settlement chain according to the fund settlement information, and judge whether the aggregate payment platform is compliant based on the business license information and fund settlement chain;

[0010] Risk monitoring module: used for obtaining static area information and dynamic area information of the aggregated payment page when the aggregated payment platform is compliant, generating the number of violations according to the static area information, generating sequence similarity according to the dynamic area information, inputting the number of violations and sequence similarity into a pre-built risk classification model, and obtaining payment risk results, which include at least low risk and high risk. The aggregated payment page is a payment page in the aggregated payment platform;

[0011] Transaction settlement module: used to complete fund settlement when the payment risk result is low risk, and interrupt fund settlement when the payment risk result is high risk.

[0012] Furthermore, the method for determining the funds settlement chain according to the funds settlement information includes:

[0013] Each element in the payment channel information is taken as a node, and the edges between the nodes are constructed according to the fund flow information, so as to construct a fund flow graph, and the fund flow graph is used as the fund settlement chain.

[0014] Furthermore, methods for determining whether the aggregate payment platform is compliant include:

[0015] Based on the business license information, determine whether the aggregate payment platform has a payment business license. If not, the aggregate payment platform is non-compliant. If so, further determine whether the funds settlement chain complies with the standard settlement process. If so, the aggregate payment platform is compliant. If not, the aggregate payment platform is non-compliant.

[0016] Further, the method for generating the number of violations according to the static area information includes:

[0017] Each element in the static area information is checked for integrity, compliance, and standardization in turn, and the inspection results are generated. The number of violations is obtained based on the inspection results. The integrity check refers to checking whether all necessary elements exist, the compliance check refers to comparing whether the attributes of each element meet the specifications, and the standardization check refers to confirming whether the arrangement of page elements meets the specifications.

[0018] Furthermore, the dynamic region information includes a payment channel sequence and a payment channel type, and the method for generating sequence similarity based on the dynamic region information includes:

[0019] Compare the dynamic area information with the preset standard dynamic area, calculate the first similarity between the payment channel order and the standard channel order, calculate the second similarity between the payment channel type and the standard channel type, and take the sum of the first similarity and the second similarity as the sequence similarity. The standard channel order is the arrangement order of different payment channels in the standard dynamic area on the page, and the standard channel type is the payment channel type in the standard dynamic area.

[0020] Furthermore, the construction method of the standard dynamic region includes:

[0021] The number of channel clicks and payment failures is obtained, a dynamic area template is pre-built according to the number of channel clicks, and the dynamic area template is optimized according to the number of payment failures and a natural inspiration algorithm to obtain a standard dynamic area.

[0022] Furthermore, the method for constructing a dynamic region template includes:

[0023] According to the number of channel clicks corresponding to each payment channel, each payment channel is arranged in descending order, and the channel width corresponding to each payment channel is generated in an arithmetic sequence.

[0024] Furthermore, the risk monitoring module includes an initialization unit, a function construction unit, an update unit, a calculation unit, a judgment unit, and a loop output unit. The method for optimizing the dynamic area template includes:

[0025] Initialization unit: used to randomly select a dynamic area template as the initial dynamic area EP initial , preset the initial temperature T0 and cooling rate α of simulated annealing, 1>α>0;

[0026] Function building unit: used to convert the initial dynamic area EP initial The number of payment failures f(EP initial ) as the objective function value;

[0027] Update unit: used to update the current dynamic area EP current Add random disturbances on the basis of to generate new dynamic area EP new , the current dynamic area EP current The starting point is the initial dynamic area EP initial ;

[0028] Calculation unit: used to calculate the new dynamic area EP new The corresponding number of new payment failures f(EP new ), the number of new payment failures f(EP new ) and the current payment failure number f(EP current ) is compared with the current payment failure number f(EPcurrent ) and the current dynamic area EP current correspond;

[0029] A judgment unit is used to judge whether to update the dynamic area, set T = α × T, and return to the update unit, where the starting point of T is T0;

[0030] Loop output unit: used to repeat the above update unit-judgment unit until T<T min , stop iteration and output the optimal dynamic area EP best , the optimal dynamic area EP best As the standard dynamic region, T is the temperature parameter, T min The preset minimum temperature value.

[0031] Further, a new dynamic area EP is generated new The methods include:

[0032] EP new =EP current +Δ EP ;

[0033] Among them, Δ EP is a random disturbance;

[0034] Methods for determining whether to update a dynamic area include:

[0035] If f(EP new )<f(EP current ), then let EP current =EP new ;

[0036] If f(EP new )≥f(EP current ), then randomly generate a probability threshold r between 0 and 1, and determine whether the probability K is greater than the probability threshold r. If so, let EP current =EP new , if not, keep EP current unchanged, the probability K is:

[0037]

[0038] Where exp(·) is an exponential function with e as the base, and e is a natural constant.

[0039] Furthermore, the method for constructing the risk classification model includes:

[0040] Obtain Q sets of training data, where Q is a positive integer greater than 1. The training data includes the number of historical violations, historical sequence similarity, and historical payment risk results.

[0041] The number of historical violations, historical sequence similarity, and historical payment risk results are used as sample sets, and the sample sets are divided into training sets and test sets to build a classifier;

[0042] The number of historical violations and historical sequence similarity in the training set are used as input data, and the historical payment risk results in the training set are used as output data to train the classifier and obtain an initial classifier;

[0043] The initial classifier is tested using the test set, and the classifier that meets the preset accuracy is output as the risk classification model.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention first determines the funds settlement chain according to the funds settlement information, determines whether the aggregate payment platform is compliant based on the business license information and the funds settlement chain, then generates the number of violations according to the static area information, generates the sequence similarity according to the dynamic area information, inputs the number of violations and the sequence similarity into the pre-built risk classification model, obtains the payment risk result, and completes the funds settlement when the payment risk result is low risk, and interrupts the funds settlement when the payment risk result is high risk. In this way, the present invention effectively responds to the unique needs of the aggregate payment platform through a multi-level transaction information management method, and analyzes and manages transaction information in multiple dimensions, such as fund flow, compliance, static and dynamic page information, etc. layer by layer, so that the system can monitor and judge the risk level of payment operations in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a structural schematic diagram of a mobile payment information management system based on aggregated transactions in the present invention;

[0047] Figure 2 It is a flow chart of a mobile payment information management method based on aggregated transactions in the present invention;

[0048] Figure 3 A schematic diagram of the illegal fund settlement mode carried out by the aggregate payment platform in the present invention;

[0049] Figure 4 It is a schematic diagram of the capital flow diagram in the present invention;

[0050] Figure 5 A schematic diagram of an aggregate payment page layout in the present invention;

[0051] Figure 6 A schematic diagram of a dynamic area layout in the present invention;

[0052] Figure 7 is a schematic diagram of a dynamic region template in the present invention;

[0053] Reference numerals:

[0054] 11. Fund input area; 12. Payment channel selection area; 13. Payment confirmation area; 20. Amount input box; 30. Payment channel title; 301. First payment channel; 302. Second payment channel; 303. Third payment channel; 40. Payment confirmation box; PW2, second channel width; PW3, first channel width. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example 1

[0057] See also Figure 1 As shown, this embodiment discloses a mobile payment information management system based on aggregated transactions, including:

[0058] Compliance detection module: used to obtain the business license information and fund settlement information of the aggregate payment platform, determine the fund settlement chain according to the fund settlement information, and judge whether the aggregate payment platform is compliant based on the business license information and fund settlement chain;

[0059] In this embodiment, the aggregate payment platform refers to a platform that relies on the payment interface of a bank or non-bank payment institution, integrates the payment methods of multiple payment institutions into a unified payment portal through the technical integration capabilities of the aggregate payment service provider, and provides merchants with a one-stop collection service, such as an existing payment platform. The business license information refers to the payment business license used to prove the qualification of a third-party payment. The payment business license is also called a payment license. It is a qualification certificate for non-financial institutions to conduct payment business. The business license information can be obtained by accessing the official website of the aggregate payment platform through the Internet and inquiring from the national or regional financial regulatory agency. This embodiment does not go into too much detail on this. The fund settlement information refers to information related to fund flow and transactions. The fund settlement information includes but is not limited to payment channel information and fund flow information. The payment channel information refers to the information of each institution or user involved in the fund settlement, and the fund flow information refers to the flow path of funds in the payment system.

[0060] Methods for determining the funds settlement chain based on funds settlement information include:

[0061] Each element in the payment channel information is taken as a node, and the edges between the nodes are constructed according to the fund flow information, so as to construct a fund flow graph, and the fund flow graph is used as the fund settlement chain.

[0062] It can be understood from the above that each element in the payment channel information refers to the information of each institution or user involved in the fund settlement, such as Figure 3 As shown, Figure 3 The illegal fund settlement model carried out by the aggregate payment platform is shown. The illegal fund settlement model refers to consumers first paying funds to the platform account, and then the unlicensed institution settles the funds from the platform account to the merchant account based on the order information. Figure 3 In the payment channel information, the elements are consumer, platform account and merchant account. The fund flow information shows that the funds are first transferred from the consumer to the platform account, and then from the platform account to the merchant account. Therefore, the constructed fund flow diagram is as follows: Figure 4 shown.

[0063] It should be added that the aggregation payment platform only acts as an intermediary for the flow of funds, providing technical support and services, but cannot directly participate in the final settlement of funds. The aggregation payment platform only serves as a channel for the transmission of information and data, ensuring that the flow of funds from consumers to merchants is compliant and secure.

[0064] Methods for determining whether an aggregate payment platform is compliant include:

[0065] Based on the business license information, determine whether the aggregate payment platform has a payment business license. If not, the aggregate payment platform is non-compliant. If so, further determine whether the funds settlement chain complies with the standard settlement process. If so, the aggregate payment platform is compliant. If not, the aggregate payment platform is non-compliant.

[0066] It should be noted that the standard settlement process refers to the established steps and specifications followed by the flow of funds from consumers to merchants within the framework of legal payment services. For example, Figure 3 As shown, the funds are first transferred from the consumer to the platform account, and then from the platform account to the merchant account. This does not comply with the standard settlement process, which requires that funds must flow directly from the consumer to the merchant account.

[0067] This embodiment can systematically ensure that the aggregated payment platform complies with regulatory requirements throughout the entire operation process by obtaining business license information and fund settlement information in sequence. First, the business license is confirmed, and then the flow of funds is analyzed, which helps to form an effective compliance review system and reduce legal risks. The payment channel information and fund flow information are combined in sequence to help build a clear fund flow diagram, which can clearly understand the flow path of funds, enhance the transparency of fund flows, and enable all parties involved (such as consumers, merchants and regulators) to monitor and confirm the flow of funds in real time, reducing the risk of funds being abused or illegally transferred.

[0068] Risk monitoring module: used for obtaining static area information and dynamic area information of the aggregated payment page when the aggregated payment platform is compliant, generating the number of violations according to the static area information, generating sequence similarity according to the dynamic area information, inputting the number of violations and sequence similarity into a pre-built risk classification model, and obtaining payment risk results, which include at least low risk and high risk. The aggregated payment page is a payment page in the aggregated payment platform;

[0069] In this embodiment, the aggregated payment page is a payment page in the aggregated payment platform. The static area refers to those areas in the aggregated payment page that remain unchanged during the user operation process. The content of these areas will not change with the user's operation during the payment process. The static area information can be specific content, layout constraints or display elements, etc. The dynamic area refers to the area in the aggregated payment page that will change according to user operations or system status changes. The content of these areas will be dynamically updated during the user interaction process. The dynamic area information can be the payment channel order and payment channel type, etc.

[0070] For example, Figure 5 As shown, Figure 5 It shows the layout of an aggregate payment page. Figure 5 The funds input area 11, the payment channel selection area 12 and the payment confirmation area 13 are shown. The funds input area 11 and the payment confirmation area 13 are both static areas, so the static area information may include the field format of the funds input area, the location of the payment confirmation button, the page title, prompt information, etc. The payment channel selection area 12 is a dynamic area, so the dynamic area information may include the arrangement order of different payment channels in the payment channel selection area 12 and the feedback information dynamically updated after the user selects the payment channel, such as the confirmation box after selection, the loading status prompt, etc.

[0071] Methods for generating violation counts based on static region information include:

[0072] Each element in the static area information is checked for integrity, compliance, and standardization in turn, and the inspection results are generated. The number of violations is obtained based on the inspection results. The integrity check refers to checking whether all necessary elements exist, the compliance check refers to comparing whether the attributes of each element meet the specifications, and the standardization check refers to confirming whether the arrangement of page elements meets the specifications.

[0073] According to the above content, static area information can include the field format of the fund input area, the location of the payment confirmation button, the page title, prompt information, etc. Then the integrity check can be to check whether there is an amount input box. If the amount input box is missing, it is recorded as a violation. Check whether the page contains a title, such as "Online Payment" or "Payment Confirmation". If the title is missing, it is recorded as a violation. The compliance check can be to check whether the text of the button meets the specifications, for example, it should be "Confirm Payment". If the text of the button is "Submit" or other non-compliant text, it is recorded as a violation. The standardization check can be to check whether the position of the amount input box is in the predetermined position of the page, such as whether it is in the top center of the page or in a designated area. If the position is incorrect, it is recorded as a violation.

[0074] The method for generating sequence similarity based on dynamic region information includes:

[0075] Compare the dynamic area information with the preset standard dynamic area, calculate the first similarity between the payment channel order and the standard channel order, calculate the second similarity between the payment channel type and the standard channel type, and take the sum of the first similarity and the second similarity as the sequence similarity. The standard channel order is the arrangement order of different payment channels in the standard dynamic area on the page, and the standard channel type is the payment channel type in the standard dynamic area.

[0076] Among them, the standard dynamic area refers to a pre-set, compliant dynamic area layout. As a reference model for evaluating the current payment page, it defines the ideal state of the dynamic area in the aggregated payment page, including the order of payment channels and the types of available payment channels, such as Figure 6 As shown, Figure 6 An amount input box 20, a payment confirmation box 40, a payment channel title 30, a first payment channel 301, a second payment channel 302 and a third payment channel 303 are shown. Figure 6 A method for constructing a standard dynamic area is shown. The types of payment channels may be Alipay and WeChat Pay, for example, the first payment channel 301 is WeChat Pay, and the second payment channel 302 is Alipay. The method for calculating the similarity may be a method for calculating the Euclidean distance or the cosine value. Calculating the Euclidean distance or the cosine value is a prior art, and this embodiment will not go into details.

[0077] The construction methods of standard dynamic regions include:

[0078] The number of channel clicks and payment failures is obtained, a dynamic area template is pre-built according to the number of channel clicks, and the dynamic area template is optimized according to the number of payment failures and a natural inspiration algorithm to obtain a standard dynamic area.

[0079] In this embodiment, the number of channel clicks refers to the total number of times the consumer clicks on each payment channel during use, and the number of payment failures refers to the number of times the consumer clicks on the wrong payment channel, resulting in payment failures. The number of channel clicks represents the consumer's preference for and frequency of selection of different payment channels, and the number of payment failures represents the degree of non-compliance of the current dynamic area. The degree of non-compliance means that the layout of the current dynamic area no longer conforms to the consumer's usage habits.

[0080] In this embodiment, by analyzing the number of channel clicks, it is possible to understand consumers' preferences for different payment channels, build a payment channel layout that conforms to consumer habits, and improve the user experience. The number of payment failures reflects the deviation between the current dynamic area layout and user expectations. Through optimization based on the number of payment failures, it is possible to effectively reduce consumer misselection or system errors, reduce the risk of payment failure, thereby increasing the rate of successful payments and ensuring a smooth payment experience. Intelligent optimization based on user behavior data (number of clicks, number of failures) can continuously adjust the dynamic area layout so that it can match user behavior at different times and situations.

[0081] Methods for building dynamic region templates include:

[0082] According to the number of channel clicks corresponding to each payment channel, each payment channel is arranged in descending order, and the channel width corresponding to each payment channel is generated in an arithmetic sequence.

[0083] In this embodiment, the payment channel with the most clicks will be arranged at the front, and the ones with fewer clicks will be arranged in the back, and the width of the payment channel will increase or decrease in sequence according to certain rules. An arithmetic sequence means that the difference between two adjacent elements is the same, so the width of the channel changes regularly. The effect of this construction is that the channel width of the more commonly used payment channels is larger, which is convenient for users to identify and click, and the channel width of the less used payment channels is smaller, ensuring the rationality and aesthetics of the page layout.

[0084] For example, Figure 7 As shown, Figure 7The first payment channel 301, the second payment channel 302 and the third payment channel 303 are shown in the figure. The third payment channel 303 has the most channel clicks, while the second payment channel 302 has the least channel clicks, and the first channel width PW3 is greater than the second channel width PW2.

[0085] The risk monitoring module includes an initialization unit, a function construction unit, an update unit, a calculation unit, a judgment unit, and a loop output unit. The method for optimizing the dynamic area template includes:

[0086] Initialization unit: used to randomly select a dynamic area template as the initial dynamic area EP initial , preset the initial temperature T0 and cooling rate α of simulated annealing, 1>α>0;

[0087] Among them, the initialization process ensures that the algorithm has enough exploration space at the beginning to avoid local optimal solutions. In subsequent iterations, as the temperature decreases, the system gradually reduces the acceptance rate of poor solutions and tends to the optimal solution.

[0088] Function building unit: used to convert the initial dynamic area EP initial The number of payment failures f(EP initial ) as the objective function value;

[0089] This embodiment takes the number of payment failures as the objective function and defines the direction of optimization, namely, reducing the number of payment failures to improve the success rate of the payment process and user experience.

[0090] Update unit: used to update the current dynamic area EP current Add random disturbances on the basis of to generate new dynamic area EP new , the current dynamic area EP current The starting point is the initial dynamic area EP initial ;

[0091] Calculation unit: used to calculate the new dynamic area EP new The corresponding number of new payment failures f(EP new ), the number of new payment failures f(EP new ) and the current payment failure number f(EP current ) is compared with the current payment failure number f(EP current ) and the current dynamic area EP current correspond;

[0092] Among them, by comparing the objective function values ​​of the new and old solutions, we can determine whether the new solution is better, and then decide whether to accept the new solution. The advantage of doing this is to ensure that each step moves in the direction of minimizing the objective function, so that the optimization process gradually approaches the optimal solution.

[0093] A judgment unit is used to judge whether to update the dynamic area, set T = α × T, and return to the update unit, where the starting point of T is T0;

[0094] This embodiment ensures that the algorithm gradually converges to the optimal solution while exploring the solution space through a gradual cooling process, thereby balancing the capabilities of global exploration and local development.

[0095] Loop output unit: used to repeat the above update unit-judgment unit until T<T min , stop iteration and output the optimal dynamic area EP best , the optimal dynamic area EP best As the standard dynamic region, T is the temperature parameter, T min The preset minimum temperature value.

[0096] Generate new dynamic area EP new The methods include:

[0097] EP new =EP current +Δ EP ;

[0098] Among them, Δ EP is a random disturbance.

[0099] Methods for determining whether to update a dynamic area include:

[0100] If f(EP new )<f(EP current ), then let EP current =EP new ;

[0101] If f(EP new )≥f(EP current ), then randomly generate a probability threshold r between 0 and 1, and determine whether the probability K is greater than the probability threshold r. If so, let EP current =EP new , if not, keep EP current unchanged, the probability K is:

[0102]

[0103] Where exp(·) is an exponential function with e as the base, and e is a natural constant.

[0104] It should be added that, by adding random perturbations, the algorithm in this embodiment can jump out of the current state in the solution space and look for possible better solutions, providing diversified solutions, making the optimization process more comprehensive and robust, and further avoiding the algorithm from falling into local optimality by introducing a probabilistic judgment mechanism, thereby increasing the chance of finding the global optimal solution.

[0105] This embodiment ensures the consistency and effectiveness of the entire optimization process from initial setting to final convergence through the above steps. By combining gradual adjustment (disturbance), local and global balance (annealing algorithm), probability mechanism (avoiding falling into local optimum) and gradual temperature reduction, the system can effectively reduce the number of payment failures and output the optimal payment channel layout that meets user needs. This optimization process improves user experience, enhances the robustness of the system, and ensures a smooth and reliable payment process.

[0106] The methods for constructing risk classification models include:

[0107] Obtain Q sets of training data, where Q is a positive integer greater than 1. The training data includes the number of historical violations, historical sequence similarity, and historical payment risk results.

[0108] The number of historical violations, historical sequence similarity, and historical payment risk results are used as sample sets, and the sample sets are divided into training sets and test sets to build a classifier;

[0109] The number of historical violations and historical sequence similarity in the training set are used as input data, and the historical payment risk results in the training set are used as output data to train the classifier and obtain an initial classifier;

[0110] The initial classifier is tested using the test set, and a classifier that meets the preset accuracy is output as the risk classification model. The classifier is preferably a naive Bayes model or a support vector machine model.

[0111] In this embodiment, the greater the number of violations, the greater the payment risk, and the greater the sequence similarity, the smaller the payment risk. This embodiment constructs a risk classification model (based on the number of violations and sequence similarity) combined with the number of payment failures, so that the entire optimization process is not only to reduce the number of payment failures, but also combines potential payment risk factors to ensure that the payment page design reduces the overall payment risk while meeting user needs.

[0112] It should be noted that the number of violations and sequence similarity are used as input data to build a payment risk classification model. From the above content, it can be seen that the number of violations refers to the compliance check results of the static area in the aggregated payment page, which is mainly obtained through the integrity, compliance and standardization checks of the page elements. The larger the number of violations, the more violations or non-standard elements there are on the page, and the system regards this as a potential risk factor. Therefore, the larger the number of violations, the more serious the deviation of the payment page from the standard specifications, and the greater the probability that the payment risk classification model will output a high risk.

[0113] Similarly, sequence similarity is calculated by comparing the dynamic area (such as the order and type of payment channels) with the standard template. Changes in dynamic area information should follow a certain logic and order in the legal payment process. Therefore, sequence similarity can reflect whether the dynamic area layout of the payment page meets the standards. The lower the sequence similarity, the greater the difference between the dynamic layout of the payment page and the standard, indicating that the page may have risks of abnormal operation, logical errors or potential fraudulent intentions, and the greater the probability that the payment risk classification model outputs high risk.

[0114] The principle of the risk classification model is to use a machine learning classification algorithm to map the input number of violations and sequence similarity to payment risk results (such as high risk or low risk). Taking the naive Bayes model as an example, the core principle of the model is based on Bayes' theorem and the "conditional independence assumption" to achieve efficient data classification. The naive Bayes model is a probability-based classification method. The key lies in the assumption of conditional independence between features, that is, under a given risk level (high risk or low risk), the number of violations and sequence similarity are independent of each other. Through this assumption, the model can simplify calculations and quickly generate classification results. The naive Bayes model as a classification algorithm belongs to the prior art and is not elaborated in this embodiment.

[0115] Transaction settlement module: used to complete fund settlement when the payment risk result is low risk, and interrupt fund settlement when the payment risk result is high risk.

[0116] It is understandable that when the payment risk result is low risk, it means that after analyzing the number of violations and sequence similarity in the static and dynamic areas of the payment page, it is determined that the risk of the payment operation is low. This means that the information in the payment process meets the standards and user expectations, the number of violations is small, and the changes in the dynamic area also meet the established specifications. Therefore, when the risk assessment result is low risk, the system considers the payment behavior to be safe and legal, and funds can be settled normally. If the assessment is high risk, there may be security risks or non-compliant operations. The system will interrupt the payment to prevent potential problems in funds settlement and protect the funds of both parties to the transaction.

[0117] This embodiment first determines the funds settlement chain based on the funds settlement information, determines whether the aggregate payment platform is compliant based on the business license information and the funds settlement chain, then generates the number of violations based on the static area information, generates the sequence similarity based on the dynamic area information, and inputs the number of violations and the sequence similarity into the pre-built risk classification model to obtain the payment risk result. When the payment risk result is low risk, the funds settlement is completed, and when the payment risk result is high risk, the funds settlement is interrupted. In this way, the unique needs of the aggregate payment platform are effectively responded to through a multi-level transaction information management method. By analyzing and managing transaction information in multiple dimensions, such as fund flow, compliance, static and dynamic page information, etc., the system can monitor and judge the risk level of payment operations in real time.

[0118] Example 2

[0119] See also Figure 2 As shown, this embodiment provides a mobile payment information management method based on aggregated transactions on the basis of Embodiment 1, including:

[0120] S10: Obtaining the business license information and fund settlement information of the aggregate payment platform, determining the fund settlement chain according to the fund settlement information, and judging whether the aggregate payment platform is compliant based on the business license information and the fund settlement chain;

[0121] Methods for determining the funds settlement chain based on funds settlement information include:

[0122] Each element in the payment channel information is taken as a node, and the edges between the nodes are constructed according to the fund flow information, so as to construct a fund flow graph, and the fund flow graph is used as the fund settlement chain.

[0123] It can be understood from the above that each element in the payment channel information refers to the information of each institution or user involved in the fund settlement, such as Figure 3 As shown, Figure 3 The illegal fund settlement model carried out by the aggregate payment platform is shown. The illegal fund settlement model refers to consumers first paying funds to the platform account, and then the unlicensed institution settles the funds from the platform account to the merchant account based on the order information. Figure 3 In the payment channel information, the elements are consumer, platform account and merchant account. The fund flow information shows that the funds are first transferred from the consumer to the platform account, and then from the platform account to the merchant account. Therefore, the constructed fund flow diagram is as follows: Figure 4 shown.

[0124] Methods for determining whether an aggregate payment platform is compliant include:

[0125] Based on the business license information, determine whether the aggregate payment platform has a payment business license. If not, the aggregate payment platform is non-compliant. If so, further determine whether the funds settlement chain complies with the standard settlement process. If so, the aggregate payment platform is compliant. If not, the aggregate payment platform is non-compliant.

[0126] S20: When the aggregated payment platform is compliant, the static area information and the dynamic area information of the aggregated payment page are obtained, the number of violations is generated according to the static area information, the sequence similarity is generated according to the dynamic area information, the number of violations and the sequence similarity are input into a pre-built risk classification model, and the payment risk result is obtained, the payment risk result includes at least low risk and high risk, and the aggregated payment page is a payment page in the aggregated payment platform;

[0127] In this embodiment, the aggregated payment page is a payment page in the aggregated payment platform. The static area refers to those areas in the aggregated payment page that remain unchanged during the user operation process. The content of these areas will not change with the user's operation during the payment process. The static area information can be specific content, layout constraints or display elements, etc. The dynamic area refers to the area in the aggregated payment page that will change according to user operations or system status changes. The content of these areas will be dynamically updated during the user interaction process. The dynamic area information can be the payment channel order and payment channel type, etc.

[0128] Methods for generating violation counts based on static region information include:

[0129] Each element in the static area information is checked for integrity, compliance, and standardization in turn, and the inspection results are generated. The number of violations is obtained based on the inspection results. The integrity check refers to checking whether all necessary elements exist, the compliance check refers to comparing whether the attributes of each element meet the specifications, and the standardization check refers to confirming whether the arrangement of page elements meets the specifications.

[0130] The method for generating sequence similarity based on dynamic region information includes:

[0131] Compare the dynamic area information with the preset standard dynamic area, calculate the first similarity between the payment channel order and the standard channel order, calculate the second similarity between the payment channel type and the standard channel type, and take the sum of the first similarity and the second similarity as the sequence similarity. The standard channel order is the arrangement order of different payment channels in the standard dynamic area on the page, and the standard channel type is the payment channel type in the standard dynamic area.

[0132] The construction methods of standard dynamic regions include:

[0133] The number of channel clicks and payment failures is obtained, a dynamic area template is pre-built according to the number of channel clicks, and the dynamic area template is optimized according to the number of payment failures and a natural inspiration algorithm to obtain a standard dynamic area.

[0134] Methods for building dynamic region templates include:

[0135] According to the number of channel clicks corresponding to each payment channel, each payment channel is arranged in descending order, and the channel width corresponding to each payment channel is generated in an arithmetic sequence.

[0136] For example, Figure 7 As shown, Figure 7 The first payment channel 301, the second payment channel 302 and the third payment channel 303 are shown in the figure. The third payment channel 303 has the most channel clicks, while the second payment channel 302 has the least channel clicks, and the first channel width PW3 is greater than the second channel width PW2.

[0137] Methods for optimizing dynamic region templates include:

[0138] S201: Randomly select a dynamic region template as the initial dynamic region EP initial , preset the initial temperature T0 and cooling rate α of simulated annealing, 1>α>0;

[0139] Among them, the initialization process ensures that the algorithm has enough exploration space at the beginning to avoid local optimal solutions. In subsequent iterations, as the temperature decreases, the system gradually reduces the acceptance rate of poor solutions and tends to the optimal solution.

[0140] S202: Initial dynamic area EP initial The number of payment failures f(EP initial ) as the objective function value;

[0141] This embodiment takes the number of payment failures as the objective function and defines the direction of optimization, namely, reducing the number of payment failures to improve the success rate of the payment process and user experience.

[0142] S203: In the current dynamic area EP current Add random disturbances on the basis of to generate new dynamic area EP new , the current dynamic area EP current The starting point is the initial dynamic area EP initial ;

[0143] S204: Calculate new dynamic area EP new The corresponding number of new payment failures f(EP new ), the number of new payment failures f(EPnew ) and the current payment failure number f(EP current ) is compared with the current payment failure number f(EP current ) and the current dynamic area EP current correspond;

[0144] Among them, by comparing the objective function values ​​of the new and old solutions, we can determine whether the new solution is better, and then decide whether to accept the new solution. The advantage of doing this is to ensure that each step moves in the direction of minimizing the objective function, so that the optimization process gradually approaches the optimal solution.

[0145] S205: Determine whether to update the dynamic area, set T = α × T, and return to S203, where the starting point of T is T0;

[0146] This embodiment ensures that the algorithm gradually converges to the optimal solution while exploring the solution space through a gradual cooling process, thereby balancing the capabilities of global exploration and local development.

[0147] S206: Repeat the above steps S203-S205 until T < T min , stop iteration and output the optimal dynamic area EP best , the optimal dynamic area EP best As the standard dynamic region, T is the temperature parameter, T min The preset minimum temperature value.

[0148] Generate new dynamic area EP new The methods include:

[0149] EP new =EP current +Δ EP ;

[0150] Among them, Δ EP is a random disturbance.

[0151] Methods for determining whether to update a dynamic area include:

[0152] If f(EP new )<f(EP current ), then let EP current =EP new ;

[0153] If f(EP new )≥f(EP current ), then randomly generate a probability threshold r between 0 and 1, and determine whether the probability K is greater than the probability threshold r. If so, let EP current =EP new , if not, keep EP current unchanged, the probability K is:

[0154]

[0155] Where exp(·) is an exponential function with e as the base, and e is a natural constant.

[0156] The methods for constructing risk classification models include:

[0157] Obtain Q sets of training data, where Q is a positive integer greater than 1. The training data includes the number of historical violations, historical sequence similarity, and historical payment risk results.

[0158] The number of historical violations, historical sequence similarity, and historical payment risk results are used as sample sets, and the sample sets are divided into training sets and test sets to build a classifier;

[0159] The number of historical violations and historical sequence similarity in the training set are used as input data, and the historical payment risk results in the training set are used as output data to train the classifier and obtain an initial classifier;

[0160] The initial classifier is tested using the test set, and a classifier that meets the preset accuracy is output as the risk classification model. The classifier is preferably a naive Bayes model or a support vector machine model.

[0161] S30: When the payment risk result is low risk, the fund settlement is completed; when the payment risk result is high risk, the fund settlement is interrupted.

[0162] It is understandable that when the payment risk result is low risk, it means that after analyzing the number of violations and sequence similarity in the static and dynamic areas of the payment page, it is determined that the risk of the payment operation is low. This means that the information in the payment process meets the standards and user expectations, the number of violations is small, and the changes in the dynamic area also meet the established specifications. Therefore, when the risk assessment result is low risk, the system considers the payment behavior to be safe and legal, and funds can be settled normally. If the assessment is high risk, there may be security risks or non-compliant operations. The system will interrupt the payment to prevent potential problems in funds settlement and protect the funds of both parties to the transaction.

[0163] Example 3

[0164] This embodiment discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned mobile payment information management method based on aggregated transactions is implemented.

[0165] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0166] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD) or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0167] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0168] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0169] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0170] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0171] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0172] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0173] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A mobile payment information management system based on aggregated transactions, characterized in that: include: Compliance detection module: used to obtain the business license information and fund settlement information of the aggregate payment platform, determine the fund settlement chain according to the fund settlement information, and judge whether the aggregate payment platform is compliant based on the business license information and fund settlement chain; Risk monitoring module: used for obtaining static area information and dynamic area information of the aggregated payment page when the aggregated payment platform is compliant, generating the number of violations according to the static area information, generating sequence similarity according to the dynamic area information, inputting the number of violations and sequence similarity into a pre-built risk classification model, and obtaining payment risk results, which include at least low risk and high risk. The aggregated payment page is a payment page in the aggregated payment platform; Transaction settlement module: used to complete fund settlement when the payment risk result is low risk, and interrupt fund settlement when the payment risk result is high risk.

2. According to claim 1, a mobile payment information management system based on aggregated transactions is characterized in that: The method for determining the funds settlement chain according to the funds settlement information includes: Each element in the payment channel information is taken as a node, and the edges between the nodes are constructed according to the fund flow information, so as to construct a fund flow graph, and the fund flow graph is used as the fund settlement chain.

3. A mobile payment information management system based on aggregated transactions according to claim 2, characterized in that: The method for determining whether the aggregate payment platform is compliant includes: Based on the business license information, determine whether the aggregate payment platform has a payment business license. If not, the aggregate payment platform is non-compliant. If so, further determine whether the funds settlement chain complies with the standard settlement process. If so, the aggregate payment platform is compliant. If not, the aggregate payment platform is non-compliant.

4. According to claim 1, a mobile payment information management system based on aggregated transactions is characterized in that: The method for generating the number of violations according to the static area information includes: Each element in the static area information is checked for integrity, compliance, and standardization in turn, and the inspection results are generated. The number of violations is obtained based on the inspection results. The integrity check refers to checking whether all necessary elements exist, the compliance check refers to comparing whether the attributes of each element meet the specifications, and the standardization check refers to confirming whether the arrangement of page elements meets the specifications.

5. A mobile payment information management system based on aggregated transactions according to claim 4, characterized in that: The dynamic area information includes the payment channel sequence and the payment channel type. The method for generating sequence similarity based on the dynamic area information includes: Compare the dynamic area information with the preset standard dynamic area, calculate the first similarity between the payment channel order and the standard channel order, calculate the second similarity between the payment channel type and the standard channel type, and take the sum of the first similarity and the second similarity as the sequence similarity. The standard channel order is the arrangement order of different payment channels in the standard dynamic area on the page, and the standard channel type is the payment channel type in the standard dynamic area.

6. A mobile payment information management system based on aggregated transactions according to claim 5, characterized in that: The method for constructing the standard dynamic region comprises: The number of channel clicks and payment failures is obtained, a dynamic area template is pre-built according to the number of channel clicks, and the dynamic area template is optimized according to the number of payment failures and a natural inspiration algorithm to obtain a standard dynamic area.

7. A mobile payment information management system based on aggregated transactions according to claim 6, characterized in that: The method for constructing a dynamic region template comprises: According to the number of channel clicks corresponding to each payment channel, each payment channel is arranged in descending order, and the channel width corresponding to each payment channel is generated in an arithmetic sequence.

8. A mobile payment information management system based on aggregated transactions according to claim 7, characterized in that: The risk monitoring module includes an initialization unit, a function construction unit, an update unit, a calculation unit, a judgment unit, and a loop output unit. The method for optimizing the dynamic area template includes: Initialization unit: used to randomly select a dynamic area template as the initial dynamic area EP initial , preset the initial temperature T0 and cooling rate α of simulated annealing, 1>α>0; Function building unit: used to convert the initial dynamic area EP initial The number of payment failures f(EP initial ) as the objective function value; Update unit: used to update the current dynamic area EP current Add random disturbances on the basis of to generate new dynamic area EP new , the current dynamic area EP current The starting point is the initial dynamic area EP initial ; Calculation unit: used to calculate the new dynamic area EP new The corresponding number of new payment failures f(EP new ), the number of new payment failures f(EP new ) and the current payment failure number f(EP current ) is compared with the current payment failure number f(EP current ) and the current dynamic area EP current correspond; A judgment unit is used to judge whether to update the dynamic area, set T = α × T, and return to the update unit, where the starting point of T is T0; Loop output unit: used to repeat the above update unit-judgment unit until T<T min , stop iteration and output the optimal dynamic area EP best , the optimal dynamic area EP best As the standard dynamic region, T is the temperature parameter, T min The preset minimum temperature value.

9. A mobile payment information management system based on aggregated transactions according to claim 8, characterized in that: The new dynamic area EP is generated new The methods include: EP new =EP current +Δ EP ; Among them, Δ EP is a random disturbance; Methods for determining whether to update a dynamic area include: If f(EP new )<f(EP current ), then let EP current =EP new ; If f(EP new )≥f(EP current ), then randomly generate a probability threshold r between 0 and 1, and determine whether the probability K is greater than the probability threshold r. If so, let EP current =EP new , if not, keep EP current unchanged, the probability K is: Where exp(·) is an exponential function with e as the base, and e is a natural constant.

10. A mobile payment information management system based on aggregated transactions according to claim 9, characterized in that: The method for constructing the risk classification model includes: Obtain Q sets of training data, where Q is a positive integer greater than 1. The training data includes the number of historical violations, historical sequence similarity, and historical payment risk results. The number of historical violations, historical sequence similarity, and historical payment risk results are used as sample sets, and the sample sets are divided into training sets and test sets to build a classifier; The number of historical violations and historical sequence similarity in the training set are used as input data, and the historical payment risk results in the training set are used as output data to train the classifier and obtain an initial classifier; The initial classifier is tested using the test set, and the classifier that meets the preset accuracy is output as the risk classification model.

Images