Online payment method based on aggregate payment system and related equipment

Abstract

The invention provides an online payment method based on an aggregation payment system, and the method comprises the steps: determining the to-be-paid amount of a to-be-paid order and the type of a to-be-paid commodity when a target user initiates payment; if the to-be-paid amount is greater than the dynamic payment amount threshold value corresponding to each available payment channel and the number of the to-be-paid commodity categories is greater than the dynamic category number threshold value corresponding to each available payment channel, based on the to-be-paid amount of the target user and the to-be-paid commodity categories, determining the to-be-paid commodity categories in the plurality of available payment channels; determining recommended payment channels, wherein each available payment channel corresponds to a dynamic payment amount threshold; and when the target user determines that the final payment channel is the recommended payment channel, executing payment operation of the target user based on the recommended payment channel. According to the method, the problem of payment failure caused by insufficient threshold or mismatching of categories of a single payment channel can be effectively avoided, the user operation complexity is reduced, the transaction processing time is shortened, and the smoothness of the payment process is improved.

Description

Technical Field

[0001] The present invention relates to the field of digital payment technology, and in particular to an online payment method and related equipment based on an aggregate payment system. Background Art

[0002] With the rapid development of mobile payment technology, users can complete online payments through a variety of payment channels (such as bank cards of various banks, third-party payment platforms, e-wallets, etc.) in daily transactions. In order to improve the convenience of payment, the aggregated payment system came into being. It integrates multiple payment channels to provide users with a unified payment portal. However, the existing aggregated payment system still has certain limitations in practical applications: when a user initiates a payment, the system usually recommends a payment channel based on fixed rules or simple conditions, and lacks dynamic adaptability to complex payment scenarios. For example, in scenarios involving high-value transactions or multiple categories of goods, the existing system may not be able to effectively match the real-time risk control strategy (such as dynamic payment amount threshold, category restriction) of the payment channel bound to the user, resulting in an increase in the payment failure rate or the user needs to manually switch the payment method multiple times, which seriously affects the payment efficiency and user experience. In addition, the existing technology lacks an intelligent decision-making mechanism for the dual dimensions of payment amount and commodity category, making it difficult to optimize the payment success rate while ensuring transaction security. Therefore, there is an urgent need for a solution that can dynamically evaluate the adaptability of payment channels and intelligently recommend the optimal payment strategy based on multi-dimensional conditions to solve the above technical defects. Summary of the invention

[0003] The embodiment of the present invention provides an online payment method based on an aggregate payment system, aiming to provide a solution that can dynamically evaluate the adaptability of payment channels and intelligently recommend the optimal payment strategy based on multi-dimensional conditions. Through multi-dimensional evaluation by dynamic thresholds and commodity categories, the dynamic payment strategy of each payment channel is matched in real time, effectively avoiding the payment failure problem caused by insufficient threshold of a single payment channel or mismatch of categories, reducing the number of times users manually switch payment methods, and shortening transaction processing time; it can avoid the waste of resources caused by frequent attempts at different payment methods, while reducing the complexity of user operations and improving the fluency and satisfaction of the payment process.

[0004] In a first aspect, an embodiment of the present invention provides an online payment method based on an aggregate payment system, wherein the aggregate payment system has multiple available payment channels, and the available payment channels are payment channels that are bound and verified by the user, and the method includes: When the target user initiates payment, the amount to be paid and the category of goods to be paid for the order to be paid are determined; If it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid by the target user and the commodity categories to be paid, a recommended payment channel is determined from the multiple available payment channels, each of which corresponds to a dynamic payment amount threshold; When the target user determines that the final payment channel is the recommended payment channel, the payment operation of the target user is performed based on the recommended payment channel.

[0005] Optionally, before the step of determining that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, the method further includes: Acquire the historical payment information of the target user, wherein the historical payment information includes historical payment amounts, historical payment channels, and historical payment commodity categories; Based on the historical payment amounts, the historical payment channels, and the historical payment commodity categories, a dynamic payment amount threshold and a dynamic category quantity threshold corresponding to each of the available payment channels are determined.

[0006] Optionally, the step of determining the dynamic payment amount threshold and the dynamic category quantity threshold corresponding to each of the available payment channels based on the historical payment amount, the historical payment channel and the historical payment commodity category specifically includes: For the historical payment channels, determining the historical payment amount set and the historical payment commodity category set corresponding to each of the available channels; Performing quantile regression calculation on the historical payment amount set to obtain a dynamic payment amount threshold corresponding to the available payment channel; An association calculation of commodity categories is performed on the historical payment commodity category set to obtain a dynamic category quantity threshold corresponding to the available payment channel.

[0007] Optionally, the step of performing quantile regression calculation on the historical payment amount set to obtain a dynamic payment amount threshold corresponding to the available payment channel includes: Determining a payment amount corresponding to a target percentile based on the historical payment amounts in the historical payment amount set; Based on the historical payment time of the historical payment amount concentration, determine the transaction activity weight, the historical total transaction frequency and the daily average transaction frequency; Obtaining a channel reliability coefficient of the available payment channel and the number of successful transactions of the channel; Based on the payment amount corresponding to the target percentile, the transaction activity weight, the total historical transaction frequency, the average daily transaction frequency, the channel reliability coefficient and the number of successful transactions of the channel, the dynamic payment amount threshold corresponding to the available payment channel is determined.

[0008] Optionally, after the step of determining the dynamic payment amount threshold corresponding to the available payment channel, the method further includes: When the current time period is reached, the channel reliability coefficient is updated according to the average daily transaction frequency to obtain the channel reliability coefficient corresponding to the current time period; When a fluctuation event is detected in the available payment channel within the current time period, the dynamic payment amount threshold is compensated and calculated using the channel reliability coefficient corresponding to the current time period to obtain the compensated dynamic payment amount threshold.

[0009] Optionally, the step of performing association calculation of commodity categories on the historical payment commodity category set to obtain a dynamic category quantity threshold corresponding to the available payment channel specifically includes: Determining a transaction preference level for each of the commodity categories based on a transaction success rate for each of the commodity categories in the historical payment commodity category set; Determining a commodity category association weight matrix based on the transaction preference levels of the commodity categories; Based on the commodity category association weight matrix, a dynamic category quantity threshold corresponding to the available payment channel is determined.

[0010] Optionally, the step of determining a recommended payment channel from the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid specifically includes: Based on the amount to be paid and the category of the goods to be paid, combined with the dynamic payment amount threshold corresponding to each of the available payment channels and the dynamic category quantity threshold corresponding to each of the available payment channels, the order to be paid is split to obtain multiple sub-orders to be paid; Based on the multiple sub-orders to be paid, a recommended payment channel is determined from the multiple available payment channels.

[0011] In a second aspect, an embodiment of the present invention further provides an online payment device based on an aggregate payment system, wherein the aggregate payment system has multiple available payment channels, wherein the available payment channels are payment channels that have been bound and verified by the user, and the online payment device based on the aggregate payment system comprises: The first processing module is used to determine the amount to be paid and the category of goods to be paid for the order to be paid when the target user initiates payment; A second processing module is configured to determine a recommended payment channel among the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid, if it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of the commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, wherein each of the available payment channels corresponds to a dynamic payment amount threshold; The payment module is used to execute the payment operation of the target user based on the recommended payment channel when the target user determines that the final payment channel is the recommended payment channel.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the online payment method based on the aggregated payment system provided in an embodiment of the present invention are implemented.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the online payment method based on the aggregated payment system provided in the embodiment of the invention are implemented.

[0014] In an embodiment of the present invention, when the target user initiates payment, the amount to be paid and the category of goods to be paid for the order to be paid are determined; if it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of the categories of goods to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid by the target user and the category of goods to be paid, a recommended payment channel is determined among the multiple available payment channels, and each of the available payment channels corresponds to a dynamic payment amount threshold; when the target user determines that the final payment channel is the recommended payment channel, the payment operation of the target user is executed based on the recommended payment channel. The present invention performs multi-dimensional evaluation through dynamic thresholds and commodity categories, matches the dynamic payment strategies of each payment channel in real time, effectively avoids payment failures caused by insufficient thresholds of a single payment channel or mismatched categories, reduces the number of times users manually switch payment methods, and shortens transaction processing time; it can avoid waste of resources caused by frequent attempts at different payment methods, while reducing the complexity of user operations and improving the fluency and satisfaction of the payment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 is a flow chart of an online payment method based on an aggregate payment system provided by an embodiment of the present invention; Figure 2 It is a structural diagram of an online payment device based on an aggregate payment system provided by an embodiment of the present invention; Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] 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.

[0018] like Figure 1 As shown, Figure 1 This is a method flow chart of an online payment method based on an aggregate payment system provided by an embodiment of the present invention. The aggregate payment system has multiple available payment channels, and the available payment channels are payment channels that the user has bound and verified. The online payment method based on the aggregate payment system includes the following steps: 101. When the target user initiates payment, the amount to be paid and the category of the goods to be paid for the order to be paid are determined; 102. If it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid of the target user and the commodity categories to be paid, a recommended payment channel is determined from the multiple available payment channels, each of which corresponds to a dynamic payment amount threshold; 103. When the target user determines that the final payment channel is the recommended payment channel, perform a payment operation of the target user based on the recommended payment channel.

[0019] In an embodiment of the invention, when a target user initiates a payment request through the aggregate payment system, the system obtains an order to be paid, which includes an amount to be paid and a category of goods to be paid.

[0020] The amount to be paid is the actual payment amount after parsing the total price, shipping fee, and discount deduction of the goods through the order transaction interface. The category of the goods to be paid is based on the classification identification in the goods database (such as food, electronic products, virtual services, etc.). The category labels of all goods in the order can be extracted and the quantity of different categories can be counted.

[0021] The system can obtain real-time dynamic threshold data of all available payment channels bound by the user from the pre-configured payment channel management module. The real-time dynamic threshold data includes: dynamic payment amount threshold and dynamic category quantity threshold.

[0022] The dynamic payment amount threshold can be understood as a dynamic calculation based on the risk control strategies of each payment channel, such as single transaction limit, daily cumulative limit, user preference amount (small payment, medium payment and large payment), user historical transaction behavior and current period risk level. The dynamic category quantity threshold can be understood as generated based on the payment channel's restriction rules on commodity categories (such as limiting the number of virtual commodities and cross-border commodities).

[0023] If the amount to be paid exceeds the dynamic payment amount threshold of all available payment channels, and the total number of commodity categories to be paid exceeds the dynamic category quantity threshold of all available payment channels, the recommended payment channel calculation module is triggered to recommend available payment channels. If any of the conditions is not met, the payment channel can be selected according to the default rules (such as user default, historical usage frequency).

[0024] Specifically, the amount to be paid is compared with the dynamic payment amount thresholds of all available payment channels. If, among all available payment channels, there is no available payment channel whose dynamic payment amount threshold is smaller than the amount to be paid, it can be determined that the amount to be paid exceeds the dynamic payment amount threshold of all available payment channels. If there is any available payment channel whose dynamic payment amount threshold is less than or equal to the amount to be paid, it can be determined that the amount to be paid does not exceed the dynamic payment amount threshold of all available payment channels.

[0025] The total number of commodity categories to be paid is compared with the dynamic category quantity threshold of all available payment channels. If among all available payment channels, there is no available payment channel whose dynamic category quantity threshold is less than the total number of commodity categories to be paid, it can be determined that the total number of commodity categories to be paid exceeds the dynamic category quantity threshold of all available payment channels. If there is any available payment channel whose total number of commodity categories to be paid is less than or equal to the amount to be paid, it can be determined that the total number of commodity categories to be paid does not exceed the dynamic category quantity threshold of all available payment channels.

[0026] The weight coefficient of each available payment channel can be set according to the payment amount adaptation (the ratio of the amount to be paid to the remaining available amount of the available payment channel) and the commodity category matching (the overlap between the order category and the category allowed by the available payment channel); the TOP-N available payment channels with the largest to smallest weight coefficients are used as recommended payment channels.

[0027] In a possible embodiment, the available payment channels are sorted in descending order according to their comprehensive scores, and the channels with the highest scores are preferentially selected as recommended payment channels; the comprehensive scores include the usage frequency of the available payment channels, the manual selection frequency of the available payment channels, and the preferential strength of the available payment channels (such as subsidies for dedicated payment channels).

[0028] In a possible embodiment, if the recommended payment channel is unavailable due to external reasons (such as temporary maintenance of the bank system), it will automatically switch to the suboptimal channel and update the recommendation result.

[0029] In a possible embodiment, for high-value multi-category orders, when automatic payment splitting is allowed in the target user's account settings, a splitting plan can be dynamically generated based on order characteristics and payment channel restrictions, including three modes: splitting by amount ratio, splitting by product category grouping, and mixed splitting.

[0030] By splitting a high-value multi-category order into multiple sub-orders, multiple available payment channels are recommended to pay the multiple sub-orders. When the target user determines that the final payment channel is the recommended payment channel, the payment operation of the multiple sub-orders is performed based on the recommended payment channel.

[0031] In an embodiment of the present invention, when the target user initiates payment, the amount to be paid and the category of goods to be paid for the order to be paid are determined; if it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of the categories of goods to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid by the target user and the category of goods to be paid, a recommended payment channel is determined among the multiple available payment channels, and each of the available payment channels corresponds to a dynamic payment amount threshold; when the target user determines that the final payment channel is the recommended payment channel, the payment operation of the target user is executed based on the recommended payment channel. The present invention performs multi-dimensional evaluation through dynamic thresholds and commodity categories, matches the dynamic payment strategies of each payment channel in real time, effectively avoids payment failures caused by insufficient thresholds of a single payment channel or mismatched categories, reduces the number of times users manually switch payment methods, and shortens transaction processing time; it can avoid waste of resources caused by frequent attempts at different payment methods, while reducing the complexity of user operations and improving the fluency and satisfaction of the payment process.

[0032] It is understandable that in the specific implementation of this application, related data such as user data, payment data, channel data, order data, behavior data, etc. are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0033] Optionally, before the step of determining that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, the method further includes: obtaining historical payment information of the target user, the historical payment information including historical payment amounts, historical payment channels, and historical payment commodity categories; determining the dynamic payment amount threshold and dynamic category number threshold corresponding to each of the available payment channels based on the historical payment amounts, the historical payment channels, and the historical payment commodity categories.

[0034] In the embodiment of the present invention, the historical payment information of the target user may be extracted from the payment log database.

[0035] The historical payment amount may be the payment amount of each transaction of the user. The single transaction amount, the average daily transaction amount, and the monthly cumulative transaction amount completed by the user through each payment channel may be counted based on the historical payment amount.

[0036] The historical payment channel may be the payment channel used for each transaction of the user, and statistics on the usage frequency, the most recent usage time and the success rate of each payment channel may be collected based on the historical payment channel.

[0037] The historical payment commodity categories may be the commodity categories included in each transaction, and the classification data of related order commodities, the commodity categories and quantity distribution paid by the user through each payment channel, etc. may be counted based on the historical payment commodity categories.

[0038] After obtaining the historical payment information, data preprocessing can be performed on the historical payment information. Data preprocessing can include removing abnormal data (such as test transactions, refund orders); desensitizing sensitive information (such as user ID, bank card number), and only using system identifiers for encoding; sliding statistics according to preset time windows (such as the last 30 days) to ensure data timeliness.

[0039] Based on historical payment information, dynamic thresholds for each payment channel can be generated through a rule engine and machine learning model.

[0040] Specifically, for the calculation of the dynamic payment amount threshold, the fixed risk control limit preset by the payment channel (such as the maximum single transaction limit of 50,000 yuan stipulated by the bank) can be taken; combined with the user behavior correction factor, for example, if the maximum successful payment amount of the user through the channel in history is M, and the recent success rate (such as within 7 days) is ≥ 95%, the dynamic threshold is increased to min (basic threshold, M×safety factor), where the safety factor is 1.1-1.3; if the user has a payment failure recently (such as insufficient balance, risk control interception), the dynamic threshold is lowered to the basic threshold×attenuation factor, and the attenuation factor decreases in steps of 0.9, 0.8... according to the number of failures. For example: the historical maximum successful payment amount of user B through payment channel P is 8,000 yuan, the basic threshold is 10,000 yuan, and the safety factor is 1.2, then the dynamic payment amount threshold = min (10,000, 8,000×1.2) = 9,600 yuan.

[0041] For the calculation of the dynamic category quantity threshold, the hard restrictions on commodity categories imposed by the payment channel can be extracted (e.g., virtual commodities are prohibited, and cross-border commodities cannot exceed 2 categories); the distribution of the number of commodity categories successfully paid by users through the payment channel is counted. If the user frequently pays for multi-category orders (e.g., an average of more than 3 categories per month) and has no risk control records, the dynamic threshold is increased to the initial threshold + elastic increment (e.g., +1 category); if the user is recently intercepted due to category exceeding the limit, the dynamic threshold is reduced to the initial threshold - penalty reduction (e.g., -1 category). For example: the initial category threshold of payment channel Q is 2 categories, and user C has successfully paid 5 3-category orders through Q in the past 30 days, then the dynamic category threshold is adjusted to 3 categories.

[0042] The calculated dynamic threshold is written into the cache (such as Redis) to ensure millisecond-level reading; upon receiving a policy change notification from the payment channel (such as a temporary limit adjustment by the bank or an active limit adjustment by the user), the threshold is triggered to be recalculated.

[0043] For example, user D binds payment channels R (basic amount threshold of 5,000 yuan) and S (basic amount threshold of 8,000 yuan). His historical payment behavior is as follows: he successfully paid 6 orders through channel R, with an amount range of 3,000-6,000 yuan (exceeding the basic threshold of channel R but not blocked by the bank); when paying through channel S, he was blocked twice by risk control due to frequent payments exceeding 8,000 yuan. Channel R dynamic amount threshold: min(5,000, 6,000×1.2)=6,000 yuan (due to historical successful payments of 6,000 yuan without failure); channel S dynamic amount threshold: 8,000×0.8=6,400 yuan (due to the recent number of failures triggering attenuation). When user D initiates a payment of 7,000 yuan, the system automatically excludes channel S (threshold 6,400 yuan) and only recommends channel R (threshold 6,000 yuan). If the user accepts the excess (7,000-6,000=1,000 yuan) to be supplemented by the balance, the payment is completed.

[0044] In some possible embodiments, a random forest or neural network model may also be trained, with input features including historical amounts, categories, payment times, user credit scores, etc., and dynamic threshold adjustment suggestions output; the model is regularly trained offline, and the threshold is corrected in real time during online inference.

[0045] In some possible embodiments, a time decay weight (such as exponential decay) may also be assigned to the historical data so that recent transactions have a greater impact on the threshold.

[0046] In some possible embodiments, when the historical payment information of the target user is relatively sparse, the dynamic threshold may be initialized with reference to the historical payment behaviors of similar user groups.

[0047] The present invention breaks through the fixed restrictions of payment channels through personalized threshold adaptation, dynamically relaxes or tightens the threshold based on the user's actual payment ability, and improves payment flexibility; the threshold is automatically adjusted according to the user's historical behavior, thereby preventing malicious cash-out or money laundering risks while improving the success rate. Optionally, the step of determining the dynamic payment amount threshold and the dynamic category quantity threshold corresponding to each of the available payment channels based on the historical payment amounts, the historical payment channels and the historical payment commodity categories specifically includes: for the historical payment channels, determining the historical payment amount set and the historical payment commodity category set corresponding to each of the available channels; performing quantile regression calculation on the historical payment amount set to obtain the dynamic payment amount threshold corresponding to the available payment channel; performing commodity category association calculation on the historical payment commodity category set to obtain the dynamic category quantity threshold corresponding to the available payment channel.

[0048] In an embodiment of the present invention, for each available payment channel, the historical payment amount set and the historical payment commodity category set corresponding to the channel are extracted from the historical payment information. The historical payment amount set is the set of all order amounts successfully paid by the user through the channel; the historical payment commodity category set is the set of commodity category labels contained in the payment order of the user through the channel. For example: the historical amount set of payment channel K is {3,000 yuan, 5,000 yuan, 7,000 yuan}, and the category set is {electronic products, home appliances, cross-border commodities}.

[0049] Set the target quantile Q based on the risk tolerance of the payment channel p , Q p The value range of is [0.85-1], which means that 85%-95% of the historical successful transaction amounts are covered; if the channel needs to allow most normal transactions but intercept transactions with extremely high failure risks, select τ=0.99.

[0050] A quantile regression model is established with time, channel score, and transaction scenario as independent variables and historical payment amount as the dependent variable. Then Q p (amount|X) = β 0 + β 1 * Time + β 2 *Credit score + β 3 *Scenario coefficient, use gradient descent method to solve parameter β, minimize the quantile loss function, input current transaction features (such as time = "holiday", credit score = 750, scenario = "big promotion"), calculate the predicted quantile value Q p ; Combined with the fixed risk control limit of the payment channel, the dynamic payment amount threshold is taken as min(Q p , fixed limit).

[0051] Apriori algorithm or FP-Growth algorithm can be used to extract frequent item sets and association rules from the historical payment commodity category set; for example, it is found that "electronic products" and "accessories" often appear at the same time (support = 30%), forming a frequent item set {electronic products, accessories}. Extract the maximum number of categories N_max that users successfully paid through this channel from the historical commodity category set; if the association confidence of a category is greater than the threshold (such as 70%), it will be merged into a logical "super category", and the dynamic category number threshold = N_max × merging coefficient (merging coefficient ∈ [0.5, 1], the stronger the association, the smaller the coefficient), the historical maximum number of categories N_max = 4, the merging coefficient = 0.75 (because there are 2 groups of strongly associated categories), then the dynamic threshold = 4×0.75 = 3 categories.

[0052] It should be noted that if the payment channel has strict restrictions on specific categories (such as virtual goods), the category quantity threshold for such orders is enforced to be 1.

[0053] Optionally, the step of performing quantile regression calculation on the historical payment amount set to obtain the dynamic payment amount threshold corresponding to the available payment channel includes: determining the payment amount corresponding to the target percentile based on the historical payment amounts in the historical payment amount set; determining the transaction activity weight, the historical total transaction frequency and the average daily transaction frequency based on the historical payment time in the historical payment amount set; obtaining the channel reliability coefficient of the available payment channel and the number of successful channel transactions; determining the dynamic payment amount threshold corresponding to the available payment channel based on the payment amount corresponding to the target percentile, the transaction activity weight, the historical total transaction frequency, the average daily transaction frequency, the channel reliability coefficient and the number of successful channel transactions.

[0054] In the embodiment of the present invention, the channel historical payment amount dataset M={m 1 ,m 2 ,...,m n}, perform sliding window quantile calculation and obtain the basic payment threshold: Th base =Q min (M [t−w:t]+ )+H⋅Q max (M [t−w:t]+ ) Among them, Th base is the basic payment threshold, Q min is the minimum quantile, Q max is the maximum quantile, t represents the tth historical payment information in the historical payment amount dataset, M t−w:t represents the first [t−w:t] +The historical payment amount corresponding to the historical payment information, w is the time window, [ ] + It means rounding up, H is the empirical coefficient, and the value range of H is [0.10, 0.40].

[0055] Calculation of dynamic payment amount threshold: Th x =Th base ⋅α⋅(1+0.1⋅log (1+f day / f T )) α= N success / N total +0.2⋅tanh(1 / UT avg ) Among them, Th x is the dynamic payment amount threshold, α is the channel reliability coefficient, and f day is the average daily transaction frequency, f T is the total historical transaction frequency, UT avg is the channel reliability coefficient, N success is the number of successful transactions on the channel, N total The channel reliability coefficient can be determined as the channel transaction number and the channel average response time.

[0056] Optionally, after the step of determining the dynamic payment amount threshold corresponding to the available payment channel, the method further includes: when the current time period is reached, updating the channel reliability coefficient according to the average daily transaction frequency, and obtaining the channel reliability coefficient corresponding to the current time period; when a fluctuation event is detected in the available payment channel within the current time period, compensating the dynamic payment amount threshold by using the channel reliability coefficient corresponding to the current time period to obtain the compensated dynamic payment amount threshold.

[0057] In the embodiment of the present invention, the calibration reliability coefficient is performed every ΔT=6 hours, as shown in the following formula: α′=α⋅[1+( N success − f day ) / ( N success +f day +ϵ) ] Among them, ϵ is the smoothing factor, and the value range of ϵ is [0.1, 0.3].

[0058] When the next time period is reached, the compensation item is injected to obtain the dynamic payment amount threshold Th after compensation. x ′, as follows: Th x ′=Th x +α′⋅Q p After the dynamic payment amount threshold after compensation is obtained, it is used to determine whether the order to be paid is a recent high-value multi-category order.

[0059] Optionally, the step of performing association calculation on the historical payment commodity category set to obtain the dynamic category quantity threshold value corresponding to the available payment channel specifically includes: determining the transaction preference level of each commodity category based on the transaction success rate of each commodity category in the historical payment commodity category set; determining the commodity category association weight matrix based on the transaction preference level of the commodity category; and determining the dynamic category quantity threshold value corresponding to the available payment channel based on the commodity category association weight matrix.

[0060] In the embodiment of the present invention, the LightGBM model can be used to extract the transaction preference level r∈[0,1] of the commodity category in the historical payment commodity category set, and calculate the commodity category association weight matrix:

[0061] in, is the association weight between product category i and product category j, d is the similarity (measured distance) between product category i and product category j, is the transaction preference level of commodity category i, is the transaction preference level of commodity category j, Indicates normalization processing, converting the value to a value in [0,1].

[0062] The dynamic category number threshold is calculated as follows:

[0063] in, is the dynamic category quantity threshold, is the standard deviation of the commodity category association weight matrix, is the transaction frequency of the day, is the preset reference transaction frequency, Indicates taking the larger of two values.

[0064] Optionally, the step of determining a recommended payment channel from the multiple available payment channels based on the amount to be paid of the target user and the category of the goods to be paid specifically includes: based on the amount to be paid and the category of the goods to be paid, combined with a dynamic payment amount threshold corresponding to each of the available payment channels, and a dynamic category quantity threshold corresponding to each of the available payment channels, splitting the order to be paid to obtain multiple sub-orders to be paid; based on the multiple sub-orders to be paid, determining a recommended payment channel from the multiple available payment channels.

[0065] In the embodiment of the present invention, for high-value multi-category orders, when the target user's account settings allow automatic payment splitting, a splitting scheme can be dynamically generated based on order features and payment channel restrictions, including splitting by amount ratio, splitting by commodity category grouping, and mixed splitting: Among them, split by amount ratio: split the amount to be paid into several sub-orders, so that the amount of each sub-order is ≤ the dynamic amount threshold of at least one available payment channel.

[0066] Example: The total amount is 12,000 yuan, and the amount thresholds of payment channels A and B are 8,000 yuan and 6,000 yuan respectively. Then it will be split into a sub-order of 8,000 yuan (using A) + a sub-order of 4,000 yuan (using B).

[0067] Group and split by product category: Group the products into multiple sub-orders by category, so that the number of categories in each sub-order is ≤ the dynamic category number threshold of the target payment channel.

[0068] Example: An order contains three categories of goods (electronic products, food, and cross-border goods). The category thresholds of payment channels C and D are 2 and 1 respectively. Then the order is split into sub-orders of "electronic products + food" (using C) + sub-orders of "cross-border goods" (using D).

[0069] Mixed splitting: Multi-dimensional splitting based on both amount and category restrictions.

[0070] Specifically, for mixed splitting, we can perform structured analysis on the original order to extract the amount distribution of each category of goods and category correlation. The amount distribution includes: total amount, independent pricing of sub-goods / services (such as 5,000 yuan for product A and 3,000 yuan for product B); category correlation includes: logical grouping of product categories (such as "electronic products" and "accessories" can be merged into the same category), cross-category restriction rules (such as some payment channels prohibit the simultaneous payment of "virtual goods" and "physical goods").

[0071] Based on the dynamic threshold of available payment channels, a split constraint model is constructed, which includes amount constraints, category constraints and business rules. The details are as follows: Amount constraint: The amount of each sub-order ≤ the dynamic payment amount threshold of at least one available payment channel.

[0072] Category constraint: The number of categories contained in each sub-order must be ≤ the dynamic category quantity threshold of at least one available payment channel.

[0073] Business rules: such as non-separable goods (for example, gifts must be paid for together with the main goods), split account requirements (for example, different merchants must settle independently).

[0074] In a possible embodiment, after the splitting constraint model is constructed, priority-driven splitting may be adopted under the constraints of the splitting constraint model, such as amount priority or category priority.

[0075] Among them, the amount is prioritized as follows: if the amount exceeds the limit and is the main contradiction, split it by the amount threshold first, and then group the sub-orders by category. Example: The total amount is 15,000 yuan (exceeding the limit), split into 10,000 yuan (using channel M) + 5,000 yuan (using channel N), and then check whether the category of the sub-order meets the target channel limit.

[0076] Category priority is: if the category exceeds the limit more seriously, group by category first, and then adjust the sub-order amount to within the channel threshold. Example: An order contains 4 categories of goods, which are split into "category 1+2" (using channel P, category threshold 2) and "category 3+4" (using channel Q, category threshold 2), and then distribute the amount so that each part is ≤ the channel amount threshold.

[0077] In another possible embodiment, after the split constraint model is constructed, under the constraints of the split constraint model, an optimization objective function can be defined to balance the amount and category fitness through weights, as follows: Objective function = α*(amount fitness) + β*(category fitness) Among them, α is the weight of the amount adaptation, β is the weight of the category adaptation, α+β=1, and it is dynamically adjusted according to user or system policies.

[0078] Amount adaptability = sub-order amount / dynamic payment amount threshold of the target available payment channel (the closer to the threshold, the higher the adaptability); Category suitability = number of sub-order categories / threshold of dynamic category quantity of target available payment channel (the closer to the threshold, the higher the suitability).

[0079] Greedy algorithm, genetic algorithm or integer programming can be used to solve the optimal split combination to obtain multiple sub-orders.

[0080] It should be noted that for non-separable items, you can mark the products that need to be paid in full (such as bundled sales packages) to ensure that they are not split into different sub-orders; for cross-channel conflicts, if a certain product category is prohibited by multiple channels, it will be automatically assigned to an unrestricted channel; for user intervention, a split preview interface can be provided, allowing users to manually adjust the split ratio or specify the payment channel.

[0081] like Figure 2 As shown, an embodiment of the present invention provides an online payment device based on an aggregate payment system, and the online payment device based on the aggregate payment system includes: The first processing module 201 is used to determine the amount to be paid and the category of goods to be paid for the order to be paid when the target user initiates payment; The second processing module 202 is used to determine a recommended payment channel from the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid, if it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of the commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, wherein each of the available payment channels corresponds to a dynamic payment amount threshold; The payment module 203 is configured to execute the payment operation of the target user based on the recommended payment channel when the target user determines that the final payment channel is the recommended payment channel.

[0082] Optionally, the device further comprises: An acquisition module, used to acquire the historical payment information of the target user, wherein the historical payment information includes historical payment amounts, historical payment channels, and historical payment commodity categories; The third processing module is used to determine the dynamic payment amount threshold and the dynamic category quantity threshold corresponding to each of the available payment channels based on the historical payment amount, the historical payment channel and the historical payment commodity category.

[0083] Optionally, the third processing module is also used to determine, for the historical payment channels, the historical payment amount set and the historical payment commodity category set corresponding to each of the available channels; perform quantile regression calculation on the historical payment amount set to obtain a dynamic payment amount threshold corresponding to the available payment channel; perform commodity category association calculation on the historical payment commodity category set to obtain a dynamic category quantity threshold corresponding to the available payment channel.

[0084] Optionally, the third processing module is also used to determine the payment amount corresponding to the target percentile based on the historical payment amount in the historical payment amount set; determine the transaction activity weight, the historical total transaction frequency and the average daily transaction frequency based on the historical payment time in the historical payment amount set; obtain the channel reliability coefficient of the available payment channel and the number of successful channel transactions; determine the dynamic payment amount threshold corresponding to the available payment channel based on the payment amount corresponding to the target percentile, the transaction activity weight, the historical total transaction frequency, the average daily transaction frequency, the channel reliability coefficient and the number of successful channel transactions.

[0085] Optionally, the third processing module is also used to update the channel reliability coefficient according to the average daily transaction frequency when the current time period is reached, so as to obtain the channel reliability coefficient corresponding to the current time period; when a fluctuation event is detected in the available payment channel within the current time period, the dynamic payment amount threshold is compensated and calculated using the channel reliability coefficient corresponding to the current time period to obtain the compensated dynamic payment amount threshold.

[0086] Optionally, the third processing module is also used to determine the transaction preference level of each commodity category based on the transaction success rate of each commodity category in the historical payment commodity category set; determine the commodity category association weight matrix based on the transaction preference level of the commodity category; and determine the dynamic category quantity threshold corresponding to the available payment channel based on the commodity category association weight matrix.

[0087] Optionally, the second processing module 202 is also used to split the order to be paid based on the amount to be paid and the category of goods to be paid, combined with the dynamic payment amount threshold corresponding to each of the available payment channels, and the dynamic category quantity threshold corresponding to each of the available payment channels, to obtain multiple sub-orders to be paid; based on the multiple sub-orders to be paid, determine the recommended payment channel from the multiple available payment channels.

[0088] It should be noted that the online payment device based on the aggregate payment system provided in the embodiment of the present invention can be applied to devices such as computers and servers that can perform online payment methods based on the aggregate payment system.

[0089] The online payment device based on the aggregate payment system provided in the embodiment of the present invention can implement each process implemented by the online payment method based on the aggregate payment system in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0090] See also Figure 3 , Figure 3is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program of an online payment method based on an aggregated payment system stored in the memory 302 and executable on the processor 301, wherein: The processor 301 is used to call the computer program stored in the memory 302 and execute the following steps: When the target user initiates payment, the amount to be paid and the category of goods to be paid for the order to be paid are determined; If it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid by the target user and the commodity categories to be paid, a recommended payment channel is determined from the multiple available payment channels, each of which corresponds to a dynamic payment amount threshold; When the target user determines that the final payment channel is the recommended payment channel, the payment operation of the target user is performed based on the recommended payment channel.

[0091] Optionally, before the step of determining that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, the method executed by the processor 301 further includes: Acquire the historical payment information of the target user, wherein the historical payment information includes historical payment amounts, historical payment channels, and historical payment commodity categories; Based on the historical payment amounts, the historical payment channels, and the historical payment commodity categories, a dynamic payment amount threshold and a dynamic category quantity threshold corresponding to each of the available payment channels are determined.

[0092] Optionally, the step of determining the dynamic payment amount threshold and the dynamic category quantity threshold corresponding to each of the available payment channels based on the historical payment amount, the historical payment channel, and the historical payment commodity category, performed by the processor 301 specifically includes: For the historical payment channels, determining the historical payment amount set and the historical payment commodity category set corresponding to each of the available channels; Performing quantile regression calculation on the historical payment amount set to obtain a dynamic payment amount threshold corresponding to the available payment channel; An association calculation of commodity categories is performed on the historical payment commodity category set to obtain a dynamic category quantity threshold corresponding to the available payment channel.

[0093] Optionally, the step of performing quantile regression calculation on the historical payment amount set to obtain the dynamic payment amount threshold corresponding to the available payment channel, performed by the processor 301, includes: Determining a payment amount corresponding to a target percentile based on the historical payment amounts in the historical payment amount set; Based on the historical payment time of the historical payment amount concentration, determine the transaction activity weight, the historical total transaction frequency and the daily average transaction frequency; Obtaining a channel reliability coefficient of the available payment channel and the number of successful transactions of the channel; Based on the payment amount corresponding to the target percentile, the transaction activity weight, the total historical transaction frequency, the average daily transaction frequency, the channel reliability coefficient and the number of successful transactions of the channel, the dynamic payment amount threshold corresponding to the available payment channel is determined.

[0094] Optionally, after the step of determining the dynamic payment amount threshold corresponding to the available payment channel, the method executed by the processor 301 further includes: When the current time period is reached, the channel reliability coefficient is updated according to the average daily transaction frequency to obtain the channel reliability coefficient corresponding to the current time period; When a fluctuation event is detected in the available payment channel within the current time period, the dynamic payment amount threshold is compensated and calculated using the channel reliability coefficient corresponding to the current time period to obtain the compensated dynamic payment amount threshold.

[0095] Optionally, the step of performing association calculation of commodity categories on the historical payment commodity category set to obtain the dynamic category quantity threshold corresponding to the available payment channel, performed by the processor 301, specifically includes: Determining a transaction preference level for each of the commodity categories based on a transaction success rate for each of the commodity categories in the historical payment commodity category set; Determining a commodity category association weight matrix based on the transaction preference levels of the commodity categories; Based on the commodity category association weight matrix, a dynamic category quantity threshold corresponding to the available payment channel is determined.

[0096] Optionally, the step of determining a recommended payment channel from the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid, performed by the processor 301, specifically includes: Based on the amount to be paid and the category of the goods to be paid, combined with the dynamic payment amount threshold corresponding to each of the available payment channels and the dynamic category quantity threshold corresponding to each of the available payment channels, the order to be paid is split to obtain multiple sub-orders to be paid; Based on the multiple sub-orders to be paid, a recommended payment channel is determined from the multiple available payment channels.

[0097] It should be noted that the electronic device provided in the embodiment of the present invention can be applied to computers, servers and other devices that can perform online payment methods based on the aggregate payment system.

[0098] The electronic device provided in the embodiment of the present invention can implement each process implemented by the online payment method based on the aggregate payment system in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0099] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the online payment method based on the aggregated payment system provided by the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0100] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the computer-readable storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0101] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. An online payment method based on an aggregate payment system, wherein the aggregate payment system has multiple available payment channels, wherein the available payment channels are payment channels that the user has bound and verified, and wherein: The method comprises the following steps: When the target user initiates payment, the amount to be paid and the category of goods to be paid for the order to be paid are determined; If it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, then based on the amount to be paid by the target user and the commodity categories to be paid, a recommended payment channel is determined from the multiple available payment channels, each of which corresponds to a dynamic payment amount threshold; When the target user determines that the final payment channel is the recommended payment channel, the payment operation of the target user is performed based on the recommended payment channel.

2. The online payment method based on the aggregate payment system as claimed in claim 1, characterized in that: Before the step of determining that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, the method further includes: Acquire the historical payment information of the target user, wherein the historical payment information includes historical payment amounts, historical payment channels, and historical payment commodity categories; Based on the historical payment amounts, the historical payment channels, and the historical payment commodity categories, a dynamic payment amount threshold and a dynamic category quantity threshold corresponding to each of the available payment channels are determined.

3. The online payment method based on the aggregate payment system as claimed in claim 2, characterized in that: The step of determining the dynamic payment amount threshold and the dynamic category quantity threshold corresponding to each of the available payment channels based on the historical payment amount, the historical payment channel and the historical payment commodity category specifically includes: For the historical payment channels, determining the historical payment amount set and the historical payment commodity category set corresponding to each of the available channels; Performing quantile regression calculation on the historical payment amount set to obtain a dynamic payment amount threshold corresponding to the available payment channel; An association calculation of commodity categories is performed on the historical payment commodity category set to obtain a dynamic category quantity threshold corresponding to the available payment channel.

4. The online payment method based on the aggregate payment system as claimed in claim 3, characterized in that: The step of performing quantile regression calculation on the historical payment amount set to obtain the dynamic payment amount threshold corresponding to the available payment channel comprises: Determining a payment amount corresponding to a target percentile based on the historical payment amounts in the historical payment amount set; Based on the historical payment time of the historical payment amount concentration, determine the transaction activity weight, the historical total transaction frequency and the daily average transaction frequency; Obtaining a channel reliability coefficient of the available payment channel and the number of successful transactions of the channel; Based on the payment amount corresponding to the target percentile, the transaction activity weight, the total historical transaction frequency, the average daily transaction frequency, the channel reliability coefficient and the number of successful transactions of the channel, the dynamic payment amount threshold corresponding to the available payment channel is determined.

5. The online payment method based on the aggregate payment system as claimed in claim 4, characterized in that: After the step of determining the dynamic payment amount threshold corresponding to the available payment channel, the method further includes: When the current time period is reached, the channel reliability coefficient is updated according to the average daily transaction frequency to obtain the channel reliability coefficient corresponding to the current time period; When a fluctuation event is detected in the available payment channel within the current time period, the dynamic payment amount threshold is compensated and calculated using the channel reliability coefficient corresponding to the current time period to obtain the compensated dynamic payment amount threshold.

6. The online payment method based on the aggregate payment system as claimed in claim 3, characterized in that: The step of performing association calculation on the commodity category set of the historical payment commodity category set to obtain the dynamic category quantity threshold corresponding to the available payment channel specifically includes: Determining a transaction preference level for each of the commodity categories based on a transaction success rate for each of the commodity categories in the historical payment commodity category set; Determining a commodity category association weight matrix based on the transaction preference levels of the commodity categories; Based on the commodity category association weight matrix, a dynamic category quantity threshold corresponding to the available payment channel is determined.

7. The online payment method based on the aggregate payment system according to any one of claims 1 to 6, characterized in that: The step of determining a recommended payment channel from the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid specifically comprises: Based on the amount to be paid and the category of the goods to be paid, combined with the dynamic payment amount threshold corresponding to each of the available payment channels and the dynamic category quantity threshold corresponding to each of the available payment channels, the order to be paid is split to obtain multiple sub-orders to be paid; Based on the multiple sub-orders to be paid, a recommended payment channel is determined from the multiple available payment channels.

8. An online payment device based on an aggregate payment system, wherein the aggregate payment system has multiple available payment channels, wherein the available payment channels are payment channels that have been bound and verified by the user, characterized in that: The online payment device based on the aggregate payment system includes: The first processing module is used to determine the amount to be paid and the category of goods to be paid for the order to be paid when the target user initiates payment; A second processing module is configured to determine a recommended payment channel among the plurality of available payment channels based on the amount to be paid by the target user and the category of the commodity to be paid, if it is determined that the amount to be paid is greater than the dynamic payment amount threshold corresponding to each of the available payment channels, and the number of the commodity categories to be paid is greater than the dynamic category number threshold corresponding to each of the available payment channels, wherein each of the available payment channels corresponds to a dynamic payment amount threshold; The payment module is used to execute the payment operation of the target user based on the recommended payment channel when the target user determines that the final payment channel is the recommended payment channel.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps in the online payment method based on the aggregated payment system as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the online payment method based on the aggregate payment system as described in any one of claims 1 to 7 are implemented.

Images