Cross-border payment processing method and system

Abstract

The invention relates to a cross-border payment processing method, and belongs to the field of data processing systems or methods specially suitable for administrative, commercial, financial, management, supervision or prediction purposes, and the method comprises the steps: carrying out each training operation on a convolutional neural network to obtain a cross-border payment intelligent identification model; and intelligently identifying whether the current cross-border payment request belongs to a cross-border high-risk payment request or not by adopting a cross-border payment intelligent identification model according to the sales data of the seller in the current cross-border payment request in multiple historical cross-border payment. The invention also relates to a cross-border payment processing system. Through the method and the device, the technical problem that in the prior art, whether a cross-border payment request relates to identification of high-risk payment is carried out on a client or not, so that too many troubles and data burdens are brought to the client is solved; an artificial intelligence model can be adopted to complete intelligent identification of whether the current cross-border payment request belongs to cross-border high-risk payment at a server side of a payment operation company, so that the technical problem is solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing systems or methods specifically suitable for administrative, commercial, financial, management, supervisory or forecasting purposes, and in particular to a cross-border payment processing method and system. Background Art

[0002] Cross-border payment refers to the act of transferring funds across borders and regions by using certain settlement tools and payment systems for international bond debts arising from international trade, international investment and other aspects between two or more countries or regions. For example, when Chinese consumers purchase products from foreign merchants online or foreign consumers purchase products from Chinese merchants, due to the difference in currency, it is necessary to use certain settlement tools and payment systems to realize the conversion of funds between the two countries or regions and finally complete the transaction. Generally, payment operation companies with cross-border payment licenses complete the cross-border payment processing between buyers and sellers.

[0003] For example, a Chinese invention patent with a publication number of CN115760100A proposes a cross-border payment processing method and device, which involves finance or other technical fields. The method includes: determining multiple cross-border payment channels for users to choose from, and transaction parameters of each cross-border payment channel; in the user's operation interface, displaying multiple cross-border payment channels and transaction parameters of each cross-border payment channel to the user, so that the user can select a cross-border payment channel through the operation interface according to the transaction parameters of each cross-border payment channel; after the user initiates a cross-border payment transaction, the cross-border payment transaction initiated by the user is processed with the cross-border payment channel selected by the user. The present invention can facilitate users to select appropriate cross-border payment channels during cross-border payment transactions, simplify user operations, and improve user experience.

[0004] For example, the Chinese invention patent publication CN111435498A proposes a cross-border payment processing method, device and electronic device, the method comprising: in response to the buyer's exchange lock operation, sending an exchange lock instruction to the exchange partner; in response to the buyer's local currency payment information, sending a currency exchange instruction to the exchange partner; remitting the local currency amount to the exchange partner, and receiving the foreign exchange amount remitted by the exchange partner; remitting the foreign exchange amount to the seller. Through the technical solution provided by the embodiment of the present invention, the cross-border payment platform acts as an intermediary to help the buyer complete the processing of exchange lock, exchange and foreign currency payment, so that the buyer can easily realize online exchange lock, and can complete the payment in local currency after the exchange lock without paying attention to the intermediate links, simplifying the complexity of cross-border payment, and through the cross-border payment platform to carry out exchange lock processing, it can give buyers more ample payment time.

[0005] However, the above technical solution only provides the specific processing procedure and simplified mechanism of each cross-border payment process. It is impossible for the payment operation company with a cross-border payment license to complete the online identification of whether the cross-border payment involves cross-border high-risk from the server side when each cross-border payment is initiated. As a result, the identification of whether the cross-border payment involves cross-border high-risk flows to the subsequent buyer customers of the payment operation company. Once the buyer customer identification fails, it is easy to cause serious troubles and huge economic losses to the buyer customer. At the same time, the subsequent buyer customer identification also brings a heavy data burden to the data processing of the entire cross-border payment. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a cross-border payment processing method and system, which uses an artificial intelligence model with a customized structure on the server side of a payment operating company to complete intelligent identification of whether a current cross-border payment request is a cross-border high-risk payment request based on a fully and comprehensively selected number of basic information, and when it is identified that the current cross-border payment request is a cross-border high-risk payment request, the current cross-border payment request is rejected, thereby completing intelligent identification and emergency processing of high-risk behaviors of high-risk cross-border payments on the server side, avoiding unnecessary troubles to the buyer customers of the payment operating company, and avoiding economic losses to the buyer customers of the payment operating company. At the same time, the identification processing on the server side also reduces the data processing volume of the entire cross-border payment compared to the subsequent identification processing of the buyer customers.

[0007] According to a first aspect of the present invention, a cross-border payment processing method is provided, which runs on a server of a payment operation company, and the method comprises: Obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment. Each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; Get all the associated information of the seller and the buyer in the current cross-border payment request; Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; A cross-border payment intelligent identification model is used to intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

[0008] According to a second aspect of the present invention, a cross-border payment processing system is provided, which is located at a server end of a payment operation company, and the system includes a memory and a plurality of processors, wherein the memory stores a computer program, and the computer program is configured to be executed by the plurality of processors to complete the following steps: Obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment. Each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; Get all the associated information of the seller and the buyer in the current cross-border payment request; Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; A cross-border payment intelligent identification model is used to intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

[0009] According to a third aspect of the present invention, a cross-border payment processing system is provided, which is located on a server side of a payment operation company, and the system comprises: The first analysis device is used to obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each sales data is the sales commodity type code data, the sales commodity quantity, the total sales commodity amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment; The second analysis device is used to obtain each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request; A target component is used to perform each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and output it as a cross-border payment intelligent identification model, and the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; An identification processing device, connected to the first analysis device, the second analysis device and the target assembly device respectively, for using a cross-border payment intelligent identification model to intelligently identify a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller association information in the current cross-border payment request, and various pieces of buyer association information; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

[0010] Compared with the prior art, the present invention has at least the following five outstanding substantive features: Substantive feature A: On the server side of the payment operating company, a customized artificial intelligence model is used to complete the intelligent identification of whether the current cross-border payment request is a cross-border high-risk payment request based on a fully and comprehensively selected number of basic information, and when the current cross-border payment request is identified as a cross-border high-risk payment request, the current cross-border payment request is rejected, thereby completing the intelligent identification and emergency processing of high-risk behaviors of high-risk cross-border payments on the server side, avoiding unnecessary troubles to the buyer customers of the payment operating company and avoiding economic losses to the buyer customers of the payment operating company; Substantive feature B: An artificial intelligence model used to intelligently identify whether the current cross-border payment request involves cross-border high-risk, and the structural customization involves the following aspects: the artificial intelligence model is a cross-border payment intelligent identification model, and the cross-border payment intelligent identification model is a convolutional neural network after various training operations. What is particularly critical is that the number of training operations of the convolutional neural network is positively correlated with the total number of merchants currently managed by the payment operating company, so that cross-border payment intelligent identification models with different structures are designed for different payment operating companies, ensuring the reliability and stability of the intelligent identification results; Substantive Feature C: For the intelligent identification of whether the current cross-border payment request involves cross-border high risk, a fully and comprehensively selected number of basic information is introduced. The multiple basic information specifically includes multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer. The full and comprehensive selection of the above multiple basic information further ensures the reliability and stability of the intelligent identification results; Substantive feature D: More specifically, the sales data corresponding to each historical cross-border payment completed by the seller in the current cross-border payment request before the current moment is the sales commodity type code data, the quantity of sales commodities, the total sales commodity amount and the time consumed from the moment the payment request is received to the moment the payment is completed in the historical cross-border payment. The seller / buyer's various pieces of associated information are the seller / buyer's registration time, the number of payments completed before the current moment and the historical average payment amount; Substantive Feature E: In each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request was received by the server of the payment operating company, the various related information of the seller in the certain historical cross-border payment request, and the various related information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network, thereby ensuring the training effect of each training operation performed on the convolutional neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 Schematic diagram of the working scenario of the cross-border payment processing method and system according to the present invention.

[0012] Figure 2 The present invention is a flowchart of the steps of a cross-border payment processing method according to Embodiment 1 of the present invention.

[0013] Figure 3 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 2 of the present invention.

[0014] Figure 4 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 3 of the present invention.

[0015] Figure 5 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 4 of the present invention.

[0016] Figure 6 It is a schematic diagram of the structure of a cross-border payment processing system according to Embodiment 5 of the present invention.

[0017] Figure 7 It is a schematic diagram of the structure of a cross-border payment processing system according to Embodiment 6 of the present invention. DETAILED DESCRIPTION

[0018] like Figure 1 As shown, a schematic diagram of the working scenario of the cross-border payment processing method and system according to the present invention is given.

[0019] The specific technical process of the present invention is as follows: Technical process 1: On the server side of the payment operator, a customized artificial intelligence model is used to intelligently identify whether the current cross-border payment request involves cross-border high-risk payment; like Figure 1 As shown, for example, the seller and buyer in the current cross-border payment request are Figure 1 Seller A and Buyer A, Figure 1 The clients corresponding to Seller A, Buyer A, Buyer B, Buyer C and Buyer D are controlled by the same server of the payment operation company. Specifically, the structural customization of the artificial intelligence model is mainly reflected in the following aspects: First: the artificial intelligence model is a cross-border payment intelligent identification model, and the cross-border payment intelligent identification model is a convolutional neural network after various training operations; Second: In the cross-border payment intelligent identification model, the number of training operations of the convolutional neural network is positively correlated with the total number of merchants currently managed by the payment operating company, so that different structures of cross-border payment intelligent identification models can be designed for different payment operating companies; like Figure 1 As shown, cross-border payment intelligent authentication models with different structures are designed for servers of different payment operating companies; Third: In each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, various pieces of associated information of the seller in the certain historical cross-border payment request, and various pieces of associated information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network, thereby ensuring the training effect of each training operation performed on the convolutional neural network; In this way, through the design of the above-mentioned customized structures of the artificial intelligence model, the reliability and stability of the intelligent identification results are guaranteed; Technical process 2: On the server side of the payment operator, multiple basic information is screened for intelligent identification of whether the current cross-border payment request involves cross-border high-risk payment; For example, the multiple pieces of basic information specifically include multiple pieces of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer; By way of further example, the sales data corresponding to each historical cross-border payment completed by the seller in the current cross-border payment request before the current moment is the sales commodity type code data, the quantity of sales commodities, the total sales commodity amount, and the time consumed from the moment the payment request is received to the moment the payment is completed in the historical cross-border payment; By way of further example, each piece of associated information of the seller / buyer is the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount; In this way, through the full and comprehensive selection of the above-mentioned multiple basic information, the reliability and stability of the intelligent identification results are further guaranteed; Technical process three: On the server side of the payment operation company, the cross-border payment intelligent identification model with customized structure designed in technical process one is used to intelligently identify whether the current cross-border payment request involves cross-border high-risk payment based on multiple basic data fully and comprehensively screened in technical process two; Specifically, the cross-border payment intelligent identification model outputs a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request; Technical process 4: On the server side of the payment operation company, based on the intelligent identification results of technical process 3, complete the dynamic payment processing of the current cross-border payment request; For example, on the server side of the payment operation company, when the received request identification identifier indicates that the current cross-border payment request belongs to a cross-border high-risk payment request, the current cross-border payment request is rejected; and when the received request identification identifier indicates that the current cross-border payment request does not belong to a cross-border high-risk payment request, the current cross-border payment request is approved; It can be seen that with the collaboration and sequential execution of the above-mentioned technical processes, an artificial intelligence model with a customized structure can be used on the server side of the payment operating company to complete the intelligent identification of whether the current cross-border payment request is a cross-border high-risk payment request based on a fully and comprehensively selected number of basic information, and when the current cross-border payment request is identified as a cross-border high-risk payment request, the current cross-border payment request will be rejected, thereby completing the intelligent identification and emergency processing of high-risk behaviors of high-risk cross-border payments on the server side, thereby avoiding unnecessary troubles to the buyer customers of the payment operating company and avoiding economic losses to the buyer customers of the payment operating company.

[0020] The key points of the present invention are: completing intelligent identification of high-risk cross-border payments on the server side to replace the subjective identification of high-risk cross-border payments completed on the buyer's client side, multiple customized structural designs of cross-border payment intelligent identification models for intelligent identification, sufficient and comprehensive screening of multiple basic data for intelligent identification, and dynamic payment processing of current cross-border payment requests based on intelligent identification results.

[0021] The cross-border payment processing method and system of the present invention will be specifically described below by way of embodiments.

[0022] Example 1 Figure 2 The present invention is a flowchart of the steps of a cross-border payment processing method according to Embodiment 1 of the present invention.

[0023] like Figure 2 As shown, the cross-border payment processing method is run on the server side of the payment operation company, and the cross-border payment processing method includes the following specific steps: Step 201: used to obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; For example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each of which is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. The current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to the same type. In this case, the sales commodity type coding data in the historical cross-border payment is the type coding data corresponding to the same type; Continuing with the example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each sales data is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. It also includes: the current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to multiple different types. In this case, the sales commodity type coding data in the historical cross-border payment is the coding data obtained by connecting the multiple type coding data corresponding to the multiple different types end to end; In order to normalize the data, the sales commodity type code data in the historical cross-border payment can be a binary value with a set number of digits. If the set number of digits is not met, the remaining digits are padded with zeros to ensure that the binary value with the set number of digits represents the sales commodity type code data; Step 202: Obtain each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request; For example, obtaining each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request includes: generally, there is only one seller in the current cross-border payment request, and there is only one buyer in the current cross-border payment request; Step 203: performing each training operation on the convolutional neural network to obtain a convolutional neural network after each training operation and outputting it as a cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; Specifically, performing each training operation on the convolutional neural network to obtain a convolutional neural network after each training operation and outputting it as a cross-border payment intelligent identification model, the number of training operations being positively correlated with the total number of merchants currently managed by the payment operating company includes: when the total number of merchants currently managed by the payment operating company is in the millions, the number of training operations selected is 2,000 times, when the total number of merchants currently managed by the payment operating company is in the five million level, the number of training operations selected is 3,000 times, when the total number of merchants currently managed by the payment operating company is in the tens of millions, the number of training operations selected is 4,000 times, and when the total number of merchants currently managed by the payment operating company is in the fifty million level, the number of training operations selected is 5,000 times, and so on; Step 204: Using a cross-border payment intelligent identification model, intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller-related information in the current cross-border payment request, and various pieces of buyer-related information; For example, a cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer. The request identification identifier includes: request identification identifiers with different values ​​are used to respectively indicate whether the current cross-border payment request belongs to a cross-border high-risk payment request; The seller / buyer's associated information includes the seller / buyer's registration time, the number of payments completed before the current moment, and the historical average payment amount; The number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operation company. The number of service areas of the payment operation company is the sum of the number of countries served by the payment operation company and the number of service areas. For example, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operating company, and the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas. The number includes: when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 8, when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 60, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 12, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 80, and the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 16, and so on; Among them, in each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, various pieces of associated information of the seller in the certain historical cross-border payment request, and various pieces of associated information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network; Among them, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information. The request identification identifier includes: performing binary value conversion processing on the multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model; And wherein, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various related information of the seller in the current cross-border payment request, and various related information of the buyer. The request identification identifier also includes: running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model indicating whether the current cross-border payment request is a cross-border high-risk payment request.

[0024] Example 2 Figure 3 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 2 of the present invention.

[0025] like Figure 3 As shown, Figure 2 Different from the embodiment in, after using the cross-border payment intelligent identification model to intelligently identify the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of seller's associated information in the current cross-border payment request, and each piece of buyer's associated information, that is, after step S204, the method further includes: Step S205: when the received request identification mark indicates that the current cross-border payment request is a cross-border high-risk payment request, reject the current cross-border payment request; when the received request identification mark indicates that the current cross-border payment request is not a cross-border high-risk payment request, approve the current cross-border payment request; For example, when the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the current cross-border payment request is rejected; when the received request identification identifier indicates that the current cross-border payment request is not a cross-border high-risk payment request, the current cross-border payment request is passed, including: after rejecting the current cross-border payment request, the seller in the current cross-border payment request is marked as a blacklisted seller.

[0026] Example 3 Figure 4 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 3 of the present invention.

[0027] like Figure 4 As shown, Figure 2Different from the embodiment in, after using the cross-border payment intelligent identification model to intelligently identify the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of seller's associated information in the current cross-border payment request, and each piece of buyer's associated information, that is, after step S204, the method further includes: Step S206: receiving a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request, and displaying in real time the request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request; For example, receiving a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request, and displaying in real time the request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request includes: selecting an LED display array to receive a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request, and displaying in real time the request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request.

[0028] Example 4 Figure 5 The present invention is a flowchart showing the steps of a cross-border payment processing method according to Embodiment 4 of the present invention.

[0029] like Figure 5 As shown, Figure 2 Different from the embodiment in, after using the cross-border payment intelligent identification model to intelligently identify the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of seller's associated information in the current cross-border payment request, and each piece of buyer's associated information, that is, after step S204, the method further includes: Step S207: when the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the request number of the current cross-border payment request and the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request are packaged together into the same network data packet, and the network data packet is sent to the financial payment terminal of the buyer in the current cross-border payment request using a wireless communication link; Specifically, when the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the request number of the current cross-border payment request and the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request are packaged together in the same network data packet, and the network data packet is sent to the financial payment terminal of the buyer in the current cross-border payment request using a wireless communication link. The method includes: when the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the request number of the current cross-border payment request and the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request are packaged together in the same IP data packet, and the IP data packet is sent to the financial payment terminal of the buyer in the current cross-border payment request using a time-division duplex communication link.

[0030] Next, various method embodiments of the present invention are described in detail.

[0031] In the cross-border payment processing method according to various method embodiments of the present invention: After performing binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model, the process includes: using a numerical conversion device to perform binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information; For example, the digital value conversion device is used to complete the binary digital value conversion processing of the multiple sales data corresponding to the multiple historical cross-border payments completed by the seller before the current moment in the current cross-border payment request, the various pieces of seller's associated information in the current cross-border payment request, and the various pieces of buyer's associated information, respectively, including: the digital value conversion device can be programmed and designed in VHDL language; Among them, performing binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model also includes: using a synchronous driving device connected to the numerical conversion device, which is used to synchronously input multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment after performing binary numerical conversion processing, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information into the cross-border payment intelligent identification model; Among them, a synchronous driving device connected to a numerical conversion device is used to synchronously input multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer into the cross-border payment intelligent identification model after performing binary numerical conversion processing respectively, including: using different programmable logic devices to respectively realize the numerical conversion device and the synchronous driving device; And wherein, running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model indicating whether the current cross-border payment request is a cross-border high-risk payment request includes: the request identification identifier is a binary value representation.

[0032] And in the cross-border payment processing method according to each method embodiment of the present invention: Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, the number of training operations being positively correlated with the total number of merchants currently managed by the payment operation company includes: using an information mapping formula to represent the information mapping relationship of the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operation company; Specifically, the information mapping relationship using the information mapping formula to express the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operation company includes: the MATLAB toolbox can be selected to complete the testing and simulation of the information mapping formula; The information mapping relationship of using an information mapping formula to express the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operation company includes: in the information mapping formula, the total number of merchants currently managed by the payment operation company is used as an input parameter of the information mapping formula; Among them, the information mapping relationship that uses the information mapping formula to represent the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operating company also includes: in the information mapping formula, the number of training operations that is positively correlated with the total number of merchants currently managed by the payment operating company is used as an output parameter of the information mapping formula.

[0033] Example 5 Figure 6 It is a schematic diagram of the structure of a cross-border payment processing system according to Embodiment 5 of the present invention.

[0034] like Figure 6 As shown, the cross-border payment processing system is located at the server side of the payment operation company, and includes a memory and multiple processors. The memory stores a computer program, and the computer program is configured to be executed by the multiple processors to complete the following steps: Step 201: used to obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; For example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each of which is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. The current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to the same type. In this case, the sales commodity type coding data in the historical cross-border payment is the type coding data corresponding to the same type; Continuing with the example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each sales data is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. It also includes: the current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to multiple different types. In this case, the sales commodity type coding data in the historical cross-border payment is the coding data obtained by connecting the multiple type coding data corresponding to the multiple different types end to end; In order to normalize the data, the sales commodity type code data in the historical cross-border payment can be a binary value with a set number of digits. If the set number of digits is not met, the remaining digits are padded with zeros to ensure that the binary value with the set number of digits represents the sales commodity type code data; Step 202: Obtain each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request; For example, obtaining each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request includes: generally, there is only one seller in the current cross-border payment request, and there is only one buyer in the current cross-border payment request; Step 203: performing each training operation on the convolutional neural network to obtain a convolutional neural network after each training operation and outputting it as a cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; Specifically, performing each training operation on the convolutional neural network to obtain a convolutional neural network after each training operation and outputting it as a cross-border payment intelligent identification model, the number of training operations being positively correlated with the total number of merchants currently managed by the payment operating company includes: when the total number of merchants currently managed by the payment operating company is in the millions, the number of training operations selected is 2,000 times, when the total number of merchants currently managed by the payment operating company is in the five million level, the number of training operations selected is 3,000 times, when the total number of merchants currently managed by the payment operating company is in the tens of millions, the number of training operations selected is 4,000 times, and when the total number of merchants currently managed by the payment operating company is in the fifty million level, the number of training operations selected is 5,000 times, and so on; Step 204: Using a cross-border payment intelligent identification model, intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller-related information in the current cross-border payment request, and various pieces of buyer-related information; For example, a cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer. The request identification identifier includes: request identification identifiers with different values ​​are used to respectively indicate whether the current cross-border payment request belongs to a cross-border high-risk payment request; The seller / buyer's associated information includes the seller / buyer's registration time, the number of payments completed before the current moment, and the historical average payment amount; The number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operation company. The number of service areas of the payment operation company is the sum of the number of countries served by the payment operation company and the number of service areas. For example, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operating company, and the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas. The number includes: when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 8, when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 60, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 12, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 80, and the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 16, and so on; Among them, in each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, various pieces of associated information of the seller in the certain historical cross-border payment request, and various pieces of associated information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network; Among them, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information. The request identification identifier includes: performing binary value conversion processing on the multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model; And wherein, using a cross-border payment intelligent identification model to intelligently identify whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information, the request identification identifier also includes: running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model indicating whether the current cross-border payment request is a cross-border high-risk payment request; like Figure 6 As shown, exemplarily, N processors are provided, and the N processors are located at the wireless routing end of the target residential user and the remote full-media content server, where N is a natural number greater than or equal to 1.

[0035] Example 6 Figure 7 It is a schematic diagram of the structure of a cross-border payment processing system according to Embodiment 6 of the present invention.

[0036] like Figure 7 As shown, the cross-border payment processing system is located on the server side of the payment operation company, and the cross-border payment processing system includes the following components: The first analysis device is used to obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each sales data is the sales commodity type code data, the sales commodity quantity, the total sales commodity amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment; For example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each of which is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. The current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to the same type. In this case, the sales commodity type coding data in the historical cross-border payment is the type coding data corresponding to the same type; Continuing with the example, multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment are obtained, each sales data is the sales commodity type coding data, the sales commodity quantity, the sales commodity total amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment. It also includes: the current cross-border payment request may include multiple sales commodities that need to be paid at the same time, and the multiple sales commodities belong to multiple different types. In this case, the sales commodity type coding data in the historical cross-border payment is the coding data obtained by connecting the multiple type coding data corresponding to the multiple different types end to end; In order to normalize the data, the sales commodity type code data in the historical cross-border payment can be a binary value with a set number of digits. If the set number of digits is not met, the remaining digits are padded with zeros to ensure that the binary value with the set number of digits represents the sales commodity type code data; The second analysis device is used to obtain each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request; For example, obtaining each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request includes: generally, there is only one seller in the current cross-border payment request, and there is only one buyer in the current cross-border payment request; A target component is used to perform each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and output it as a cross-border payment intelligent identification model, and the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; Specifically, performing each training operation on the convolutional neural network to obtain a convolutional neural network after each training operation and outputting it as a cross-border payment intelligent identification model, the number of training operations being positively correlated with the total number of merchants currently managed by the payment operating company includes: when the total number of merchants currently managed by the payment operating company is in the millions, the number of training operations selected is 2,000 times, when the total number of merchants currently managed by the payment operating company is in the five million level, the number of training operations selected is 3,000 times, when the total number of merchants currently managed by the payment operating company is in the tens of millions, the number of training operations selected is 4,000 times, and when the total number of merchants currently managed by the payment operating company is in the fifty million level, the number of training operations selected is 5,000 times, and so on; An identification processing device, connected to the first analysis device, the second analysis device and the target assembly device respectively, for using a cross-border payment intelligent identification model to intelligently identify a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller association information in the current cross-border payment request, and various pieces of buyer association information; For example, a cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer. The request identification identifier includes: request identification identifiers with different values ​​are used to respectively indicate whether the current cross-border payment request belongs to a cross-border high-risk payment request; The seller / buyer's associated information includes the seller / buyer's registration time, the number of payments completed before the current moment, and the historical average payment amount; The number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operation company. The number of service areas of the payment operation company is the sum of the number of countries served by the payment operation company and the number of service areas. For example, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operating company, and the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas. The number includes: when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 8, when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 60, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 12, and when the number of service areas of the payment operating company is the sum of the number of countries served by the payment operating company and the number of service areas, the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 80, and the number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is 16, and so on; Among them, in each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, various pieces of associated information of the seller in the certain historical cross-border payment request, and various pieces of associated information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network; Among them, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information. The request identification identifier includes: performing binary value conversion processing on the multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model; And wherein, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various related information of the seller in the current cross-border payment request, and various related information of the buyer. The request identification identifier also includes: running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model indicating whether the current cross-border payment request is a cross-border high-risk payment request.

[0037] In addition, the present invention may also cite the following technical contents to further demonstrate the outstanding substantial progress of the present invention: In each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, each piece of associated information of the seller in the certain historical cross-border payment request, and each piece of associated information of the buyer are used as multiple input contents of the convolutional neural network. The training operation performed on the convolutional neural network includes: the known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is represented in a binary value form; And wherein, in each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, each piece of associated information of the seller in the certain historical cross-border payment request, and each piece of associated information of the buyer are used as multiple input contents of the convolutional neural network, and completing this training operation performed on the convolutional neural network also includes: the multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, each piece of associated information of the seller in the certain historical cross-border payment request, and each piece of associated information of the buyer are respectively represented in the form of binary values; For example, a numerical simulation mode can be selected to execute the known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request as the single output content of the convolutional neural network, and use the multiple sales data corresponding to the multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the time when the certain historical cross-border payment request is received by the server of the payment operating company, the various related information of the seller in the certain historical cross-border payment request, and the various related information of the buyer as the multiple input contents of the convolutional neural network to complete the simulation and test of the data processing process of this training operation performed by the convolutional neural network.

[0038] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device / electronic device / computer-readable storage medium / computer program product embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The above is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A cross-border payment processing method, running on the server side of a payment operation company, characterized in that: The method comprises: Obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment. Each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; Get all the associated information of the seller and the buyer in the current cross-border payment request; Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; A cross-border payment intelligent identification model is used to intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

2. The cross-border payment processing method according to claim 1, characterized in that: The number of historical cross-border payments completed by the seller in the current cross-border payment request before the current moment is proportional to the number of service areas of the payment operation company. The number of service areas of the payment operation company is the sum of the number of countries served by the payment operation company and the number of service areas. Among them, in each training operation performed on the convolutional neural network, a known request identification identifier indicating whether a certain historical cross-border payment request is a cross-border high-risk payment request is used as a single output content of the convolutional neural network, and multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the certain historical cross-border payment request before the moment when the certain historical cross-border payment request is received by the server of the payment operating company, various related information of the seller in the certain historical cross-border payment request, and various related information of the buyer are used as multiple input contents of the convolutional neural network to complete this training operation performed on the convolutional neural network.

3. The cross-border payment processing method according to claim 2, characterized in that: A cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request belongs to a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller before the current moment in the current cross-border payment request, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information. The request identification identifier includes: performing binary value conversion processing on the multiple sales data corresponding to multiple historical cross-border payments completed by the seller before the current moment in the current cross-border payment request, various pieces of seller's associated information in the current cross-border payment request, and various pieces of buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model; Among them, the cross-border payment intelligent identification model is used to intelligently identify whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various related information of the seller in the current cross-border payment request, and various related information of the buyer. The request identification identifier also includes: running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model indicating whether the current cross-border payment request is a cross-border high-risk payment request.

4. The cross-border payment processing method according to claim 3, characterized in that: After intelligently identifying a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer using a cross-border payment intelligent identification model, the method further includes: When the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the current cross-border payment request is rejected. When the received request identification identifier indicates that the current cross-border payment request is not a cross-border high-risk payment request, the current cross-border payment request is approved.

5. The cross-border payment processing method according to claim 3, characterized in that: After intelligently identifying a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer using a cross-border payment intelligent identification model, the method further includes: A request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request is received, and the request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request is displayed in real time.

6. The cross-border payment processing method according to claim 3, characterized in that: After intelligently identifying a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer using a cross-border payment intelligent identification model, the method further includes: When the received request identification identifier indicates that the current cross-border payment request is a cross-border high-risk payment request, the request number of the current cross-border payment request and the request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request are packaged together into the same network data packet, and the network data packet is sent to the financial payment terminal of the buyer in the current cross-border payment request using a wireless communication link.

7. The cross-border payment processing method according to any one of claims 3 to 6, characterized in that: After performing binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model, the process includes: using a numerical conversion device to perform binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information; Among them, performing binary numerical conversion processing on multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information, and then synchronously inputting them into the cross-border payment intelligent identification model also includes: using a synchronous driving device connected to the numerical conversion device, which is used to synchronously input multiple copies of sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment after performing binary numerical conversion processing, each copy of the seller's associated information in the current cross-border payment request, and each copy of the buyer's associated information into the cross-border payment intelligent identification model; Among them, a synchronous driving device connected to a numerical conversion device is used to synchronously input multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each piece of associated information of the seller in the current cross-border payment request, and each piece of associated information of the buyer into the cross-border payment intelligent identification model after performing binary numerical conversion processing respectively, including: using different programmable logic devices to respectively realize the numerical conversion device and the synchronous driving device; Among them, running the cross-border payment intelligent identification model to obtain the request identification identifier output by the cross-border payment intelligent identification model, which indicates whether the current cross-border payment request is a cross-border high-risk payment request, includes: the request identification identifier is a binary value representation.

8. The cross-border payment processing method according to any one of claims 3 to 6, characterized in that: Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, the number of training operations being positively correlated with the total number of merchants currently managed by the payment operation company includes: using an information mapping formula to represent the information mapping relationship of the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operation company; The information mapping relationship of using an information mapping formula to express the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operation company includes: in the information mapping formula, the total number of merchants currently managed by the payment operation company is used as an input parameter of the information mapping formula; Among them, the information mapping relationship that uses the information mapping formula to represent the positive correlation between the number of training operations and the total number of merchants currently managed by the payment operating company also includes: in the information mapping formula, the number of training operations that is positively correlated with the total number of merchants currently managed by the payment operating company is used as an output parameter of the information mapping formula.

9. A cross-border payment processing system located on the server side of a payment operation company, characterized in that: The system comprises a memory and a plurality of processors, wherein the memory stores a computer program, and the computer program is configured to be executed by the plurality of processors to complete the following steps: Obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment. Each sales data is the sales commodity type code data, sales commodity quantity, sales commodity total amount and the time consumed from the payment request receiving moment to the payment completion moment in the corresponding historical cross-border payment; Get all the associated information of the seller and the buyer in the current cross-border payment request; Performing each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and outputting it as the cross-border payment intelligent identification model, wherein the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; A cross-border payment intelligent identification model is used to intelligently identify a request identification mark indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of associated information of the seller in the current cross-border payment request, and various pieces of associated information of the buyer; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

10. A cross-border payment processing system, located on the server side of a payment operation company, characterized in that: The system comprises: The first analysis device is used to obtain multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, each sales data is the sales commodity type code data, the sales commodity quantity, the total sales commodity amount and the time consumed from the payment request reception moment to the payment completion moment in the corresponding historical cross-border payment; The second analysis device is used to obtain each piece of associated information of the seller and each piece of associated information of the buyer in the current cross-border payment request; A target component is used to perform each training operation on the convolutional neural network to obtain the convolutional neural network after each training operation and output it as a cross-border payment intelligent identification model, and the number of training operations is positively correlated with the total number of merchants currently managed by the payment operation company; An identification processing device, connected to the first analysis device, the second analysis device and the target assembly device respectively, for using a cross-border payment intelligent identification model to intelligently identify a request identification identifier indicating whether the current cross-border payment request is a cross-border high-risk payment request based on multiple sales data corresponding to multiple historical cross-border payments completed by the seller in the current cross-border payment request before the current moment, various pieces of seller association information in the current cross-border payment request, and various pieces of buyer association information; The seller / buyer's associated information includes the seller / buyer's registration duration, the number of payments completed before the current moment, and the historical average payment amount.

Images