Risk control method and device for payment transactions based on blockchain

Abstract

The present invention discloses a risk control method and device for payment transactions based on blockchain, relating to the field of blockchain technology. The method comprises: classifying a bank's historical payment data within a predetermined time range to obtain multiple payment data categories; determining the risk probability and corresponding probability convergence value of each payment data category in various dimensions; selecting a low-risk payment data category from the multiple payment data categories based on the risk probability and probability convergence value; screening the low-risk payment data from a customer's historical payment data; identifying the counterparty corresponding to the low-risk payment data as the customer's safe counterparty and recommending it to the customer. After the customer digitally signs and confirms the information, the information is uploaded to the blockchain; and distributing the safe counterparty to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal falls below a specified strength value, risk control is performed on the customer's transactions based on the safe counterparty stored on the mobile terminal. The present invention can improve customer experience.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and in particular to a risk control method and device for payment transactions based on blockchain. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. No statement herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] Because banks' risk control functions reside on their servers, customers must be connected to the internet to conduct transactions. Poor network signals prevent transactions. Some users may frequently be in specific locations at specific times, where weak wireless signals can prevent them from using their terminals to conduct transactions, resulting in a poor customer experience. Summary of the Invention

[0004] An embodiment of the present invention provides a risk control method for payment transactions based on blockchain, the method comprising:

[0005] Obtain historical payment data of the bank within a predetermined time range;

[0006] Classify the acquired historical payment data to obtain multiple payment data categories;

[0007] For each payment data category, determine the risk probability of the payment data category in each dimension, as well as the probability convergence value corresponding to each risk probability;

[0008] Selecting a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability;

[0009] For each customer, filter out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank;

[0010] Determine the counterparty corresponding to the low-risk payment data of the customer as the safe counterparty corresponding to the customer;

[0011] Recommend the client's corresponding security counterparty to the client, and upload it to the blockchain after the client's digital signature confirmation;

[0012] The security counterparty corresponding to the customer is sent to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than the specified strength value, risk control of the customer's transaction is performed based on the security counterparty stored on the mobile terminal.

[0013] An embodiment of the present invention further provides a risk control device for payment transactions based on blockchain, the device comprising:

[0014] A data acquisition module, used to obtain the bank's historical payment data within a predetermined time range;

[0015] A classification module is used to classify the acquired historical payment data to obtain multiple payment data categories;

[0016] A risk probability determination module is used to determine, for each payment data category, the risk probability of the payment data category in each dimension, and the probability convergence value corresponding to each risk probability;

[0017] a category selection module, configured to select a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability;

[0018] A screening module is used to screen low-risk payment data corresponding to each customer from the customer's historical payment data at the bank;

[0019] A safe counterparty determination module, configured to determine a counterparty corresponding to the low-risk payment data corresponding to the customer as the safe counterparty corresponding to the customer;

[0020] The recommendation upload module is used to recommend the client's corresponding security counterparty to the client, and upload it to the blockchain after the client's digital signature is confirmed;

[0021] The sending module is used to send the security counterparty corresponding to the customer to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than the specified strength value, risk control is performed on the customer's transaction based on the security counterparty stored on the mobile terminal.

[0022] An embodiment of the present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the risk control method for payment transactions based on blockchain is implemented.

[0023] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned risk control method for blockchain-based payment transactions.

[0024] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned risk control method for blockchain-based payment transactions.

[0025] In an embodiment of the present invention, compared with the technical solution in the prior art in which customers cannot trade when the network signal is weak, resulting in a poor customer experience, the method obtains historical payment data of a bank within a predetermined time range; classifies the obtained historical payment data to obtain multiple payment data categories; for each payment data category, determines the risk probability of the payment data category in each dimension and the probability convergence value corresponding to each risk probability; selects a low-risk payment data category from the multiple payment data categories based on the risk probability in each dimension and the probability convergence value corresponding to each risk probability; for each customer, filters out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank; determines the counterparty corresponding to the low-risk payment data corresponding to the customer as the customer's corresponding secure counterparty; recommends the customer's corresponding secure counterparty to the customer, and uploads it to the blockchain after the customer's digital signature is confirmed; and sends the customer's corresponding secure counterparty to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than a specified strength value, risk control is performed on the customer's transactions based on the secure counterparty stored on the mobile terminal. Transactions can be conducted even when the network signal is weak, thereby improving the customer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0027] Figure 1 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 1 ;

[0028] Figure 2 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 2 ;

[0029] Figure 3 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 3 ;

[0030] Figure 4 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 4 ;

[0031] Figure 5 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 5 ;

[0032] Figure 6 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 6 ;

[0033] Figure 7 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 7 ;

[0034] Figure 8 This is a structural block diagram of a risk control device for payment transactions based on blockchain in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0036] Figure 1 The risk control method process of the payment transaction based on blockchain in the embodiment of the present invention Figure 1 ,like Figure 1 As shown, the method includes:

[0037] Step 101: Obtain the bank's historical payment data within a predetermined time range;

[0038] Step 102: Classify the acquired historical payment data to obtain multiple payment data categories;

[0039] Step 103: For each payment data category, determine the risk probability of the payment data category in each dimension, and the probability convergence value corresponding to each risk probability;

[0040] Step 104: selecting a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability;

[0041] Step 105: For each customer, filter out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank;

[0042] Step 106: Determine the counterparty corresponding to the low-risk payment data corresponding to the customer as the safe counterparty corresponding to the customer;

[0043] Step 107: Recommend the client's corresponding security counterparty to the client. After the client digitally signs and confirms, upload the data to the blockchain.

[0044] Step 108: The security counterparty corresponding to the customer is sent to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than a specified strength value, risk control is performed on the customer's transaction based on the security counterparty stored on the mobile terminal.

[0045] Specifically, dimensions are used to categorize transaction data, such as transaction scenario, business category, transaction time, and transaction channel. Because customer transactions may pose security risks for both transactions and funds, it's necessary to determine the risk probability of transactions. This can be determined using different dimensions. Risk varies across different dimensions, and by calculating risk probabilities across these dimensions, we can determine the security of each customer relationship category.

[0046] In the embodiment of the present invention, Figure 2 As shown, step 102 classifies the acquired historical payment data to obtain multiple payment data categories, including:

[0047] Step 201: For each historical payment data, extract the payer, counterparty and payment amount level of the historical payment data;

[0048] Step 202: Determine, based on the payer's historical transaction data, a payment direction quantity corresponding to the historical payment data, wherein each component of the payment direction quantity corresponds to a transaction category on a one-to-one basis, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the payer's historical transaction data;

[0049] Step 203: Determine the counterparty quantity corresponding to the historical payment data based on the counterparty's historical transaction data, wherein each component of the counterparty quantity corresponds to a transaction category on a one-to-one basis, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data;

[0050] Step 204: Determine a payer distance function and a counterparty distance function based on the payment direction quantity and the counterparty direction quantity, wherein for any two payment data, the function value of the payer distance function corresponding to the two payment data is the distance between the payment direction quantities corresponding to the two payment data, and the function value of the counterparty distance function corresponding to the two payment data is the distance between the counterparty direction quantities corresponding to the two payment data;

[0051] Step 205: Cluster the acquired historical payment data based on the payer distance function, the counterparty distance function, and the payment amount level to obtain multiple payment data categories.

[0052] In one embodiment, the acquired historical payment data is clustered based on the payer distance function, the counterparty distance function, and the payment amount level to obtain multiple payment data categories, including:

[0053] Determine the distance function corresponding to the payment data based on the payer distance function and the counterparty distance function;

[0054] Based on the distance function corresponding to the payment data and the payment amount level, the acquired historical payment data is clustered to obtain multiple payment data categories (for example, K-means is selected, or the payment amount level is used as the category identifier, and learning vector quantization is selected to cluster the acquired historical payment data).

[0055] In one embodiment, the historical payment data is clustered based on the distance function corresponding to the payment data and the payment amount level to obtain multiple payment data categories, including:

[0056] 1. Select multiple historical payment data from the acquired historical payment data as payment data category centers, each payment data category center corresponds to a payment data category, and the initial element of the payment data category only includes the corresponding payment data category center;

[0057] 2. For each historical payment data obtained, perform the following steps:

[0058] Calculating the distance between each payment data category center and the historical payment data based on the distance function corresponding to the payment data, and determining the distance as the distance corresponding to the payment data category center;

[0059] Selecting multiple payment data category centers that are consistent with the payment amount level corresponding to the historical payment data from all payment data category centers;

[0060] The minimum value of the distances corresponding to the selected payment data category centers is used as the distance S corresponding to the historical payment data, and the payment data category center corresponding to the minimum value is used as the payment data category center corresponding to the historical payment data; the minimum value of the distances corresponding to the payment data category centers that are not selected is used as the distance T corresponding to the historical payment data;

[0061] If the difference between the corresponding distance T and the corresponding distance S is greater than a specified threshold, a new payment data category center is created based on the historical payment data. The newly created payment data category center corresponds to a new payment data category, and the initial elements of the new payment data category only contain the historical payment data. Otherwise, the historical payment data is classified into the payment data category corresponding to the payment data category center corresponding to the historical payment data.

[0062] 3. After executing the above steps for all historical payment data, for each payment data category, update the payment direction quantity corresponding to the payment data category center corresponding to the payment data category to the average of the payment direction quantities corresponding to all historical payment data of the payment data category, and update the opponent direction quantity corresponding to the payment data category center corresponding to the payment data category to the average of the opponent direction quantities corresponding to all historical payment data of the payment data category; update the payment amount level corresponding to the payment data category center corresponding to the payment data category to the data value with the largest number among the data values of the payment amount level of all historical payment data of the payment data category;

[0063] 4. Repeat step 2 for each historical payment data and step 3 for each payment data category until the changes in the payment direction quantities and the corresponding opponent direction quantities corresponding to the centers of all payment data categories are less than the set threshold, thereby obtaining multiple payment data categories.

[0064] In the embodiment of the present invention, Figure 3 As shown, step 103 determines, for each payment data category, the risk probability of the payment data category in each dimension, and the probability convergence value corresponding to each risk probability, including:

[0065] Step 301: Divide the payment data category into multiple payment sub-data, with the payment sub-data corresponding to each dimension one by one;

[0066] Step 302: For each dimension, divide the payment sub-data corresponding to the dimension into multiple sample sub-data in chronological order, so that the number of transactions in each sample sub-data is greater than a set value;

[0067] Step 303: For each sample sub-data, determine the proportion of risk data in the sample sub-data as the risk probability corresponding to the sample sub-data;

[0068] Step 304: Determine the corresponding mean and variance based on the risk probabilities corresponding to the multiple sample sub-data corresponding to the dimension;

[0069] Step 305: Determine the corresponding mean as the risk probability of the payment data category in the dimension;

[0070] Step 306: Determine a probability convergence value corresponding to the risk probability of the dimension based on the corresponding variance.

[0071] In one embodiment, determining a probability convergence value corresponding to the risk probability of the dimension based on the corresponding variance includes:

[0072] The probability convergence value corresponding to the risk probability of this dimension is determined as Where σ is the corresponding variance, and n is the number of sample sub-data corresponding to this dimension.

[0073] In the embodiment of the present invention, Figure 4 As shown, step 104 selects a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability for each payment data category, including:

[0074] Step 401: Determine a partial order of payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability. For any two payment data categories, the partial order can be used to determine whether the first payment data category is superior to the second payment data category.

[0075] Step 402: Selecting a plurality of maximum payment data categories of the partial order from the plurality of payment data categories, wherein the maximum payment data category is a maximum element of the partial order;

[0076] Step 403: Determine the multiple extremely large payment data categories as low-risk payment data categories.

[0077] In the embodiment of the present invention, Figure 5 As shown, step 401 determines the partial order of the payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability, including:

[0078] Step 501: For any two payment data categories, if for each dimension, the risk probability of the first payment data category of the two payment data categories corresponding to the dimension is less than or equal to the risk probability of the second payment data category of the two payment data categories corresponding to the dimension, and the probability convergence values corresponding to the risk probabilities of the first payment data category corresponding to each dimension are all less than an acceptable probability convergence threshold, then it is determined that the first payment data category is better than the second payment data category.

[0079] In one embodiment, an acceptable probability convergence threshold is determined as follows:

[0080] Set the acceptable risk probability error threshold ε and the probability P that the acceptable risk probability error is greater than ε;

[0081] The acceptable probability convergence threshold is determined as ε 2 ×P.

[0082] In the embodiment of the present invention, Figure 6 As shown, step 105 is to screen out low-risk payment data corresponding to each customer from the customer's historical payment data at the bank, including:

[0083] Step 601: Determine a transaction vector corresponding to the customer based on the customer's historical payment data at the bank, wherein each component of the transaction vector corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the customer's historical payment data at the bank;

[0084] Step 602: For each historical payment record of the customer at the bank, extract the counterparty of the historical payment record; and determine the counterparty quantity corresponding to the historical payment record based on the counterparty's historical transaction data, wherein each component of the counterparty quantity corresponds to a transaction category on a one-to-one basis, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data.

[0085] Step 603: Determine a partial order corresponding to the historical payment data of the customer at the bank based on the transaction vector corresponding to the customer and the counterparty direction quantity corresponding to the historical payment data of the customer at the bank. For any two historical payment data of the bank within a predetermined time range, the partial order can be used to determine whether the first of the two historical payment data is closer to the second historical payment data.

[0086] Step 604: Determine the maximum historical payment data of the partial sequence according to the partial sequence corresponding to the historical payment data of the customer at the bank, wherein the maximum historical payment data is the maximum element of the partial sequence;

[0087] Step 605: Based on the maximum historical payment data, determine whether the historical payment data of the customer at the bank is low-risk payment data corresponding to the customer.

[0088] It should be noted that the maximum element of a partial order is the element in the set corresponding to the partial order that has no other elements superior to it. Obviously, the number of maximum elements may be greater than 1.

[0089] In one embodiment, determining the maximum historical payment data of the partial sequence based on the partial sequence corresponding to the historical payment data of the customer at the bank includes:

[0090] 1. Initialize the partial order maximum value corresponding to each historical payment data of the bank within a predetermined time range to possible, and initialize the partial order comparison value corresponding to each historical payment data to yes;

[0091] 2. For each historical payment data of the bank within the predetermined time range, perform the following steps:

[0092] 2.1 If the partial order maximum value corresponding to the historical payment data is not possible, continue to perform step 2 for the next historical payment data;

[0093] 2.2 If the partial order maximum value corresponding to the historical payment data is possible, then compare the historical payment data (except the historical payment data) whose corresponding partial order comparison value is yes with the historical payment data in sequence; if the historical payment data whose corresponding partial order comparison value is yes is closer to the historical payment data, then set the partial order maximum value corresponding to the historical payment data to no, and then continue to perform the above step 2 on the next historical payment data; if the historical payment data is closer to the historical payment data whose corresponding partial order comparison value is yes, then set the partial order maximum value corresponding to the historical payment data whose corresponding partial order comparison value is yes to no, and determine the historical payment data whose corresponding partial order comparison value is yes as the next historical payment data of the historical payment data;

[0094] 2.3 If it is confirmed that all historical payment data with corresponding partial order comparison values of yes are not close to the historical payment data, then the historical payment data is determined as the maximum historical payment data of the partial order, and the partial order comparison value of each secondary historical payment data of the maximum historical payment data is updated to no.

[0095] In one embodiment, determining whether the historical payment data of the customer at the bank is low-risk payment data corresponding to the customer based on the maximum historical payment data includes:

[0096] When the extremely large historical payment data all belongs to the low-risk payment data category, it is determined that the historical payment data of the customer at the bank is the low-risk payment data corresponding to the customer.

[0097] It should be noted that the number of extremely large historical payment data may be greater than 1, so the above method can be improved as follows:

[0098] Determining whether each historical payment data in the large amount of historical payment data belongs to a low-risk payment data category;

[0099] When the proportion of historical payment data belonging to the low-risk payment data category in the extremely large historical payment data is greater than a set ratio value, it is determined that the historical payment data of the customer at the bank is the low-risk payment data corresponding to the customer.

[0100] In the embodiment of the present invention, Figure 7 As shown, step 603 determines the partial order corresponding to the historical payment data of the customer at the bank based on the transaction vector corresponding to the customer and the counterparty direction quantity corresponding to the historical payment data of the customer at the bank, including:

[0101] Step 701: For each historical payment data of the bank within a predetermined time range, determine the distance between the payment direction quantity corresponding to the historical payment data and the transaction vector corresponding to the customer, and determine the distance as the payer distance corresponding to the historical payment data; determine the distance between the counterparty direction quantity corresponding to the historical payment data and the counterparty direction quantity corresponding to the historical payment data of the customer at the bank, and determine the distance as the counterparty distance corresponding to the historical payment data;

[0102] Step 702: For any two historical payment data of the bank within a predetermined time range, if the payer distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the payer distance corresponding to the second historical payment data of the two historical payment data, and the counterparty distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the counterparty distance corresponding to the second historical payment data of the two historical payment data, then it is determined that the first historical payment data is closer to the second historical payment data.

[0103] The acquisition, storage, use, and processing of data in this application's technical solution comply with relevant national laws and regulations.

[0104] The present invention also provides a risk control device for blockchain-based payment transactions, as described in the following embodiments. Because the principles underlying the device are similar to those of the risk control method for blockchain-based payment transactions, the implementation of the device can be referenced to the implementation of the risk control method for blockchain-based payment transactions, and any repetitions will not be repeated.

[0105] Figure 8 This is a structural diagram of a risk control device for payment transactions based on blockchain in an embodiment of the present invention. Figure 8 As shown, the device includes:

[0106] Data acquisition module 02, used to obtain the bank's historical payment data within a predetermined time range;

[0107] Classification module 04, used to classify the acquired historical payment data to obtain multiple payment data categories;

[0108] The risk probability determination module 06 is used to determine the risk probability of each payment data category in each dimension and the probability convergence value corresponding to each risk probability for each payment data category;

[0109] Category selection module 08, configured to select a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability;

[0110] A screening module 10 is used to screen low-risk payment data corresponding to each customer from the customer's historical payment data at the bank;

[0111] A safe counterparty determination module 12, configured to determine the counterparty corresponding to the low-risk payment data corresponding to the customer as the safe counterparty corresponding to the customer;

[0112] The recommendation upload module 14 is used to recommend the client's corresponding security counterparty to the client, and upload it to the blockchain after the client's digital signature is confirmed;

[0113] The sending module 16 is used to send the security counterparty corresponding to the customer to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than a specified strength value, risk control is performed on the customer's transaction based on the security counterparty stored on the mobile terminal.

[0114] In the embodiment of the present invention, the classification module 04 is specifically used to:

[0115] For each historical payment data, extract the payer, counterparty and payment amount level of the historical payment data;

[0116] Determine, based on the payer's historical transaction data, a payment direction quantity corresponding to the historical payment data, wherein each component of the payment direction quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the payer's historical transaction data;

[0117] Determining, based on the counterparty's historical transaction data, a counterparty quantity corresponding to the historical payment data, wherein each component of the counterparty quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data;

[0118] Determine a payer distance function and a counterparty distance function based on the payment direction quantity and the counterparty direction quantity, wherein, for any two payment data, the function value of the payer distance function corresponding to the two payment data is the distance between the payment direction quantities corresponding to the two payment data, and the function value of the counterparty distance function corresponding to the two payment data is the distance between the counterparty direction quantities corresponding to the two payment data;

[0119] Based on the payer distance function, the counterparty distance function, and the payment amount level, the acquired historical payment data is clustered to obtain multiple payment data categories.

[0120] In the embodiment of the present invention, the risk probability determination module 06 is specifically configured to:

[0121] The payment data category is divided into multiple payment sub-data, and the payment sub-data corresponds to each dimension one by one;

[0122] For each dimension, the payment sub-data corresponding to the dimension is divided into multiple sample sub-data in chronological order, so that the number of transactions in each sample sub-data is greater than the set value;

[0123] For each sample sub-data, the proportion of risk data in the sample sub-data is determined as the risk probability corresponding to the sample sub-data;

[0124] Based on the risk probabilities corresponding to the multiple sample sub-data corresponding to the dimension, the corresponding mean and variance are determined;

[0125] The corresponding mean is determined as the risk probability of the payment data category in the dimension;

[0126] The probability convergence value corresponding to the risk probability of the dimension is determined based on the corresponding variance.

[0127] In the embodiment of the present invention, the category selection module 08 is specifically configured to:

[0128] Determine a partial order of the payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability, wherein, for any two payment data categories, the partial order can be used to determine whether a first payment data category of the two payment data categories is superior to a second payment data category;

[0129] Selecting a plurality of maximum payment data categories of the partial order from the plurality of payment data categories, wherein the maximum payment data category is a maximum element of the partial order;

[0130] The plurality of extremely large payment data categories are determined as low-risk payment data categories.

[0131] In the embodiment of the present invention, the category selection module 08 is specifically configured to:

[0132] For any two payment data categories, if for each dimension, the risk probability of the first payment data category of the two payment data categories corresponding to the dimension is less than or equal to the risk probability of the second payment data category of the two payment data categories corresponding to the dimension, and the probability convergence values corresponding to the risk probabilities of the first payment data category corresponding to each dimension are less than an acceptable probability convergence threshold, then the first payment data category is determined to be superior to the second payment data category.

[0133] In the embodiment of the present invention, the screening module 10 is specifically used to:

[0134] Determine a transaction vector corresponding to the customer based on the customer's historical payment data at the bank, where each component of the transaction vector corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the customer's historical payment data at the bank;

[0135] For each historical payment record of the customer at the bank, extract the counterparty of the historical payment data; and determine the counterparty direction quantity corresponding to the historical payment data based on the counterparty's historical transaction data, wherein each component of the counterparty direction quantity corresponds to a transaction category on a one-to-one basis, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data;

[0136] Determine, based on the transaction vector corresponding to the customer and the counterparty quantity corresponding to the historical payment data of the customer at the bank, a partial order corresponding to the historical payment data of the customer at the bank, wherein, for any two historical payment data of the bank within a predetermined time range, the partial order can be used to determine whether a first historical payment data of the two historical payment data is closer to a second historical payment data;

[0137] Determining, based on the partial sequence corresponding to the historical payment data of the customer at the bank, a maximum historical payment data of the partial sequence, wherein the maximum historical payment data is a maximum element of the partial sequence;

[0138] Based on the extremely large historical payment data, determine whether the historical payment data of the customer at the bank is the low-risk payment data corresponding to the customer.

[0139] In the embodiment of the present invention, the screening module 10 is specifically used to:

[0140] For each historical payment data of the bank within a predetermined time range, determine the distance between the payment direction quantity corresponding to the historical payment data and the transaction vector corresponding to the customer, and determine the distance as the payer distance corresponding to the historical payment data; determine the distance between the counterparty direction quantity corresponding to the historical payment data and the counterparty direction quantity corresponding to the historical payment data of the customer at the bank, and determine the distance as the counterparty distance corresponding to the historical payment data;

[0141] For any two historical payment data of the bank within a predetermined time range, if the payer distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the payer distance corresponding to the second historical payment data of the two historical payment data, and the counterparty distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the counterparty distance corresponding to the second historical payment data of the two historical payment data, then it is determined that the first historical payment data is closer to the second historical payment data.

[0142] An embodiment of the present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the risk control method for payment transactions based on blockchain is implemented.

[0143] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned risk control method for blockchain-based payment transactions.

[0144] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned risk control method for blockchain-based payment transactions.

[0145] In an embodiment of the present invention, compared with the technical solution in the prior art in which customers cannot trade when the network signal is weak, resulting in a poor customer experience, the method obtains historical payment data of a bank within a predetermined time range; classifies the obtained historical payment data to obtain multiple payment data categories; for each payment data category, determines the risk probability of the payment data category in each dimension and the probability convergence value corresponding to each risk probability; selects a low-risk payment data category from the multiple payment data categories based on the risk probability in each dimension and the probability convergence value corresponding to each risk probability; for each customer, filters out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank; determines the counterparty corresponding to the low-risk payment data corresponding to the customer as the customer's corresponding secure counterparty; recommends the customer's corresponding secure counterparty to the customer, and uploads it to the blockchain after the customer's digital signature is confirmed; and sends the customer's corresponding secure counterparty to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than a specified strength value, risk control is performed on the customer's transactions based on the secure counterparty stored on the mobile terminal. Transactions can be conducted even when the network signal is weak, thereby improving the customer experience.

[0146] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0150] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific 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 should be included in the scope of protection of the present invention.

Claims

1. A risk control method for payment transactions based on blockchain, characterized in that: include: Obtain historical payment data of the bank within a predetermined time range; Classify the acquired historical payment data to obtain multiple payment data categories; For each payment data category, determine the risk probability of the payment data category in each dimension, as well as the probability convergence value corresponding to each risk probability; Selecting a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability; For each customer, filter out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank; Determine the counterparty corresponding to the low-risk payment data of the customer as the safe counterparty corresponding to the customer; Recommend the client's corresponding security counterparty to the client, and upload it to the blockchain after the client's digital signature confirmation; The corresponding security counterparty of the customer is sent to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than the specified strength value, risk control is performed on the customer's transaction based on the security counterparty stored on the mobile terminal; Categorize the acquired historical payment data to obtain multiple payment data categories, including: For each historical payment data, extract the payer, counterparty and payment amount level of the historical payment data; Determine, based on the payer's historical transaction data, a payment direction quantity corresponding to the historical payment data, wherein each component of the payment direction quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the payer's historical transaction data; Determining, based on the counterparty's historical transaction data, a counterparty quantity corresponding to the historical payment data, wherein each component of the counterparty quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data; Determine a payer distance function and a counterparty distance function based on the payment direction quantity and the counterparty direction quantity, wherein, for any two payment data, the function value of the payer distance function corresponding to the two payment data is the distance between the payment direction quantities corresponding to the two payment data, and the function value of the counterparty distance function corresponding to the two payment data is the distance between the counterparty direction quantities corresponding to the two payment data; Based on the payer distance function, the counterparty distance function, and the payment amount level, the acquired historical payment data is clustered to obtain multiple payment data categories.

2. The method according to claim 1, wherein For each payment data category, determine the risk probability of the payment data category in each dimension, as well as the probability convergence value corresponding to each risk probability, including: The payment data category is divided into multiple payment sub-data, and the payment sub-data corresponds to each dimension one by one; For each dimension, the payment sub-data corresponding to the dimension is divided into multiple sample sub-data in chronological order, so that the number of transactions in each sample sub-data is greater than the set value; For each sample sub-data, the proportion of risk data in the sample sub-data is determined as the risk probability corresponding to the sample sub-data; Based on the risk probabilities corresponding to the multiple sample sub-data corresponding to the dimension, the corresponding mean and variance are determined; The corresponding mean is determined as the risk probability of the payment data category in the dimension; The probability convergence value corresponding to the risk probability of the dimension is determined based on the corresponding variance.

3. The method according to claim 1, wherein For each payment data category, based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability, a low-risk payment data category is selected from the multiple payment data categories, including: Determine a partial order of the payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability, wherein, for any two payment data categories, the partial order is used to determine whether a first payment data category of the two payment data categories is superior to a second payment data category; Selecting a plurality of maximum payment data categories of the partial order from the plurality of payment data categories, wherein the maximum payment data category is a maximum element of the partial order; The plurality of extremely large payment data categories are determined as low-risk payment data categories.

4. The method according to claim 3, wherein Based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability, the partial order of the payment data categories is determined, including: For any two payment data categories, if for each dimension, the risk probability of the first payment data category of the two payment data categories corresponding to the dimension is less than or equal to the risk probability of the second payment data category of the two payment data categories corresponding to the dimension, and the probability convergence values corresponding to the risk probabilities of the first payment data category corresponding to each dimension are less than an acceptable probability convergence threshold, then the first payment data category is determined to be superior to the second payment data category.

5. The method according to claim 1, wherein For each customer, filter out the low-risk payment data corresponding to the customer from the customer's historical payment data at the bank, including: Determine a transaction vector corresponding to the customer based on the customer's historical payment data at the bank, where each component of the transaction vector corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the customer's historical payment data at the bank; For each historical payment record of the customer at the bank, extract the counterparty of the historical payment data; and determine the counterparty direction quantity corresponding to the historical payment data based on the counterparty's historical transaction data, wherein each component of the counterparty direction quantity corresponds to a transaction category on a one-to-one basis, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data; Determine a partial order corresponding to the historical payment data of the customer at the bank based on the transaction vector corresponding to the customer and the counterparty quantity corresponding to the historical payment data of the customer at the bank, wherein, for any two historical payment data of the bank within a predetermined time range, the partial order is used to determine whether a first historical payment data of the two historical payment data is closer to a second historical payment data; Determining, based on the partial sequence corresponding to the historical payment data of the customer at the bank, a maximum historical payment data of the partial sequence, wherein the maximum historical payment data is a maximum element of the partial sequence; Based on the extremely large historical payment data, determine whether the historical payment data of the customer at the bank is the low-risk payment data corresponding to the customer.

6. The method according to claim 1, wherein Determining a partial order corresponding to the historical payment data of the customer at the bank based on the transaction vector corresponding to the customer and the counterparty quantity corresponding to the historical payment data of the customer at the bank includes: For each historical payment data of the bank within a predetermined time range, determine the distance between the payment direction quantity corresponding to the historical payment data and the transaction vector corresponding to the customer, and determine the distance as the payer distance corresponding to the historical payment data; determine the distance between the counterparty direction quantity corresponding to the historical payment data and the counterparty direction quantity corresponding to the historical payment data of the customer at the bank, and determine the distance as the counterparty distance corresponding to the historical payment data; For any two historical payment data of the bank within a predetermined time range, if the payer distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the payer distance corresponding to the second historical payment data of the two historical payment data, and the counterparty distance corresponding to the first historical payment data of the two historical payment data is less than or equal to the counterparty distance corresponding to the second historical payment data of the two historical payment data, then it is determined that the first historical payment data is closer to the second historical payment data.

7. A risk control device for payment transactions based on blockchain, characterized in that: include: A data acquisition module is used to obtain the bank's historical payment data within a predetermined time range; A classification module is used to classify the acquired historical payment data to obtain multiple payment data categories; A risk probability determination module is used to determine, for each payment data category, the risk probability of the payment data category in each dimension, and the probability convergence value corresponding to each risk probability; a category selection module, configured to select a low-risk payment data category from the multiple payment data categories based on the risk probabilities in each dimension and the probability convergence values corresponding to each risk probability; A screening module is used to screen low-risk payment data corresponding to each customer from the customer's historical payment data at the bank; A safe counterparty determination module, configured to determine a counterparty corresponding to the low-risk payment data corresponding to the customer as the safe counterparty corresponding to the customer; The recommendation upload module is used to recommend the client's corresponding security counterparty to the client, and upload it to the blockchain after the client's digital signature is confirmed; A sending module is used to send the security counterparty corresponding to the customer to the customer's mobile terminal. When the network signal strength of the customer's mobile terminal is lower than a specified strength value, risk control is performed on the customer's transaction based on the security counterparty stored on the mobile terminal; Classification module, specifically used for: For each historical payment data, extract the payer, counterparty and payment amount level of the historical payment data; Determine, based on the payer's historical transaction data, a payment direction quantity corresponding to the historical payment data, wherein each component of the payment direction quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the payer's historical transaction data; Determining, based on the counterparty's historical transaction data, a counterparty quantity corresponding to the historical payment data, wherein each component of the counterparty quantity corresponds one-to-one to a transaction category, and the value of each component is equal to the number of transactions of the transaction category corresponding to the component in the counterparty's historical transaction data; Determine a payer distance function and a counterparty distance function based on the payment direction quantity and the counterparty direction quantity, wherein, for any two payment data, the function value of the payer distance function corresponding to the two payment data is the distance between the payment direction quantities corresponding to the two payment data, and the function value of the counterparty distance function corresponding to the two payment data is the distance between the counterparty direction quantities corresponding to the two payment data; Based on the payer distance function, the counterparty distance function, and the payment amount level, the acquired historical payment data is clustered to obtain multiple payment data categories.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images