Block chain-based cross-border payment intelligent risk control method, system and device, and medium

Abstract

The invention relates to a block chain-based cross-border payment intelligent risk control method, system and device, and a medium. The method comprises the steps of obtaining transaction data related to cross-border payment, encrypting the transaction data through a block chain node and then performing uplink storage, performing consistency verification on the transaction data in different block chain networks, and when the verification result is consistent, analyzing the transaction data on the block chain to obtain risk characteristics; based on the risk characteristics, judging whether the transaction of the cross-border payment has potential risks or not, and obtaining a judgment result; based on the judgment result, obtaining risk information; based on the risk information, triggering a corresponding intervention measure according to a preset risk control strategy, and meanwhile, performing uplink storage on the risk information and an intervention record corresponding to the preset risk control strategy; by adopting the method, the security and the credibility of the transaction data can be improved, the consistency of the transaction data is ensured, the risk identification capability and the management capability are enhanced, the supervision and auditing efficiency is improved, and the business process is optimized.

Description

Technical Field

[0001] The present invention belongs to the field of financial data processing, and in particular relates to a blockchain-based cross-border payment intelligent risk control method, system, device and medium. Background Art

[0002] With the acceleration of global economic integration, cross-border payments are experiencing explosive growth. Traditional cross-border payment methods face numerous challenges in terms of efficiency, cost, and security, such as cumbersome transaction processes, long processing times, high fees, and the risk of fraud and money laundering. The emergence of blockchain technology has brought new opportunities for cross-border payments. Its decentralized, tamper-proof, secure, and transparent nature can effectively address these pain points.

[0003] However, cross-border payments involve different blockchain networks across multiple countries and regions, which may employ varying blockchain technology architectures and consensus mechanisms. In this multi-chain environment, ensuring the consistency of transaction data and risk control information across these networks is a pressing issue. Existing technologies lack effective cross-chain data synchronization and verification mechanisms, making it difficult to ensure data integrity and consistency across these networks. This can lead to omissions or inaccuracies in risk information, hindering intelligent risk control decisions.

[0004] Furthermore, traditional cross-border payment risk control methods, which primarily rely on manual review and simple rule matching, are unable to cope with the increasingly complex transaction environment and diverse risk types. Manual review is inefficient and prone to human error, while rule matching has limitations and cannot adapt to rapidly changing markets and new risk models. Summary of the Invention

[0005] Based on this, it is necessary to provide a blockchain-based cross-border payment intelligent risk control method, system, equipment and medium that can ensure data integrity and consistency in response to the above technical problems.

[0006] In the first aspect, this application provides a blockchain-based cross-border payment intelligent risk control method, including:

[0007] Obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes, and store it on the chain. Perform consistency verification on transaction data in different blockchain networks to obtain verification results; the verification results are used to indicate whether the transaction data on different blockchains is consistent;

[0008] When the verification results are consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics;

[0009] Based on risk characteristics, determine whether there are potential risks in cross-border payment transactions and obtain the judgment results;

[0010] Based on the judgment results, risk information is obtained;

[0011] Based on risk information, corresponding intervention measures are triggered according to pre-set risk control strategies, and the risk information and intervention records corresponding to the pre-set risk control strategies are saved on the chain; intervention measures include suspending transactions, requiring additional verification information, and notifying relevant personnel for manual review.

[0012] Furthermore, the transaction data related to cross-border payments is obtained, the transaction data is encrypted by the blockchain node and then stored on the chain, and the transaction data in different blockchain networks is subjected to consistency verification to obtain the verification results, including the following steps:

[0013] Obtain transaction data related to cross-border payments; transaction data includes transaction amount, identity information of both parties, transaction time, transaction purpose, and recipient bank information;

[0014] Utilize the encryption function of the blockchain node and adopt a specific encryption algorithm to convert the acquired transaction data into ciphertext form, and then store the encrypted transaction data on the blockchain;

[0015] When there are multiple blockchain networks involved in cross-border payment transactions, consistency verification is performed on the transaction data of the cross-border payments in each blockchain network to verify whether the data of different transaction data copies are consistent and obtain the verification results.

[0016] Furthermore, when the verification results are consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics, including:

[0017] Use graph neural network models to extract risk features of transaction data on blockchain;

[0018] Among them, the graph neural network model includes:

[0019] Input layer, used to input node features and edge information in the blockchain transaction graph;

[0020] The graph convolution layer is used to update the feature representation of the node by aggregating the information of neighboring nodes. It uses multiple layers of graph convolution layers, and the output of each layer serves as the input of the next layer. The graph convolution network update formula in the multi-layer graph convolution layer is: Among them H (l) is the node feature matrix of the lth layer, is the adjacency matrix with self-loops added, is the corresponding degree matrix, W (l) is the weight matrix of the lth layer, σ is the nonlinear activation function;

[0021] The output layer is used to output the risk feature representation of each node; the risk features are used in the subsequent risk assessment model.

[0022] Furthermore, based on the risk characteristics, it is determined whether the cross-border payment transaction has potential risks, and the judgment results are obtained, including:

[0023] Use the logistic regression model to determine whether there are potential risks in cross-border payment transactions and obtain the judgment results;

[0024] Among them, the logistic regression model includes:

[0025] Input layer, used to input risk characteristics;

[0026] The logistic regression layer is used to calculate the probability that a transaction is a risky transaction based on the input risk characteristics through a logistic regression function; the logistic regression function is: Where P(Y=1|X) represents the probability of risk in cross-border payment transactions, X1, X2, ..., X n is the risk feature of the input, β0, β1, ..., β n are model parameters, learned through training data;

[0027] The output layer is used to determine that there is a potential risk when the probability value is greater than the set threshold, and output the judgment result on whether there is a potential risk in the cross-border payment transaction.

[0028] Furthermore, based on the judgment result, risk information is obtained, including the following steps:

[0029] When a cross-border payment transaction is judged to have potential risks, a risk score is assigned to the cross-border payment transaction based on the preset risk assessment rules;

[0030] Based on the risk score, the risk level of the cross-border payment transaction is obtained;

[0031] Generate risk information based on risk level.

[0032] Based on risk information, corresponding intervention measures are triggered according to pre-set risk control strategies, and risk information and intervention records are saved on the chain, including:

[0033] Match risk information with pre-set risk control strategies to determine whether intervention measures need to be triggered;

[0034] According to the risk level, different intervention measures are matched and executed through smart contracts;

[0035] Record the execution of intervention measures and generate intervention records; intervention records include transaction ID, intervention measures, execution time and execution results;

[0036] Encrypt the intervention record and write it into the blockchain through the blockchain consensus mechanism;

[0037] The intervention records are verified through the hash verification mechanism of the blockchain to obtain the verification results.

[0038] Furthermore, it also includes:

[0039] Through the blockchain’s event monitoring mechanism, the on-chain status of risk information and intervention records is monitored in real time, generating audit logs. The audit logs include the entire process of risk information collection, risk control strategy triggering, intervention measures execution, and on-chain recording.

[0040] Regularly audit risk information and intervention records on the blockchain according to a preset time period to obtain audit results;

[0041] Analyze audit results, adjust pre-set risk control strategies and risk assessment models, and optimize intervention measures;

[0042] Update risk assessment models based on market changes and new risk characteristics.

[0043] Secondly, this application also provides a blockchain-based cross-border payment intelligent risk control system, including:

[0044] The data collection and data verification module is used to obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes, and store it on the chain. It also performs consistency verification on transaction data in different blockchain networks to obtain verification results. The verification results are used to indicate whether the transaction data on different blockchains is consistent.

[0045] The risk feature extraction module is used to analyze the transaction data on the blockchain and obtain risk features when the verification results are consistent;

[0046] The risk assessment module is used to determine whether a cross-border payment transaction has potential risks based on risk characteristics and obtain the assessment results;

[0047] A risk information acquisition module is used to obtain risk information based on the judgment results;

[0048] Decision-making execution module: used to trigger corresponding intervention measures based on risk information and pre-set risk control strategies, and at the same time save the risk information and intervention records corresponding to the pre-set risk control strategies on the chain; intervention measures include suspending transactions, requiring additional verification information, and notifying relevant personnel for manual review.

[0049] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform any of the methods described in the first aspect.

[0050] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method described in any one of the first aspects is executed.

[0051] The above-mentioned blockchain-based cross-border payment intelligent risk control method, system, device and medium, BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 Schematic diagram of an application environment of a blockchain-based cross-border payment intelligent risk control method in one embodiment;

[0054] Figure 2 The figure is a flowchart of a blockchain-based cross-border payment intelligent risk control method in one embodiment;

[0055] Figure 3 This is a schematic diagram of the structure of a blockchain-based cross-border payment intelligent risk control system in one embodiment;

[0056] Figure 4 A schematic diagram of the computer device structure for blockchain-based cross-border payment intelligent risk control in one embodiment. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] The cross-border payment intelligent risk control method based on blockchain provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the risk control personnel terminal 101, the payer terminal 102, the payee terminal 103, and the financial institution terminal 104 communicate with the blockchain 105 via the blockchain network to obtain or store blockchain resources. The terminal devices may include, but are not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices.

[0059] In an exemplary embodiment, Figure 2 As shown in the figure, a cross-border payment intelligent risk control method based on blockchain is provided. Figure 1 The risk control personnel terminal 101 in FIG. 1 is used as an example to illustrate, including the following S201 to S205:

[0060] S201, obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes and store it on the chain, perform consistency verification on transaction data in different blockchain networks, and obtain verification results.

[0061] The verification result is used to indicate whether the transaction data is consistent on different blockchains.

[0062] Exemplarily, obtaining transaction data related to cross-border payments, encrypting the transaction data through blockchain nodes and then storing it on-chain, performing consistency verification on the transaction data in different blockchain networks, and obtaining verification results may include the following steps:

[0063] Step 1001: Collect all data related to cross-border payments from various channels or systems related to cross-border payment services; the data may include the transaction amount, information of both parties to the transaction, and the transaction time.

[0064] In step 1002, blockchain nodes (the computers or servers that make up the blockchain network) acquire cross-border payment transaction data and use their encryption capabilities to encrypt the transaction data using a specific encryption algorithm, converting it into ciphertext to protect data privacy and security.

[0065] In step 1003, after encryption is completed, the encrypted transaction data is uploaded to the blockchain network for storage. The distributed ledger feature of the blockchain makes it difficult to tamper with the data once it is stored, ensuring the security and non-repudiation of the data.

[0066] Step 1004: In actual cross-border payment transaction scenarios, there may be multiple different blockchain networks involved. Each blockchain network records some data related to cross-border payments. To ensure that the data recorded on different blockchain networks for the same cross-border payment transaction is consistent, a consistency check is required. The check process involves comparing key information of the transaction data in different blockchain networks, such as the transaction amount, the identifiers of the two parties to the transaction, and the transaction timestamp, to check whether the key information matches.

[0067] In step 1005, after consistency verification of the transaction data in different blockchain networks, a verification result is obtained. The verification result is used to indicate that the transaction data is consistent, indicating that the records of the cross-border payment transaction in different blockchain networks are accurate.

[0068] S202: When the verification results are consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics.

[0069] For example, when the verification result is consistent, analyzing the transaction data on the blockchain to obtain risk characteristics may include the following steps:

[0070] Step 2001: After completing the consistency check of cross-border payment transaction data in different blockchain networks and determining that the verification results show that the data is consistent, the relevant transaction data stored on the blockchain is analyzed and studied;

[0071] In step 2002, risk-related features are extracted from a large amount of transaction data by using a graph neural network model. The extracted features are risk features. Among them, risk features may include abnormal fluctuations in transaction amounts and unreasonable transaction time and location.

[0072] S203: Based on the risk characteristics, determine whether the cross-border payment transaction has potential risks and obtain a determination result.

[0073] For example, based on the risk characteristics, it is determined whether a cross-border payment transaction has potential risks, and the determination result includes:

[0074] Step 3001: Evaluate a specific cross-border payment transaction based on risk characteristics derived from transaction data on the blockchain.

[0075] Step 3002: By comparing and considering the actual circumstances of the cross-border payment transaction with the risk characteristics, a logistic regression model is used to determine whether the transaction has potential risks. Potential risks may include fraud risk, money laundering risk, and illegal operation risk.

[0076] In step 3003, a clear judgment result is finally obtained, which is either a judgment that the cross-border payment transaction has potential risks, or a judgment that the cross-border payment transaction does not have potential risks, providing a basis for subsequent decision-making.

[0077] S204: Obtain risk information based on the judgment result.

[0078] Exemplarily, based on the judgment result, risk information is obtained, including:

[0079] In step 4001, if the result of the judgment is that there is a potential risk, the risk type is determined based on the risk characteristics. If the risk characteristics show that the transaction amount fluctuates frequently and significantly within a short period of time and the counterparty is an anonymous account, it is determined to be a money laundering risk. If the transaction suddenly shows an abnormal amount that far exceeds the historical average and the payment purpose is unclear, it is determined to be a fraud risk.

[0080] Step 4002: Obtain a risk score based on the transaction amount, transaction frequency, and counterparty reputation in the risk information;

[0081] Step 4003, determine the risk level based on the risk score, and finally generate risk information; wherein the risk information may include risk type, risk score and risk level.

[0082] S205: Based on the risk information and according to the pre-set risk control strategy, corresponding intervention measures are triggered, and the risk information and the intervention records corresponding to the pre-set risk control strategy are saved on the chain.

[0083] For example, based on risk information and according to pre-set risk control strategies, corresponding intervention measures are triggered, and the risk information and intervention records corresponding to the pre-set risk control strategies are saved on the chain, including the following steps:

[0084] Step 5001: After obtaining risk information, the risk information is matched with pre-set risk control strategies based on a series of pre-set strategies corresponding to different risk situations, thereby initiating corresponding actions. If the risk information indicates that a cross-border payment transaction is a high-risk money laundering transaction, the transaction funds are frozen.

[0085] In step 5002, while taking intervention measures, the risk information related to the transaction and the actual intervention measures implemented will be recorded, and this information will be uploaded to the blockchain network through the blockchain node for storage; the characteristics of the blockchain ensure that the records are tamper-proof and traceable.

[0086] The embodiment of the present application provides a blockchain-based intelligent risk control method for cross-border payments, which obtains transaction data related to cross-border payments, encrypts the transaction data through blockchain nodes and then stores it on the chain, performs consistency verification on the transaction data in different blockchain networks, and when the verification results are consistent, analyzes the transaction data on the blockchain to obtain risk characteristics; based on the risk characteristics, determines whether there are potential risks in the cross-border payment transactions and obtains a judgment result; based on the judgment result, obtains risk information; based on the risk information, triggers corresponding intervention measures according to a pre-set risk control strategy, and at the same time, stores the risk information and the intervention records corresponding to the pre-set risk control strategy on the chain; the use of this method can improve the security and credibility of transaction data, ensure the consistency of transaction data, enhance risk identification and management capabilities, improve supervision and audit efficiency, and optimize business processes.

[0087] Optionally, obtaining transaction data related to cross-border payments, encrypting the transaction data through blockchain nodes and storing the encrypted data on the blockchain, performing consistency verification on the transaction data in different blockchain networks, and obtaining verification results may include the following steps:

[0088] Step 6001: Obtain transaction data related to cross-border payments; transaction data includes transaction amount, identity information of both parties to the transaction, transaction time, transaction purpose, and beneficiary bank information;

[0089] Risk control personnel cooperate with participants in cross-border payment business such as banks and payment institutions. Risk control personnel terminals obtain cross-border payment transaction data recorded in their systems, clean the obtained data, remove duplicate, erroneous and incomplete data, and verify the authenticity and legality of the data.

[0090] Step 6002: Using the encryption function of the blockchain node, a specific encryption algorithm is used to convert the acquired transaction data into ciphertext, and the encrypted transaction data is stored on the blockchain.

[0091] Based on the security requirements of cross-border payment transaction data, the risk control personnel terminal selects the asymmetric encryption algorithm RSA, generates a public key (for encrypting data) and a private key (for decrypting), encrypts the cross-border payment transaction data, and ensures the security of the data during transmission and storage.

[0092] Step 6003: When multiple blockchain networks participate in a cross-border payment transaction, a consistency check is performed on the transaction data of the cross-border payment in each blockchain network to verify whether the data in different transaction data copies are consistent;

[0093] Risk control personnel write smart contracts to define the storage structure, access rights and operation rules of the data; risk control personnel deploy the written smart contracts to the selected blockchain platform, which may include Ethereum; the risk control personnel's terminal encapsulates the encrypted transaction data according to the requirements of the smart contract to form a transaction request, and broadcasts it to other nodes through the P2P network of the blockchain network; after receiving the transaction request, other nodes in the network will verify it, and the verification method may include the format and signature of the data; after the verification is passed, the data will be stored in the local blockchain database and a new block will be formed; in different blockchain networks, hash values ​​are calculated for the cross-border payment transaction data, and these hash values ​​are compared to see if they are consistent. If they are consistent, it means that the data is consistent in different networks, and a verification result is obtained.

[0094] Optionally, when the verification result is consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics, including:

[0095] Use graph neural network models to extract risk features of transaction data on blockchain;

[0096] Among them, the graph neural network model includes:

[0097] Input layer, used to input node features and edge information in the blockchain transaction graph;

[0098] Extract the features of each node from the blockchain transaction graph, extract the edge information between nodes, organize the node features into a structured data format (vector or matrix), organize the edge information into an adjacency matrix or edge list to represent the connection relationship between the nodes, input the formatted node features into the input layer of the model, and input the edge information into the model; the input layer structures and standardizes the node features and edge information in the blockchain transaction graph so that the data can be effectively processed by the model.

[0099] The graph convolution layer is used to update the feature representation of the node by aggregating the information of neighboring nodes. It uses multiple layers of graph convolution layers, and the output of each layer serves as the input of the next layer. The graph convolution network update formula in the multi-layer graph convolution layer is: Among them H (l) is the node feature matrix of the lth layer, is the adjacency matrix with self-loops added, is the corresponding degree matrix, W (l) is the weight matrix of the lth layer, σ is the nonlinear activation function;

[0100] In a graph convolutional network, the initial feature vector of each node is obtained from the input layer, which contains the basic attributes of the node. The graph convolution layer updates the features of the current node by aggregating information from neighboring nodes; by multiplying with the weight matrix, the aggregated information is mapped to a new feature space; and nonlinear activation functions (such as ReLU) are applied to introduce nonlinearity, enabling the model to learn more complex feature representations.

[0101] The graph convolution layer can effectively capture the complex relationships between nodes; by aggregating neighbor information, the graph convolution network can improve the model's ability to analyze node features, helping to identify abnormal trading behaviors and potential risks; the introduction of nonlinear activation functions enables the model to learn nonlinear relationships in the data, improving the model's expressive power and prediction accuracy; the weight matrix is ​​learned through training, allowing the model to adaptively adjust the focus of feature extraction and improve its ability to capture key features.

[0102] The output layer is used to output the risk feature representation of each node; the risk feature is used in the subsequent logistic regression model.

[0103] The node features processed by the graph convolution layer are finally integrated and output to obtain the risk feature representation of each node; the risk feature representation will be used in the subsequent risk assessment model to determine whether the node (such as a transaction) has potential risks.

[0104] Optionally, based on the risk characteristics, determine whether the cross-border payment transaction has potential risks, and obtain a determination result, including:

[0105] Use the logistic regression model to determine whether there are potential risks in cross-border payment transactions and obtain the judgment results;

[0106] Among them, the logistic regression model includes:

[0107] Input layer, used to input risk characteristics;

[0108] The input layer is used to input risk features. Risk features are node feature representations processed by the graph convolution layer, which contain comprehensive information about the node itself and its neighboring nodes.

[0109] The logistic regression layer is used to calculate the probability that a transaction is a risky transaction based on the input risk characteristics through a logistic regression function; the logistic regression function is: Where P(Y=1|X) represents the probability of risk in cross-border payment transactions, X1, X2, ..., X n is the risk feature of the input, β0, β1, ..., β n are model parameters, learned through training data;

[0110] By designing and applying a logistic regression layer, we can effectively calculate the probability of a transaction being risky based on the input risk characteristics, providing a reliable basis for risk assessment of cross-border payment transactions. This process improves the accuracy of risk assessment and enhances the model's interpretability and computational efficiency.

[0111] The output layer is used to determine that there is a potential risk when the probability value is greater than the set threshold, and output the judgment result on whether there is a potential risk in the cross-border payment transaction.

[0112] The output layer converts the probability value into a clear risk judgment result by setting a threshold, making it easier to take appropriate measures.

[0113] Optionally, obtaining risk information based on the judgment result includes the following steps:

[0114] Step 7001: When it is determined that a cross-border payment transaction has potential risks, a risk score is assigned to the cross-border payment transaction based on a preset risk assessment model. The risk score is calculated using a weighted summation formula, where each factor has a weight indicating its contribution to the risk.

[0115] The risk score S can be calculated using the following formula:

[0116] S = w1 × transaction amount weight + w2 × transaction frequency weight + w3 × counterparty reputation weight, where w1, w2, and w3 are preset weight values.

[0117] Step 7002: Obtain the risk level of the cross-border payment transaction based on the risk score;

[0118] Based on the risk score, the risk level is divided into different levels based on pre-set thresholds; when the risk score is less than 30, it is low risk; when the risk score is between 30 and 70, it is medium risk; when the risk score is greater than 70, it is high risk.

[0119] Step 7003: Generate risk information based on the risk level.

[0120] Generate detailed risk information based on the risk level, where the risk information may include the risk level, risk score, and risk cause.

[0121] Optionally, based on risk information and according to pre-set risk control strategies, corresponding intervention measures are triggered, and risk information and intervention records are saved on-chain, including:

[0122] Step 8001: Match risk information with pre-set risk control strategies to determine whether intervention measures need to be triggered;

[0123] Based on the generated risk information (including risk level, risk score, etc.), it is matched with the pre-set risk control strategy to determine whether intervention measures need to be triggered; among them, the risk control strategy defines the intervention measures corresponding to different risk levels.

[0124] Step 8002: Match different intervention measures based on the risk level and execute the corresponding intervention measures through smart contracts;

[0125] When risk information matches the corresponding risk control strategy, the corresponding intervention measures are automatically executed through smart contracts. Smart contracts can ensure the rapid and accurate execution of intervention measures and reduce delays and errors caused by manual intervention.

[0126] Step 8003: Record the execution of the intervention measures and generate an intervention record; the intervention record includes the transaction ID, intervention measures, execution time, and execution results;

[0127] Step 8004: Encrypt the intervention record and write the encrypted intervention record into the blockchain through the blockchain consensus mechanism.

[0128] The intervention records are encrypted to ensure data security and privacy; then, the encrypted intervention records are written into the blockchain through the blockchain’s consensus mechanism; the blockchain’s immutability ensures the authenticity and integrity of the intervention records.

[0129] Step 8005: Verify the intervention record through the hash verification mechanism of the blockchain to obtain the verification result.

[0130] The hash verification mechanism of the blockchain is used to verify the data of the intervention records. By comparing the hash value stored on the blockchain with the newly calculated hash value, it is ensured that the data has not been tampered with. This process improves the efficiency and accuracy of verification.

[0131] Optionally, it also includes:

[0132] Step 9001: Using the blockchain’s event monitoring mechanism, monitor the on-chain status of risk information and intervention records in real time and generate an audit log. The audit log includes the entire process of risk information collection, risk control strategy triggering, intervention measure execution, and on-chain recording.

[0133] The event monitoring mechanism can detect specific events on the blockchain, such as the update of risk information or the addition of intervention records, and generate audit logs to ensure the transparency and traceability of the entire process.

[0134] Step 9002: Regularly audit the risk information and intervention records on the blockchain according to a preset time period to obtain audit results.

[0135] Leveraging the blockchain's immutability and timestamp features, we verify the authenticity of risk information. We check whether data has been tampered with and whether it aligns with actual business activities. We assess the integrity of risk information to ensure all relevant data is recorded on the blockchain without omissions. We check whether intervention records comply with relevant laws, regulations, policies, and business rules, ensuring that all interventions have a legitimate basis. We also assess the effectiveness of intervention measures to ensure they achieve the intended risk control and management objectives. This helps verify the integrity and accuracy of data and ensure that all operations comply with regulations.

[0136] Step 9003: Analyze the audit results, adjust the pre-set risk control strategy and risk assessment model, and optimize intervention measures;

[0137] Identify key risk points and existing problems from the audit report; conduct quantitative analysis of the problems found in the audit and assess their potential impact on the business; analyze the trend of risk occurrence to determine whether it is increasing, decreasing or remaining stable; based on the audit results, adjust the parameters in the risk assessment model to more accurately reflect the current risk situation; verify the adjusted model to ensure its effectiveness and accuracy; based on the verification results, further optimize the model to improve its predictive ability; update the rules in the risk control strategy based on the audit results and the adjusted risk assessment model; adjust the thresholds in the strategy to adapt to the new risk level; test the updated strategy in the actual business process and evaluate its effectiveness; review the existing intervention process and identify areas for improvement; evaluate the effectiveness of existing intervention measures to determine which measures are effective and which need improvement; based on the evaluation results, update the intervention measures to improve their relevance and effectiveness.

[0138] Step 9004: Update the risk assessment model based on market changes and new risk characteristics.

[0139] Regularly review and update risk models to identify sources of risk that may affect the achievement of objectives, such as market risk, operational risk, financial risk, and legal risk; re-examine and update each link of risk identification, risk analysis, and risk assessment to ensure that the risk assessment model can reflect the latest market changes and risk characteristics; and ensure that the model can continuously and effectively identify and manage risks.

[0140] In the above-mentioned blockchain-based cross-border payment intelligent risk control method, the blockchain's encryption technology ensures the security of transaction data during transmission and storage, preventing data leakage and tampering; through the consistency verification mechanism, it ensures that transaction data in different blockchain networks remains consistent, avoiding risks caused by data inconsistency; blockchain-based smart contracts and data analysis capabilities can quickly identify potential risks in transactions and trigger corresponding intervention measures to improve the efficiency of risk management; the immutability and transparency of blockchain make all transactions and intervention records traceable, enhancing the credibility of the system; the automated execution capability of smart contracts reduces manual intervention and improves the efficiency of transaction processing and risk management.

[0141] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0142] Based on the same inventive concept, the embodiments of the present application also provide a blockchain-based cross-border payment intelligent risk control system for implementing the aforementioned blockchain-based cross-border payment intelligent risk control method. The implementation solution provided by this system is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of one or more blockchain-based cross-border payment intelligent risk control system embodiments provided below can be found in the limitations of the blockchain-based cross-border payment intelligent risk control method above and will not be repeated here.

[0143] In an exemplary embodiment, Figure 3 As shown, a cross-border payment intelligent risk control system 300 based on blockchain is provided, including:

[0144] The data collection and data verification module 301 is used to obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes, and store it on the blockchain. It also performs consistency verification on the transaction data in different blockchain networks to obtain verification results. The verification results are used to indicate whether the transaction data on different blockchains is consistent.

[0145] The risk feature extraction module 302 is used to analyze the transaction data on the blockchain to obtain risk features when the verification result is consistent;

[0146] The risk judgment module 303 is used to judge whether there is a potential risk in the cross-border payment transaction based on the risk characteristics and obtain a judgment result;

[0147] The risk information acquisition module 304 is used to obtain risk information based on the judgment result;

[0148] The decision execution module 305 is used to trigger corresponding intervention measures based on risk information and pre-set risk control strategies, and at the same time save the risk information and intervention records corresponding to the pre-set risk control strategies on the chain; intervention measures include suspending transactions, requesting additional verification information, and notifying relevant personnel for manual review.

[0149] Furthermore, the data collection and data verification module 301 is also used to:

[0150] Obtain transaction data related to cross-border payments; transaction data includes transaction amount, identity information of both parties, transaction time, transaction purpose, and recipient bank information;

[0151] Utilize the encryption function of the blockchain node and adopt a specific encryption algorithm to convert the acquired transaction data into ciphertext form, and then store the encrypted transaction data on the blockchain;

[0152] When there are multiple blockchain networks involved in cross-border payment transactions, consistency verification is performed on the transaction data of the cross-border payments in each blockchain network to verify whether the data of different transaction data copies are consistent and obtain the verification results.

[0153] Furthermore, the risk feature extraction module 302 is further configured to:

[0154] Use graph neural network models to extract risk features of transaction data on blockchain;

[0155] Among them, the graph neural network model includes:

[0156] Input layer, used to input node features and edge information in the blockchain transaction graph;

[0157] The graph convolution layer is used to update the feature representation of the node by aggregating the information of neighboring nodes. It uses multiple layers of graph convolution layers, and the output of each layer serves as the input of the next layer. The graph convolution network update formula in the multi-layer graph convolution layer is: Among them H (l) is the node feature matrix of the lth layer, is the adjacency matrix with self-loops added, is the corresponding degree matrix, W (l) is the weight matrix of the lth layer, σ is the nonlinear activation function;

[0158] The output layer is used to output the risk feature representation of each node; the risk features are used in the subsequent risk assessment model.

[0159] Furthermore, the risk judgment module 303 is further configured to:

[0160] Use the logistic regression model to determine whether there are potential risks in cross-border payment transactions and obtain the judgment results;

[0161] Among them, the logistic regression model includes:

[0162] Input layer, used to input risk characteristics;

[0163] The logistic regression layer is used to calculate the probability that a transaction is a risky transaction based on the input risk characteristics through a logistic regression function; the logistic regression function is: Where P(Y=1|X) represents the probability of risk in cross-border payment transactions, X1, X2, ..., X n is the risk feature of the input, β0, β1, ..., β n are model parameters, learned through training data;

[0164] The output layer is used to determine that there is a potential risk when the probability value is greater than the set threshold, and output the judgment result on whether there is a potential risk in the cross-border payment transaction.

[0165] Furthermore, the risk information acquisition module 304 is further configured to:

[0166] When a cross-border payment transaction is judged to have potential risks, a risk score is assigned to the cross-border payment transaction based on the preset risk assessment rules;

[0167] Based on the risk score, the risk level of the cross-border payment transaction is obtained;

[0168] Generate risk information based on risk level.

[0169] Furthermore, the decision execution module 305 is further configured to:

[0170] Match risk information with pre-set risk control strategies to determine whether intervention measures need to be triggered;

[0171] According to the risk level, different intervention measures are matched and executed through smart contracts;

[0172] Record the execution of intervention measures and generate intervention records; intervention records include transaction ID, intervention measures, execution time and execution results;

[0173] Encrypt the intervention record and write it into the blockchain through the blockchain consensus mechanism;

[0174] The intervention records are verified through the hash verification mechanism of the blockchain to obtain the verification results.

[0175] Furthermore, the system also includes:

[0176] The event monitoring module is used to monitor the on-chain status of risk information and intervention records in real time through the blockchain's event monitoring mechanism, and generate audit logs. The audit logs include the entire process of risk information collection, risk control strategy triggering, intervention measure execution, and on-chain recording.

[0177] The audit module is used to regularly audit risk information and intervention records on the blockchain according to a preset time period and obtain audit results;

[0178] The optimization and adjustment module is used to analyze audit results, adjust pre-set risk control strategies and risk assessment models, and optimize intervention measures;

[0179] The model update module is used to update the risk assessment model based on market changes and new risk characteristics.

[0180] In one embodiment, Figure 4 A computer device 400 is provided, comprising:

[0181] at least one processor 401;

[0182] and a memory 402 communicatively connected to at least one of the processors 401;

[0183] The memory stores application code that can be executed by at least one of the processors, and the application code is executed by at least one of the processors so that at least one of the processors can perform the steps of the blockchain-based cross-border payment intelligent risk control method as described above.

[0184] The computer device may further include a transceiver 403 .

[0185] The processor 401, the memory 402 and the transceiver 403 may be connected via a bus 404 or other means. In the figure, the bus 404 is used as an example. Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0186] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0187] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0188] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A cross-border payment intelligent risk control method based on blockchain, characterized in that: The method comprises: Obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes, and then store it on-chain. Perform consistency checks on the transaction data in different blockchain networks to obtain verification results; the verification results are used to indicate whether the transaction data on different blockchains is consistent; When the verification results are consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics; Based on the risk characteristics, determine whether the cross-border payment transaction has potential risks and obtain a determination result; Based on the judgment result, risk information is obtained; Based on the risk information, according to the pre-set risk control strategy, corresponding intervention measures are triggered, and the risk information and the intervention records corresponding to the pre-set risk control strategy are saved on the chain; the intervention measures include suspending transactions, requiring additional verification information and notifying relevant personnel for manual review.

2. The method according to claim 1, characterized in that The method of obtaining transaction data related to cross-border payments, encrypting the transaction data through blockchain nodes and then storing the data on the blockchain, and performing consistency verification on the transaction data in different blockchain networks to obtain verification results includes the following steps: Obtain transaction data related to cross-border payments; such transaction data includes transaction amount, identity information of both parties to the transaction, transaction time, transaction purpose, and recipient bank information; Utilize the encryption function of the blockchain node and adopt a specific encryption algorithm to convert the acquired transaction data into ciphertext form, and then store the encrypted transaction data on the blockchain; When there are multiple blockchain networks participating in a cross-border payment transaction, a consistency check is performed on the transaction data of the cross-border payment in each of the blockchain networks to verify whether the data of different transaction data copies are consistent and obtain a verification result.

3. The method according to claim 1, characterized in that When the verification results are consistent, the transaction data on the blockchain is analyzed to obtain risk characteristics, including: Using a graph neural network model to extract risk features of transaction data on the blockchain; The graph neural network model includes: Input layer, used to input node features and edge information in the blockchain transaction graph; The graph convolution layer is used to update the feature representation of the node by aggregating the information of neighboring nodes. It uses multiple layers of graph convolution layers, and the output of each layer serves as the input of the next layer. The graph convolution network update formula in the multi-layer graph convolution layer is: Among them H (l) is the node feature matrix of the lth layer, is the adjacency matrix with self-loops added, is the corresponding degree matrix, W (l) is the weight matrix of the lth layer, σ is the nonlinear activation function; The output layer is used to output the risk feature representation of each node; the risk feature is used in the subsequent risk assessment model.

4. The method according to claim 1, wherein The determining whether the cross-border payment transaction has potential risks based on the risk characteristics and obtaining a determination result include: Using a logistic regression model, determine whether the cross-border payment transaction has potential risks and obtain a determination result; Wherein, the logistic regression model includes: An input layer, used for inputting the risk characteristics; The logistic regression layer is used to calculate the probability that a transaction is a risky transaction using a logistic regression function based on the input risk characteristics; the logistic regression function is: Where P(Y=1|X) represents the probability of risk in cross-border payment transactions, X1, X2, ..., X n is the risk feature of the input, β0, β1, ..., β n are model parameters, learned through training data; The output layer is used to determine that there is a potential risk when the probability value is greater than the set threshold, and output the judgment result of whether the cross-border payment transaction has a potential risk.

5. The method according to claim 1, characterized in that The step of obtaining risk information based on the judgment result includes the following steps: When it is determined that the cross-border payment transaction has potential risks, a risk score is assigned to the cross-border payment transaction according to a preset risk assessment model; Obtaining a risk level for the cross-border payment transaction based on the risk score; Risk information is generated according to the risk level.

6. The method according to claim 1, characterized in that Based on the risk information, according to the pre-set risk control strategy, corresponding intervention measures are triggered, and the risk information and intervention records are saved on the chain, including: Matching the risk information with the pre-set risk control strategy to determine whether intervention measures need to be triggered; According to the risk level, different intervention measures are matched and executed through smart contracts; Recording the execution of intervention measures and generating intervention records; the intervention records include transaction ID, intervention measures, execution time, and execution results; Encrypting the intervention record and writing the encrypted intervention record into the blockchain through the consensus mechanism of the blockchain; The intervention record is data verified through the hash verification mechanism of the blockchain to obtain a verification result.

7. The method according to claim 1, characterized in that Also includes: Through the blockchain’s event monitoring mechanism, the on-chain status of risk information and intervention records is monitored in real time, generating audit logs. The audit logs include the entire process of risk information collection, risk control strategy triggering, intervention measures execution, and on-chain recording. Regularly audit the risk information and intervention records on the blockchain according to a preset time period to obtain audit results; Analyze the audit results, adjust the pre-set risk control strategy and the risk assessment model, and optimize intervention measures; The risk assessment model is updated based on market changes and new risk characteristics.

8. A blockchain-based cross-border payment intelligent risk control system, characterized by: The system comprises: The data collection and data verification module is used to obtain transaction data related to cross-border payments, encrypt the transaction data through blockchain nodes, and store it on the blockchain. It also performs consistency verification on transaction data in different blockchain networks to obtain verification results; the verification results are used to indicate whether the transaction data on different blockchains is consistent; a risk feature extraction module, configured to analyze the transaction data on the blockchain to obtain risk features when the verification result is consistent; A risk judgment module is used to judge whether the cross-border payment transaction has potential risks based on risk characteristics and obtain a judgment result; A risk information acquisition module, configured to obtain risk information based on the judgment result; Decision-making execution module: used to trigger corresponding intervention measures based on the risk information and according to the pre-set risk control strategy, and at the same time save the intervention records corresponding to the risk information and the pre-set risk control strategy on the chain; the intervention measures include suspending transactions, requiring additional verification information and notifying relevant personnel for manual review.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

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

Images