Blockchain-based Third-party Payment Security Authentication Method and System

Abstract

The present invention relates to the field of secure payment technologies, and particularly to a third-party payment security authentication method and system based on blockchain. The method includes the following steps: obtaining real-time payment information of a user; generating multi-dimensional payment information based on the real-time payment information of the user; performing time series fitting on the multi-dimensional payment information to generate a payment behavior sequence; obtaining local dependencies of the payment sequence based on the payment behavior sequence; performing implicit risk feature analysis on the payment behavior sequence by using the local dependencies of the payment sequence to obtain implicit payment risk behavior feature data; obtaining a plurality of payment behavior nodes based on the payment behavior sequence; dynamically connecting the plurality of payment behavior nodes through a topological network by using the implicit payment risk behavior feature data to construct a payment behavior node blockchain network; performing risk probability distribution analysis on the payment behavior node blockchain network to obtain node risk probability distribution data; the present invention realizes a highly secure authentication method.

Description

Technical Field

[0001] The present invention relates to the technical field of secure payment, and particularly to a third-party payment security authentication method and system based on blockchain. Background Art

[0002] In recent years, as a fast and secure payment method, third-party payment has become an indispensable and important part of modern e-commerce. With the continuous expansion of the payment scale and the increasing diversification of payment needs, the traditional third-party payment security authentication mechanism has been difficult to meet the increasingly complex security requirements. Most existing third-party payment systems adopt a centralized authentication mode. The payment platform, as a centralized third-party institution, undertakes the responsibilities of payment verification and fund custody. This mode has problems such as high single-point failure risk, difficulty in ensuring privacy and data security. Once the payment platform system fails or is attacked, it will cause serious losses to a large number of users and merchants. With the continuous emergence of emerging payment scenarios such as mobile payment and cross-border payment, security issues in aspects such as user identity recognition, transaction information transmission, and privacy protection have become increasingly prominent. Existing payment authentication methods, such as passwords and SMS verification codes, have disadvantages such as low security and easy to be stolen. Therefore, an intelligent third-party payment security authentication method is needed. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention proposes a third-party payment security authentication method and system based on blockchain to solve at least one of the above technical problems.

[0004] To achieve the above object, the present invention provides a third-party payment security authentication method based on blockchain, including the following steps:

[0005] Step S1: Obtain the user's real-time payment information; generate multi-dimensional payment information based on the user's real-time payment information; perform time series fitting on the multi-dimensional payment information to generate a payment behavior sequence;

[0006] Step S2: Obtain the local dependence relationship of the payment sequence based on the payment behavior sequence; perform implicit risk feature analysis on the payment behavior sequence using the local dependence relationship of the payment sequence to obtain implicit payment risk behavior feature data;

[0007] Step S3: Obtain multiple payment behavior nodes based on the payment behavior sequence; perform dynamic topological network connection on the multiple payment behavior nodes using the implicit payment risk behavior feature data to construct a payment behavior node blockchain network;

[0008] Step S4: Analyze the risk probability distribution of the payment behavior node blockchain network to obtain node risk probability distribution data; perform risk propagation dynamic evolution analysis based on the node risk probability distribution data to generate risk propagation evolution data;

[0009] Step S5: Perform individual risk trend quantitative prediction based on the risk propagation evolution data to obtain payment node risk trend prediction values; mark high-risk payment nodes based on the payment node risk trend prediction values; reconstruct the global potential association path for high-risk payment nodes to obtain a payment risk deduction chain;

[0010] Step S6: Perform risk monitoring decision mining on the payment behavior node blockchain network according to the payment risk deduction chain to construct an intelligent payment risk monitoring engine; execute user payment operations based on the intelligent payment risk monitoring engine.

[0011] In this specification, a third-party payment security authentication system based on blockchain is also provided, including:

[0012] A behavior sequence module, configured to obtain real-time payment information of a user; generate multi-dimensional payment information based on the real-time payment information of the user; perform time series fitting on the multi-dimensional payment information to generate a payment behavior sequence;

[0013] A risk behavior feature module, configured to obtain local dependence relationships of the payment sequence based on the payment behavior sequence; perform implicit risk feature analysis on the payment behavior sequence using the local dependence relationships of the payment sequence to obtain implicit payment risk behavior feature data;

[0014] A topology network module, configured to obtain multiple payment behavior nodes based on the payment behavior sequence; perform dynamic topology network connection on the multiple payment behavior nodes using the implicit payment risk behavior feature data to construct a payment behavior node blockchain network;

[0015] A risk dynamic evolution module, configured to analyze the risk probability distribution of the payment behavior node blockchain network to obtain node risk probability distribution data; perform risk propagation dynamic evolution analysis based on the node risk probability distribution data to generate risk propagation evolution data;

[0016] A risk trend prediction module, configured to perform individual risk trend quantitative prediction based on the risk propagation evolution data to obtain payment node risk trend prediction values; mark high-risk payment nodes based on the payment node risk trend prediction values; reconstruct the global potential association path for high-risk payment nodes to obtain a payment risk deduction chain;

[0017] A risk monitoring module, configured to perform risk monitoring decision mining on the distributed behavior node network according to the payment risk deduction chain to construct an intelligent payment risk monitoring engine; execute user payment operations based on the intelligent payment risk monitoring engine.

[0018] The third-party payment security authentication method and system based on blockchain provided by the present invention have the following beneficial effects:

[0019] The present invention obtains the real-time payment information of users, timely understands the payment behaviors and transaction situations of users, provides a data basis for subsequent risk analysis and decision-making, generates payment information including multiple dimensions such as transaction amount, transaction time, and transaction object based on the real-time payment information of users, enriches the description and analysis of payment behaviors, arranges the payment information in chronological order by performing time series fitting on the multi-dimensional payment information to generate a payment behavior sequence, which helps analyze the evolution and trend of payment behaviors, analyzes the local dependence relationship between payment behaviors based on the payment behavior sequence, that is, the influence degree of a certain payment behavior on other payment behaviors, which helps understand the mutual relationship and influence between payment behaviors, uses the local dependence relationship of the payment sequence to perform implicit risk feature analysis on the payment behavior sequence, discovers potential payment risk behaviors by analyzing the associations and features between payment behaviors, and extracts relevant risk feature data, takes each payment behavior in the sequence as a node based on the payment behavior sequence to obtain multiple payment behavior nodes, which helps the individual analysis and modeling of payment behaviors, uses the implicit payment risk behavior feature data to perform dynamic topological network connection on multiple payment behavior nodes, reveals the network structure and relevance between payment behaviors by establishing the connection relationship between nodes, helps understand the complexity of payment behaviors and the path of risk propagation, obtains the risk probability distribution data of each node by performing risk probability distribution analysis on the payment behavior node blockchain network, which helps evaluate the risk degree of each node and the possibility of potential risks, performs dynamic evolution analysis of risk propagation based on the node risk probability distribution data, understands the propagation process and evolution trend of risks in the network by analyzing the risk propagation path and probability change between nodes, provides a basis for risk prediction and decision-making, performs risk trend quantitative prediction on each payment node one by one based on the risk propagation evolution data, which helps identify possible high-risk payment nodes and predict the development trend of their risks, marks high-risk payment nodes based on the predicted values of payment node risk trends, which helps quickly identify and pay attention to potential high-risk payment behaviors, reconstructs the global potential association path of high-risk payment nodes to reveal the associations and paths between high-risk payment nodes, which helps understand the deduction chain of payment risks and the influence and propagation paths between different nodes, performs risk monitoring decision mining on the payment behavior node blockchain network based on the payment risk deduction chain, formulates corresponding risk monitoring strategies and decision rules by analyzing the deduction chain of payment risks and the associations between nodes, constructs an intelligent payment risk monitoring engine by integrating the results of risk monitoring decision mining, the engine automatically monitors the payment behaviors of users and performs real-time risk assessment and decision-making according to preset rules and algorithms, executes the payment security authentication operation of users based on the results of the intelligent payment risk monitoring engine, and the engine performs real-time interception, warning or other necessary operations on high-risk payment behaviors according to the risk assessment results to ensure payment security authentication and risk control. Description of the Drawings

[0020] Figure 1 It is a schematic flow chart of the steps of a method for third-party payment security authentication based on blockchain according to the present invention;

[0021] Figure 2 It is a schematic flow chart of the detailed implementation steps of step S1;

[0022] Figure 3 It is a schematic flow chart of the detailed implementation steps of step S2;

[0023] Figure 4 It is a schematic flow chart of the detailed implementation steps of step S3. Detailed Implementation Manner

[0024] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] The embodiments of the present application provide a method and system for third-party payment security authentication based on blockchain. The execution subjects of the method and system for third-party payment security authentication based on blockchain include but are not limited to: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. that can be regarded as general computing nodes of the present application. The data processing platform includes but is not limited to at least one of: audio and image management systems, information management systems, and cloud data management systems.

[0026] Please refer to Figures 1 to 4 , the present invention provides a method for third-party payment security authentication based on blockchain, and the method for third-party payment security authentication based on blockchain includes the following steps:

[0027] To achieve the above object, the present invention provides a method for third-party payment security authentication based on blockchain, including the following steps:

[0028] Step S1: Obtain the user's real-time payment information; generate multi-dimensional payment information based on the user's real-time payment information; perform time series fitting on the multi-dimensional payment information to generate a payment behavior sequence;

[0029] Step S2: Obtain the local dependence relationship of the payment sequence based on the payment behavior sequence; perform implicit risk feature analysis on the payment behavior sequence by using the local dependence relationship of the payment sequence to obtain implicit payment risk behavior feature data;

[0030] Step S3: Obtain multiple payment behavior nodes based on the payment behavior sequence; perform dynamic topological network connection on the multiple payment behavior nodes by using the implicit payment risk behavior feature data to construct a distributed behavior node network;

[0031] Step S4: Conduct a risk probability distribution analysis on the distributed behavior node network to obtain node risk probability distribution data; perform a risk propagation dynamic evolution analysis based on the node risk probability distribution data to generate risk propagation evolution data;

[0032] Step S5: Perform a quantitative prediction of each risk trend based on the risk propagation evolution data to obtain the risk trend prediction value of the payment node; mark the high-risk payment nodes based on the risk trend prediction value of the payment node; reconstruct the global potential association path for the high-risk payment nodes to obtain the payment risk deduction chain;

[0033] Step S6: Conduct risk monitoring decision mining on the distributed behavior node network according to the payment risk deduction chain to construct an intelligent payment risk monitoring engine; execute user payment operations based on the intelligent payment risk monitoring engine.

[0034] The present invention obtains the real-time payment information of users, timely understands the payment behaviors and transaction situations of users, and provides a data basis for subsequent risk analysis and decision-making. Based on the real-time payment information of users, payment information including multiple dimensions such as transaction amount, transaction time, and transaction object is generated, enriching the description and analysis of payment behaviors. By performing time series fitting on the multi-dimensional payment information, the payment information is arranged in chronological order to generate a payment behavior sequence, which helps analyze the evolution and trend of payment behaviors. Based on the payment behavior sequence, the local dependence relationship between payment behaviors, that is, the influence degree of a certain payment behavior on other payment behaviors, is analyzed, which helps understand the mutual relationship and influence between payment behaviors. Using the local dependence relationship of the payment sequence, implicit risk feature analysis is performed on the payment behavior sequence. By analyzing the associations and features between payment behaviors, potential payment risk behaviors are discovered, and relevant risk feature data is extracted. Based on the payment behavior sequence, each payment behavior in the sequence is used as a node to obtain multiple payment behavior nodes, which helps the individual analysis and modeling of payment behaviors. Using the implicit payment risk behavior feature data, dynamic topological network connections are made for multiple payment behavior nodes. By establishing the connection relationships between nodes, the network structure and relevance between payment behaviors are revealed, helping to understand the complexity of payment behaviors and the path of risk propagation. By performing risk probability distribution analysis on the payment behavior node blockchain network, risk probability distribution data for each node is obtained, which helps evaluate the risk degree of each node and the possibility of potential risks. Based on the node risk probability distribution data, dynamic evolution analysis of risk propagation is carried out. By analyzing the risk propagation path and probability change between nodes, the propagation process and evolution trend of risks in the network are understood, providing a basis for risk prediction and decision-making. Based on the risk propagation evolution data, risk trend quantification prediction is performed for each payment node one by one, which helps identify possible high-risk payment nodes and predict the development trend of their risks. Based on the predicted values of the payment node risk trends, high-risk payment nodes are marked, which helps quickly identify and pay attention to potential high-risk payment behaviors. The global potential association path of high-risk payment nodes is reconstructed to reveal the associations and paths between high-risk payment nodes, which helps understand the deduction chain of payment risks and the influence and propagation paths between different nodes. Based on the payment risk deduction chain, risk monitoring decision mining is carried out on the payment behavior node blockchain network. By analyzing the deduction chain of payment risks and the associations between nodes, corresponding risk monitoring strategies and decision rules are formulated. By integrating the results of risk monitoring decision mining, an intelligent payment risk monitoring engine is constructed. This engine automatically monitors the payment behaviors of users and performs real-time risk assessment and decision-making according to pre-set rules and algorithms. Based on the results of the intelligent payment risk monitoring engine, the payment security authentication operation of users is executed. The engine performs real-time interception, warning or other necessary operations on high-risk payment behaviors according to the risk assessment results to ensure payment security authentication and risk control.

[0035] In the embodiment of the present invention, refer to Figure 1 , is a flowchart of a third-party payment security authentication method based on blockchain of the present invention. In this example, the steps of the third-party payment security authentication method based on blockchain include:

[0036] Step S1: obtaining user real-time payment information; generating multi-dimensional payment information based on the user real-time payment information; performing time series fitting on the multi-dimensional payment information to generate a payment behavior sequence;

[0037] In this embodiment, after obtaining the authorization of the user to the platform, the user's payment transaction data, including transaction amount, transaction time, transaction object, transaction type and other information, is collected in real time from the user's payment system, financial system and other channels. The collected payment information is cleaned and preprocessed to ensure data integrity and accuracy. According to the collected payment transaction data, a payment information matrix including multiple dimensions such as transaction amount, transaction frequency, transaction object, transaction type, and transaction time is constructed. The payment information matrix is ​​feature extracted and dimensionally reduced using methods such as principal component analysis to generate a more compact multi-dimensional payment information representation. The generated multi-dimensional payment information is arranged in chronological order to form time series data of user payment behavior. Time series analysis methods such as ARIMA and exponential smoothing are applied to the time series data for fitting to obtain a time series model of payment behavior. The user's future payment behavior trend is predicted through the time series model, providing a basis for subsequent payment risk analysis.

[0038] Step S2: obtaining a local dependency of the payment sequence based on the payment behavior sequence; performing implicit risk feature analysis on the payment behavior sequence using the local dependency of the payment sequence to obtain implicit payment risk behavior feature data;

[0039] In this embodiment, time series analysis methods, such as Granger causality test, are used to analyze the local dependencies between payment behaviors in the payment behavior sequence. This local dependency reflects the inherent logic of the user's payment behavior and provides a basis for the subsequent implicit risk feature analysis. Combined with the local dependencies of the payment sequence, in-depth analysis is performed on abnormal patterns, abnormal nodes, etc. in the payment behavior sequence to mine hidden payment risk features. Abnormal patterns reflect abnormal correlations between payment behaviors, and abnormal nodes may indicate payment behaviors with hidden payment risk hazards. Anomaly detection, abnormal pattern mining and other methods are used to extract implicit risk feature data from the payment behavior sequence. The implicit risk features such as abnormal patterns and abnormal nodes obtained from the above analysis are integrated into a payment risk behavior feature data set. These feature data will provide an important basis for the subsequent construction of the payment risk network and risk prediction.

[0040] Step S3: Obtain multiple payment behavior nodes based on the payment behavior sequence; use the implicit payment risk behavior feature data to perform dynamic topological network connection on the multiple payment behavior nodes, and construct a distributed behavior node network;

[0041] In this embodiment, from the user's payment behavior sequence, each independent payment behavior event is identified as a payment behavior node. Each payment behavior node contains key information such as user ID, payment time, payment amount, and payment merchant. Through the analysis of the payment behavior sequence, multiple independent payment behavior nodes are obtained. Based on the obtained implicit payment risk behavior feature data, the risk features of each payment behavior node are analyzed, and the nodes with relevant risk features are dynamically topologically connected to construct an association network between the nodes. This dynamic association based on implicit risk features reflects the potential connections between payment behaviors, rather than relying only on surface payment information. The above dynamically associated payment behavior nodes are organized into a distributed peer-to-peer network structure. Each node, as an independent participant in the network, stores its own payment behavior data and risk features. Nodes exchange and spread dynamic risk information through encrypted peer-to-peer communication. The entire network adopts a decentralized architecture, with high autonomy and fault tolerance.

[0042] Step S4: Perform risk probability distribution analysis on the distributed behavior node network to obtain node risk probability distribution data; perform dynamic evolution analysis of risk propagation based on the node risk probability distribution data to generate risk propagation evolution data;

[0043] In this embodiment, using graph theory and probability models, the constructed distributed behavior node network is deeply analyzed. According to the topological features of the network nodes, such as metrics like degree centrality, PageRank, etc., combined with implicit risk features, the risk probability distribution of each node is calculated to obtain a complete node risk probability distribution data, providing a basis for subsequent risk propagation analysis. A network-based propagation model, such as SIR, SIS, etc., is used to simulate the dynamic propagation process of risk in the network. Using the previously obtained node risk probability distribution data as the initial condition, through iterative simulation calculations, the propagation and evolution process of risk in the network is predicted, and the characteristics such as the speed, scope, and intensity of risk propagation are analyzed to form risk propagation evolution data. The results of the dynamic evolution analysis of risk propagation are integrated into a risk propagation evolution data set. The data set contains information such as the risk probability distribution, risk propagation path, and propagation intensity at each time point in the network.

[0044] Step S5: Quantitatively predict the risk trend of each risk one by one based on the risk propagation and evolution data to obtain the predicted value of the risk trend of the payment node; mark the high-risk payment nodes based on the predicted value of the risk trend of the payment node; reconstruct the global potential association path for the high-risk payment nodes to obtain the payment risk deduction chain;

[0045] In this embodiment, the risk trend of each payment node in the network is quantitatively predicted one by one. Methods such as time series analysis and machine learning are used, combined with the historical risk probability distribution of the node, to predict the risk change trend of the node within a certain period of time in the future. In this way, the predicted value of the risk trend of each payment node is obtained, providing a basis for the subsequent identification of high-risk nodes. According to the obtained predicted value of the risk trend of the payment node, a suitable risk threshold is set to identify the high-risk payment nodes. High-risk payment nodes refer to the nodes whose predicted value of the risk trend exceeds the preset threshold, indicating that these nodes may face greater payment risks in the future. These high-risk payment nodes are focused on and analyzed to provide targeted support for subsequent risk control. Based on the identified high-risk payment nodes, further analyze their global potential association paths in the entire payment behavior network, and use graph theory analysis methods such as the shortest path algorithm and community detection to discover the relevance between these high-risk nodes and other nodes in the network, forming the payment risk deduction chain of high-risk nodes, and revealing the risk propagation path and evolution mechanism in the network.

[0046] Step S6: Mine the risk monitoring decision for the distributed behavior node network according to the payment risk deduction chain, and construct an intelligent payment risk monitoring engine; execute the user payment operation based on the intelligent payment risk monitoring engine.

[0047] In this embodiment, according to the payment risk deduction chain, deeply analyze the risk characteristics and correlation rules of each link in the network, combine industry experience and expert knowledge, design targeted risk monitoring rules and decision logics, and use technologies such as machine learning and knowledge graphs to automatically mine and optimize the risk monitoring decision logic, forming a set of intelligent risk monitoring decision mechanisms for adaptive learning. Integrate the above risk monitoring decision mechanisms into an intelligent payment risk monitoring engine. This engine can real-time monitor the risk changes in the payment network, timely warn and identify high-risk nodes. The engine is built-in with complex risk propagation analysis models and decision rule libraries, and has strong autonomous learning and decision-making capabilities. Deploy the constructed intelligent payment risk monitoring engine in the user payment system to real-time monitor and control payment risks. The engine will comprehensively analyze the real-time data of the payment network, identify potential high-risk nodes and risk propagation chains, and combine the user's risk preferences to automatically execute risk control strategies such as intercepting high-risk transactions and adjusting payment paths. Through continuous operation and optimization, the engine can continuously improve the security and efficiency of the user payment operation.

[0048] In this embodiment, refer to Figure 2 which is a schematic diagram of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0049] Step S11: Obtain the user's real-time payment information;

[0050] Step S12: Identify the payment attributes of the user's real-time payment information to obtain payment attribute data;

[0051] Step S13: Perform multi-dimensional partitioning on the user's real-time payment information based on the payment attribute data to generate multi-dimensional payment information;

[0052] Step S14: Mine the payment behavior characteristics of the multi-dimensional payment information to obtain payment behavior characteristic data;

[0053] Step S15: Perform time series fitting on the payment behavior characteristic data to generate a payment behavior sequence.

[0054] In this embodiment, payment transaction data is collected in real time from the user payment system, including key information such as payment time, transaction amount, payment method, transaction location, etc. A data collection channel is established to ensure the timely, accurate, and complete collection of payment data. The collected payment data is subjected to preliminary cleaning and formatting processing to lay a foundation for subsequent payment attribute identification and multi-dimensional partitioning. According to the semantic and structural characteristics of the payment data, payment attribute identification rules are designed, and technologies such as text analysis and knowledge graphs are used to automatically identify and label attributes such as payment method, transaction object, and payment purpose in the payment data to form structured payment attribute data, providing a basis for the subsequent generation of multi-dimensional payment information. Combining business requirements, a multi-dimensional partitioning scheme for payment information is designed, such as classification according to dimensions such as payment method, transaction object, and geographical location. According to the obtained payment attribute data, the real-time payment information is multi-dimensionally segmented and aggregated to generate a multi-dimensional payment information view, providing a data basis for subsequent payment behavior characteristic mining. Data mining and machine learning technologies are applied to mine various characteristics of payment behavior, such as transaction frequency, transaction amount distribution, and payment period habits, for the multi-dimensional payment information. Combining expert experience, a targeted feature extraction algorithm is designed to extract the key features of the payment behavior to form structured payment behavior characteristic data, laying a foundation for subsequent time series analysis. The obtained payment behavior characteristic data is organized into a time series in chronological order, and time series analysis methods such as autoregressive models and ARIMA models are applied to model and predict the payment behavior characteristic sequence, generating a time series prediction result of the payment behavior, reflecting the evolution law of the payment behavior in the time dimension.

[0055] In this embodiment, refer to Figure 3, which is a schematic diagram of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0056] Step S21: Perform time-domain convolution filtering on the payment behavior sequence to obtain the local dependence relationship of the payment sequence;

[0057] Step S22: Perform recursive feature transfer on the local dependence relationship of the payment sequence to extract the global trend law;

[0058] Step S23: Use the global trend law to perform implicit risk feature analysis on the payment behavior sequence to obtain implicit payment risk behavior feature data.

[0059] In this embodiment, starting from the payment behavior time series, the time-domain convolution filtering technology is applied to process the sequence. Time-domain convolution can identify the local correlation and dependence relationship in the sequence, highlight the short-term behavior patterns within the time window, and extract the local features in the payment behavior sequence by designing appropriate convolution kernels and filtering parameters. These local dependence relationships provide the basis for subsequent global trend analysis. Through recursive feature transfer and aggregation, using technologies such as graph neural networks, a global association map of the payment behavior sequence is established to capture the overall trend in the sequence. Feature propagation and aggregation are performed on the association map to identify the latent global laws in the payment behavior sequence. A targeted implicit risk feature extraction method is designed. By deeply analyzing the abnormal patterns, abnormal fluctuations, abnormal associations, etc. in the payment behavior sequence, the hidden risk signals are mined. These implicit risk features may be difficult to directly observe and identify, but contain important risk warning information. The extracted implicit risk behavior feature data is used as an important input for subsequent intelligent risk prediction.

[0060] In this embodiment, refer to Figure 4 , which is a schematic diagram of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0061] Step S31: Divide the payment behavior sequence into timestamp nodes to obtain multiple payment behavior nodes;

[0062] Step S32: Use the implicit payment risk behavior feature data to perform implicit feature virtual projection on multiple payment behavior nodes to obtain implicit feature behavior nodes;

[0063] Step S33: Calculate the feature similarity of the implicit feature behavior nodes to obtain the feature similarity between nodes;

[0064] Step S34: Analyze the node link association strength of the implicit feature behavior nodes based on the feature similarity between nodes to obtain the node association strength data;

[0065] Step S35: Based on the node association strength data, perform dynamic topological network connection on the implicit feature behavior nodes to construct a distributed behavior node network.

[0066] In this embodiment, the sequence is divided into multiple discrete nodes according to timestamps. Each node represents the payment behavior state at a specific time point and contains various payment features at that time. Through this timestamp-based node formation, the continuous payment behavior sequence is transformed into a discrete node set. The extracted implicit payment risk behavior features are virtually projected onto each corresponding payment behavior node. Through this virtual projection of implicit features, each discrete payment behavior node is given an implicit risk feature. These implicit feature behavior nodes can more comprehensively depict the potential risk information in the payment process, which provides an important feature basis for subsequent node association analysis and dynamic network construction. Calculate the feature similarity between implicit feature behavior nodes using common similarity measurement methods such as cosine similarity and Euclidean distance. This inter-node feature similarity reflects the correlation degree of payment behavior nodes in terms of implicit risk features. Construct the association strength between nodes, and assign stronger association strength to nodes with high feature similarity, and vice versa with weaker association strength. Through this node association strength analysis based on implicit features, the potential connections between payment behavior nodes are mined, and the dynamic topological network connection between payment behavior nodes is established. Nodes with stronger association strength are connected together in a closer manner to construct a distributed behavior node network. This network structure can better depict the dynamic evolution process of payment behavior and reveal its internal association rules. The constructed distributed behavior node network provides an important topological basis for subsequent intelligent risk prediction and decision-making.

[0067] In this embodiment, the specific steps of step S35 are as follows:

[0068] Perform contract analysis on the implicit feature behavior nodes based on the node association strength data to generate smart contract logic;

[0069] Perform smart contract editing processing on the implicit feature behavior nodes according to the smart contract logic to obtain a payment behavior smart contract;

[0070] Reconstruct the dynamic topological network architecture of the implicit feature behavior nodes through the payment behavior smart contract to construct a payment behavior node blockchain network.

[0071] In this embodiment, features such as transaction frequency, amount, and time among nodes are analyzed, the association strength between nodes is calculated, node pairs with potential risk connections are identified based on the level of association strength, attribute indicators such as the credit rating and risk-bearing capacity of nodes are determined, rules for the rights, obligations, and behavior constraints of nodes in the network are formulated, including information sharing strategies, transaction approval mechanisms, fund transfer controls, etc., an independent smart contract instance is created for each node to ensure that the content of the smart contract of each node matches its own characteristics, the correctness and executability of the contract logic are checked to ensure that the contract content can accurately reflect the role and behavior limitations of the node in the network, the smart contract contracts of each node are deployed to the blockchain network, a peer-to-peer communication mechanism based on smart contracts is established between nodes, the change situation of the association strength between nodes is analyzed in real time, the execution of the smart contract is automatically triggered according to the change, and the rights and obligation relationships between nodes are adjusted to ensure that the network topology can be dynamically reconstructed with the change of payment behavior.

[0072] In this embodiment, step S4 includes the following steps:

[0073] Step S41: Conduct a quantitative risk assessment of each node in the distributed behavior node network based on payment behavior characteristic data to obtain payment node risk data;

[0074] Step S42: Conduct a risk probability distribution analysis on the payment node risk data to obtain node risk probability distribution data;

[0075] Step S43: Identify the risk propagation path response of the distributed behavior node network according to the node risk probability distribution data, thereby generating a node risk propagation path;

[0076] Step S44: Conduct a risk propagation dynamic evolution analysis on the node risk propagation path to generate risk propagation evolution data.

[0077] In this embodiment, a quantitative risk assessment is performed on each node in the constructed distributed behavior node network. Using machine learning or statistical models, based on the payment characteristic data of the nodes, the potential risk level of each node is predicted or evaluated. This quantitative assessment of node risk based on payment behavior can assign a clear risk value to each network node. The payment node risk data is statistically analyzed and modeled to obtain the probability distribution characteristics of node risks. A parametric probability distribution model (such as normal distribution, lognormal distribution, etc.) is used to fit the probability distribution of node risk data. This risk probability distribution analysis can comprehensively characterize the statistical characteristics of the risk levels of nodes in the network, identify possible risk propagation paths, and by simulating the propagation and diffusion process of the node risk probability distribution, discover the key paths where risks may spread in the network. This identification of propagation paths based on risk probability distribution can reveal the potential spread mechanism of risks in the payment behavior network. A dynamic model based on network diffusion is used to simulate the propagation mechanism and evolution trend under different node risk levels. Through this analysis of the dynamic evolution of risk propagation, the possible spread situation and evolution process of risks in the network are predicted.

[0078] In this embodiment, step S5 includes the following steps:

[0079] Step S51: Based on the risk propagation evolution data, perform a quantitative prediction of the risk trend for each node in the distributed behavior node network to obtain the predicted value of the payment node risk trend;

[0080] Step S52: Compare the predicted value of the payment node risk trend with a preset payment behavior risk prediction threshold. When the predicted value of the payment node risk trend is greater than or equal to the preset payment behavior risk prediction threshold, it is marked as a high-risk payment node;

[0081] Step S53: Trace the payment link of the high-risk payment node to obtain the risky payment link;

[0082] Step S54: Reconstruct the global potential association path for the risky payment link to obtain the payment risk deduction chain.

[0083] In this embodiment, a quantitative prediction of the risk trend is performed for each payment node in the distributed behavior node network. Methods such as time series analysis and machine learning are used to predict the risk change trend of each node within a certain period of time in the future based on the historical risk data of the node. This kind of risk trend prediction can provide a basis for subsequent risk threshold comparison and high-risk node identification. The predicted value of the payment node risk trend obtained describes the possible future risk level change of each node. A reasonable payment behavior risk prediction threshold is set as the basis for determining whether a node is a high-risk node. The predicted value of the payment node risk trend obtained is compared with the preset risk prediction threshold. When the predicted value of the risk trend of the node is greater than or equal to the preset threshold, it is marked as a high-risk payment node. This identification of high-risk nodes based on the threshold can effectively screen out potential high-risk payment behaviors in the network. For the payment nodes marked as high-risk, trace their payment links. Using the constructed distributed behavior node network, along the association relationship between payment nodes, trace and discover the complete payment links participated by high-risk nodes. This tracing of payment links can show the propagation and diffusion process of high-risk payment behaviors. Based on the risk payment links and the distributed behavior node network topology established in the previous steps, globally reconstruct the potential association paths in the entire network. By analyzing the interaction relationship and propagation mechanism between nodes, identify the key paths that may lead to the deduction of payment risks. This reconstruction of the global potential association paths can reveal the complex mechanism of risk evolution in the payment behavior network. The obtained payment risk deduction chain provides an important basis for subsequent intelligent risk early warning and decision-making.

[0084] In this embodiment, step S6 includes the following steps:

[0085] Step S61: Identify abnormal risk factors for the payment risk deduction chain to generate payment risk factors;

[0086] Step S62: Conduct risk monitoring decision mining on the distributed behavior node network according to the payment risk factors to construct an intelligent payment risk monitoring engine;

[0087] Step S63: Perform payment security authentication processing on the user's real-time payment information based on the intelligent payment risk monitoring engine to obtain a payment security authentication result; the payment security authentication result specifically includes a normal payment result and a payment result with risks;

[0088] Step S64: When the payment security authentication result is a normal payment result, allow the user to conduct a payment transaction to complete the user payment security authentication operation;

[0089] Step S65: When the payment security authentication result is a payment result with risks, generate a payment risk warning signal and prohibit the user from making a payment.

[0090] In this embodiment, methods such as data mining and machine learning are adopted to identify the key factors in the payment behavior network that may lead to risk propagation. The identification of such abnormal risk factors, including abnormal payment behavior patterns, suspicious transaction characteristics, key node risk status, etc., integrates and summarizes the identified abnormal risk factors into payment risk factors, providing a basis for subsequent risk monitoring decisions. Based on the obtained payment risk factors, technical means such as data analysis and machine learning are used to comprehensively monitor the entire distributed behavior node network. By deeply mining the risk characteristics in aspects such as node behavior, transaction patterns, and association relationships in the network, a monitoring engine capable of intelligently identifying payment risks is constructed. This intelligent payment risk monitoring engine can real-time monitor user payment information, automatically detect and warn of potential risk hazards. The payment information generated by users in real time is input into the constructed intelligent payment risk monitoring engine for analysis. The monitoring engine will combine payment risk factors to conduct real-time intelligent identification and risk assessment of user payment behavior. According to the monitoring results, the payment security authentication results are divided into two types: normal payment results and payment results with risks. This payment risk identification can timely discover and identify abnormal payment behaviors in the network. When the payment risk identification result is a normal payment result, it indicates that there are no risk hazards in this payment behavior, and the user is allowed to smoothly conduct the payment transaction and execute the corresponding payment operation. If the payment risk identification result is a normal payment result, it indicates that there are no risk hazards in this payment behavior, and the user is allowed to smoothly conduct the payment transaction and execute the corresponding payment operation.

[0091] In this specification, a third-party payment security authentication system based on blockchain is also provided, including:

[0092] A behavior sequence module, used to obtain real-time payment information of users; generate multi-dimensional payment information based on the real-time payment information of users; perform time series fitting on the multi-dimensional payment information to generate a payment behavior sequence;

[0093] A risk behavior feature module, used to obtain the local dependence relationship of the payment sequence based on the payment behavior sequence; perform implicit risk feature analysis on the payment behavior sequence using the local dependence relationship of the payment sequence to obtain implicit payment risk behavior feature data;

[0094] A topology network module, used to obtain multiple payment behavior nodes based on the payment behavior sequence; perform dynamic topology network connection on the multiple payment behavior nodes using the implicit payment risk behavior feature data to construct a payment behavior node blockchain network;

[0095] A risk dynamic evolution module, used to perform risk probability distribution analysis on the payment behavior node blockchain network to obtain node risk probability distribution data; perform risk propagation dynamic evolution analysis based on the node risk probability distribution data to generate risk propagation evolution data;

[0096] A risk trend prediction module, which is used to perform quantitative prediction of each risk trend based on risk propagation evolution data to obtain a predicted value of the payment node risk trend; mark high-risk payment nodes based on the predicted value of the payment node risk trend; reconstruct the global potential association path for the high-risk payment nodes to obtain a payment risk deduction chain;

[0097] A risk monitoring module, which is used to perform risk monitoring decision mining on the distributed behavior node network according to the payment risk deduction chain to construct an intelligent payment risk monitoring engine; execute user payment operations based on the intelligent payment risk monitoring engine.

[0098] Those skilled in the art clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, systems, and units refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0099] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application essentially, or the part that contributes to the prior art, or all or part of the technical solution is embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that store program codes.

[0100] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to cover all changes falling within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0101] As described above, these are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A third-party payment security authentication method based on blockchain, characterized in that: The following steps are involved: Step S1: obtaining user real-time payment information; generating multi-dimensional payment information based on the user real-time payment information; performing time series fitting on the multi-dimensional payment information to generate a payment behavior sequence; Step S2: obtaining the local dependency of the payment sequence based on the payment behavior sequence; The implicit risk characteristics of payment behavior sequence are analyzed by using the local dependency of payment sequence to obtain implicit payment risk behavior characteristic data; Step S3: obtaining multiple payment behavior nodes based on the payment behavior sequence; using implicit payment risk behavior feature data to dynamically connect multiple payment behavior nodes to form a payment behavior node blockchain network; Step S4: Perform risk probability distribution analysis on the payment behavior node blockchain network to obtain node risk probability distribution data; Conduct risk propagation dynamic evolution analysis based on node risk probability distribution data to generate risk propagation evolution data; Step S5: Quantitatively predict the risk trends one by one based on the risk propagation evolution data to obtain the risk trend prediction value of the payment node; Mark high-risk payment nodes based on payment node risk trend prediction values; Reconstruct the global potential association path of high-risk payment nodes to obtain the payment risk deduction chain; Step S6: Conduct risk monitoring decision mining on the payment behavior node blockchain network according to the payment risk deduction chain to build an intelligent payment risk monitoring engine; Execute user payment operations based on the intelligent payment risk monitoring engine; the specific steps of step S3 are: Step S31: dividing the payment behavior sequence into time stamp nodes, thereby obtaining a plurality of payment behavior nodes; Step S32: using the implicit payment risk behavior feature data to perform implicit feature virtual projection on multiple payment behavior nodes to obtain implicit feature behavior nodes; Step S33: Calculating feature similarity of implicit feature behavior nodes to obtain feature similarity between nodes; Step S34: performing node link association strength analysis on implicit feature behavior nodes based on feature similarity between nodes to obtain node association strength data; Step S35: Based on the node association strength data, the implicit characteristic behavior nodes are dynamically connected to the topological network to build a behavior node blockchain network; the specific steps of step S35 are: Conduct contract analysis on implicit feature behavior nodes based on node association strength data to generate smart contract logic; According to the smart contract logic, the implicit feature behavior node is edited and processed by the smart contract, and the payment behavior smart contract has been obtained; Through the payment behavior smart contract, the implicit feature behavior nodes are dynamically reconstructed into a topological network architecture to build a payment behavior node blockchain network.

2. The blockchain-based third-party payment security authentication method according to claim 1 is characterized in that: The specific steps of step S1 are: Step S11: Obtaining user real-time payment information; Step S12: Identify the payment attribute of the user's real-time payment information to obtain payment attribute data; Step S13: dividing the user's real-time payment information into multiple dimensions based on the payment attribute data to generate multi-dimensional payment information; Step S14: mining payment behavior characteristics on the multi-dimensional payment information to obtain payment behavior characteristic data; Step S15: Perform time series fitting on the payment behavior characteristic data to generate a payment behavior sequence.

3. The blockchain-based third-party payment security authentication method according to claim 1 is characterized in that: The specific steps of step S2 are: Step S21: performing time-domain convolution filtering on the payment behavior sequence to obtain the local dependency of the payment sequence; Step S22: recursively transfer features of the local dependencies of the payment sequence to extract global trend rules; Step S23: Utilize the global trend law to perform implicit risk feature analysis on the payment behavior sequence to obtain implicit payment risk behavior feature data.

4. The blockchain-based third-party payment security authentication method according to claim 1 is characterized in that: The specific steps of step S4 are: Step S41: Based on the payment behavior feature data, quantitative risk assessment is performed on each node of the distributed behavior node network to obtain payment node risk data; Step S42: performing risk probability distribution analysis on the payment node risk data to obtain node risk probability distribution data; Step S43: performing risk propagation path response identification on the distributed behavior node network according to the node risk probability distribution data, thereby generating a node risk propagation path; Step S44: Perform risk propagation dynamic evolution analysis on the node risk propagation path to generate risk evolution data.

5. The blockchain-based third-party payment security authentication method according to claim 1 is characterized in that: The specific steps of step S5 are: Step S51: Quantitatively predict the risk trends of the payment behavior node blockchain network one by one based on the risk propagation evolution data to obtain the payment node risk trend prediction value; Step S52: comparing the payment node risk trend prediction value based on the preset payment behavior risk prediction threshold, and marking the payment node as a high-risk payment node when the payment node risk trend prediction value is greater than or equal to the preset payment behavior risk prediction threshold; Step S53: tracing the payment link of high-risk payment nodes to eliminate risky payment links; Step S54: reconstruct the global potential association path of the risk payment link to obtain the payment risk deduction chain.

6. The blockchain-based third-party payment security authentication method according to claim 1 is characterized in that: The specific steps of step S6 are: Step S61: Identify abnormal risk factors in the payment risk deduction chain and generate payment risk factors; Step S62: Perform risk monitoring decision mining on the distributed behavior node network according to payment risk factors to build an intelligent payment risk monitoring engine; Step S63: Perform payment security authentication processing on the user's real-time payment information based on the intelligent payment risk monitoring engine, thereby obtaining a payment security authentication result; The payment security authentication result specifically includes a normal payment result and a risky payment result; Step S64: When the payment security authentication result is a normal payment result, the user is allowed to make a payment transaction, and the user payment security authentication operation is completed; Step S65: When the payment security authentication result is a risky payment result, a payment risk warning signal is generated to prohibit the user from making payments.

7. A third-party payment security authentication system based on blockchain, characterized in that: Used to execute the blockchain-based third-party payment security authentication method as claimed in claim 1, comprising: The behavior sequence module is used to obtain the user's real-time payment information; generate multi-dimensional payment information based on the user's real-time payment information; perform time series fitting on the multi-dimensional payment information to generate a payment behavior sequence; The risk behavior feature module is used to obtain the local dependency of the payment sequence based on the payment behavior sequence; the payment behavior sequence is subjected to implicit risk feature analysis using the local dependency of the payment sequence to obtain implicit payment risk behavior feature data; A topological network module is used to obtain multiple payment behavior nodes based on the payment behavior sequence; use implicit payment risk behavior feature data to dynamically connect multiple payment behavior nodes to form a payment behavior node blockchain network; The risk dynamic evolution module is used to perform risk probability distribution analysis on the payment behavior node blockchain network to obtain node risk probability distribution data; perform risk propagation dynamic evolution analysis based on the node risk probability distribution data to generate risk propagation evolution data; The risk trend prediction module is used to quantitatively predict risk trends one by one based on risk propagation evolution data to obtain the risk trend prediction value of the payment node; mark high-risk payment nodes based on the risk trend prediction value of the payment node; reconstruct the global potential association path of the high-risk payment node to obtain the payment risk deduction chain; The risk monitoring module is used to conduct risk monitoring decision mining on the distributed behavior node network according to the payment risk deduction chain, build an intelligent payment risk monitoring engine, and execute user payment operations based on the intelligent payment risk monitoring engine.

Images