Risk control method and system for real-time transfer transaction

Abstract

The invention discloses a risk control method and system for real-time transfer transaction, and relates to the technical field of finance. The method comprises the steps of collecting transaction related information, and establishing an initial transaction intention fingerprint; obtaining transfer execution behavior information, updating the initial transaction intention fingerprint, and generating a transaction intention evolution track; performing consistency analysis on the transfer execution behavior information and the transaction intention evolution trajectory to obtain a transaction intention deviation result; generating a corresponding risk energy variation according to a preset risk mapping rule, and constructing a risk energy change track; judging the risk state of the transfer transaction request, and determining a risk control level; setting a risk control permission condition, judging whether the risk control permission condition corresponding to the risk control level is met or not, and executing a risk control action matched with the risk control level according to a judgment result. According to the method and the device, the accuracy of transfer transaction risk identification and the performability of risk control are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of financial technology, and in particular to a risk control method and system for real-time transfer transactions. Background Technology

[0002] With the rapid development of mobile payment, internet banking, and third-party payment platforms, real-time fund transfers have become a common form of transaction in financial services. Real-time fund transfers are typically characterized by short time intervals between transaction initiation and fund arrival, high concurrent transaction volumes, and complex transaction chains. Once abnormal transaction behavior occurs, it can easily lead to financial losses in a short period of time. Therefore, this places high demands on the risk control of real-time fund transfers.

[0003] In related technologies, existing real-time transfer transaction risk control methods are mostly concentrated at the transaction initiation stage. They mainly rely on static rules or risk scoring models to conduct a one-time assessment of information such as transfer amount, account attributes, and historical behavior. They lack the ability to continuously monitor and analyze dynamic behavioral changes during the execution of transfer transactions, making it difficult to identify risks that gradually emerge during the transaction execution stage in a timely manner. In addition, most risk control methods use discrete threshold triggering methods, which are difficult to characterize the cumulative and volatile nature of risk evolution over time. At the level of risk control action execution, they lack comprehensive constraints on the transaction subject's authority, transaction scenario, and system operating status, which can easily lead to problems such as rigid risk control strategies, insufficient interpretability, or excessive intervention in normal transactions. There is room for improvement. Summary of the Invention

[0004] The purpose of this invention is to provide a risk control method and system for real-time transfer transactions to solve the problems mentioned in the background art.

[0005] Firstly, this application provides a risk control method for real-time transfer transactions, which adopts the following technical solution: When a transfer transaction request is received, transaction-related information corresponding to the transfer transaction request is collected, and an initial transaction intent fingerprint of the transfer transaction request is established; During the execution of a transfer transaction, transfer execution behavior information is obtained, the initial transaction intent fingerprint is updated, and a corresponding transaction intent evolution trajectory is generated. By performing a consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory, the corresponding transaction intent deviation results are obtained; Based on the deviation of the transaction intention, the corresponding risk energy change is generated according to the preset risk mapping rules, and the corresponding risk energy change trajectory is constructed. Based on the risk energy change trajectory, the risk status of the transfer transaction request is determined, and the risk control level corresponding to the transfer transaction request is determined. Set risk control permission conditions, further determine whether the risk control permission conditions corresponding to the risk control level are met, and execute risk control actions matching the risk control level based on the determination result.

[0006] Preferably, the step of collecting transaction-related information corresponding to the transfer transaction request and establishing the initial transaction intent fingerprint of the transfer transaction request when accepting the transfer transaction request specifically includes: When accepting a transfer transaction request, the transaction-related information corresponding to the transfer transaction request is collected. The transaction-related information includes transfer amount information, recipient information, transfer initiation time, and transfer operation environment information. The transaction-related information is structured and parsed, and different types of transaction-related information are mapped to a preset set of information dimensions to form a standard set of transaction features; Based on the standard transaction feature set, intent feature parameters that characterize the expected behavior of the transfer transaction are extracted. According to the preset intent feature combination rules, the intent feature parameters are structurally combined to generate the initial transaction intent fingerprint corresponding to the prime number transfer transaction request.

[0007] Preferably, the steps of obtaining transfer execution behavior information, updating the initial transaction intent fingerprint, and generating the corresponding transaction intent evolution trajectory during the execution of a transfer transaction are as follows: During the execution of a transfer transaction, transfer execution behavior information is obtained, including transfer operation sequence change information, transfer confirmation behavior information, transfer execution delay information, and transfer amount adjustment information. The transfer execution behavior information is processed into events, converting it into a sequence of execution behavior events ordered by time. Based on the sequence of execution behavior events, behavioral evolution feature parameters reflecting the changing characteristics of transfer execution behavior are extracted. The behavioral evolution feature parameters include behavioral occurrence sequence features, behavioral type change features, and behavioral rhythm change features. The behavioral evolution feature parameters are associated and mapped with the initial transaction intent fingerprint. According to the preset intent update rule, the corresponding intent feature parameters in the initial transaction intent fingerprint are adjusted to generate the corresponding intermediate transaction intent state. The intermediate transaction intent states are continuously recorded according to the chronological order of the transfer execution, forming a transaction intent evolution trajectory that represents the changes in transaction intent as the transfer execution process progresses.

[0008] Preferably, the step of performing consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation result specifically includes: Based on the sequence of execution events corresponding to the transfer execution behavior information, a sequence of transaction intent state features corresponding to the sequence of execution events in the time dimension is extracted from the transaction intent evolution trajectory; Based on the sequence of execution behavior events and the sequence of transaction intent states, the degree of matching between the transfer execution behavior events and the corresponding transaction intent states is calculated according to preset consistency analysis rules, thereby obtaining the corresponding intent state. Figure 1 Consistency measurement results; Regarding the meaning stated Figure 1 Consistency measurement results are used for time-series trend analysis to identify changes in consistency between transfer execution behavior and the evolution trajectory of transaction intent, and to generate corresponding intent... Figure 1 Characteristic parameters of consistency change; Based on the above intention Figure 1 The consistency change characteristic parameters are used to determine whether the current transfer execution behavior deviates from the intention evolution direction corresponding to the transaction intention evolution trajectory, and generate the corresponding transaction intention deviation result.

[0009] Preferably, based on the stated intention Figure 1 The steps of determining whether the current transfer execution behavior deviates from the intent evolution direction corresponding to the transaction intent evolution trajectory, based on the consistency change characteristic parameters, and generating the corresponding transaction intent deviation result, are as follows: Based on the above intention Figure 1 Consistency change characteristic parameters are used to construct an intention deviation assessment parameter set, which includes consistency change direction parameters, consistency change magnitude parameters, and consistency change rate parameters. Obtain the intention evolution direction feature parameters corresponding to the transaction intention evolution trajectory. The intention evolution direction feature parameters are used to characterize the subjective evolution direction of the transaction intention state as time changes. The consistency change direction parameter is compared with the intention evolution direction feature parameter to determine whether the consistency change direction of the transfer execution behavior is consistent with the intention evolution direction of the transaction intention evolution trajectory. If the direction of consistency change is inconsistent with the direction of intent evolution, the deviation strength of the transfer execution behavior relative to the trajectory of transaction intent evolution is comprehensively evaluated by combining the consistency change magnitude parameter and the consistency change rate parameter, and the deviation strength evaluation result is obtained to generate the corresponding transaction intent deviation result.

[0010] Preferably, the step of generating a corresponding risk energy change based on the deviation result of the transaction intention according to a preset risk mapping rule, and constructing a corresponding risk energy change trajectory, specifically includes: Based on the deviation results of the transaction intent, analyze the deviation type parameters and deviation intensity parameters of the deviation status of the transfer execution behavior; Based on the deviation type parameter and deviation intensity parameter, according to the preset risk mapping rule, the deviation result of the transaction intention is mapped to the corresponding risk energy change, which is used to characterize the incremental or decremental change of the current transfer transaction risk status. Based on the change in risk energy, the current risk energy status corresponding to the transfer transaction is updated, and the corresponding risk energy status value is generated. According to the time sequence of the transfer transaction execution process, the risk energy state value generated at each moment is continuously recorded to construct a risk energy change trajectory that represents the dynamic change of risk state as the transfer execution process progresses.

[0011] Preferably, the step of determining the risk status of the transfer transaction request and identifying the risk control level corresponding to the transfer transaction request based on the risk energy change trajectory specifically includes: Based on the risk energy change trajectory, risk trajectory feature parameters that characterize the evolution of risk status are extracted. The risk trajectory feature parameters include risk energy change trend parameters, risk energy fluctuation amplitude parameters, and risk energy accumulation degree parameters. The risk trajectory feature parameters are matched with preset risk status determination rules to determine the current risk status type of the transfer transaction request; Based on the risk status type, obtain the risk status identifier parameter corresponding to the risk status type; Based on the risk status identifier parameter, the risk control level corresponding to the transfer transaction request is determined according to the preset risk control level mapping rule. The risk control level is used to indicate the level of risk control strategy implemented for the transfer transaction request.

[0012] Preferably, the steps include setting risk control permission conditions, further determining whether the risk control permission conditions corresponding to the risk control level are met, and executing risk control actions matching the risk control level based on the determination result. Specifically, this involves: Based on the risk control level, obtain the risk control permission condition set corresponding to the risk control level. The risk control permission condition set includes transaction entity permission conditions, transaction scenario permission conditions, and system operation status permission conditions. Based on the risk control permission condition set, the current transaction environment status, transaction subject status and system status of the transfer transaction request are matched and determined to generate the corresponding risk control permission determination result. When the risk control authority determination result meets the risk control authority conditions corresponding to the risk control level, a risk control action matching the risk control level is executed. When the risk control authority determination result does not meet the risk control authority conditions corresponding to the risk control level, the execution method of the risk control action is adjusted or the execution of the risk control action is delayed. The risk control actions include one or more of the following: delayed processing of transfer transactions, secondary verification of transfer transactions, restriction of transfer transaction execution, and blocking of transfer transactions.

[0013] Secondly, the risk control system for real-time transfer transactions provided in this application adopts the following technical solution: A risk control system for real-time transfer transactions includes: The intent fingerprint construction module, upon receiving a transfer transaction request, collects transaction-related information corresponding to the transfer transaction request and establishes an initial transaction intent fingerprint for the transfer transaction request; The intent trajectory generation module acquires transfer execution behavior information during the execution of a transfer transaction, updates the initial transaction intent fingerprint, and generates the corresponding transaction intent evolution trajectory. The intent deviation analysis module performs consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation results. The risk energy change module generates the corresponding risk energy change amount according to the deviation result of the transaction intention and constructs the corresponding risk energy change trajectory according to the preset risk mapping rules. The control level determination module determines the risk status of the transfer transaction request based on the risk energy change trajectory, and determines the risk control level corresponding to the transfer transaction request. The control permission determination module sets risk control permission conditions, further determines whether the risk control permission conditions corresponding to the risk control level are met, and executes risk control actions matching the risk control level based on the determination result.

[0014] In summary, this application includes at least one of the following beneficial technical effects: 1. By constructing an initial transaction intent fingerprint during the transaction request acceptance phase and dynamically updating the transaction intent in conjunction with transfer execution behavior information during the transaction execution process, a transaction intent evolution trajectory is formed. This achieves continuous characterization of the expected intent of a transfer transaction and its changes during execution, effectively solving the problem that existing risk control methods rely solely on static judgment at the transaction initiation stage, and significantly improving the ability to identify hidden risks during transfer execution. By performing consistency analysis between transfer execution behavior information and the transaction intent evolution trajectory, deviations between transaction execution behavior and transaction intent are identified, enabling the system to... Figure 1This approach identifies abnormal transfer behavior from a consistency perspective, thereby improving the accuracy and interpretability of risk assessment. Furthermore, it introduces risk energy change and risk energy change trajectory, transforming discrete risk assessment results into a continuously accumulating and traceable risk quantification process. This allows the risk status to dynamically evolve during transaction execution, avoiding misjudgments or excessive interventions caused by traditional threshold triggering methods. Based on this, risk status is determined and risk control levels are established according to the risk energy change trajectory, providing a clear hierarchical output for risk assessment results, facilitating their correspondence with specific risk control strategies. Simultaneously, risk control permission conditions are introduced during the risk control action execution phase, comprehensively constraining the transaction entity's permissions, transaction scenario conditions, and system operating status. This makes the execution of risk control actions more reasonable and controllable, effectively reducing the impact on normal transfer transactions. While ensuring the security of real-time transfer transactions, it also considers transaction efficiency and system stability, making it particularly suitable for high-concurrency, high-real-time transfer transaction scenarios.

[0015] 2. By analyzing the deviation type and intensity parameters of the transfer execution behavior based on the deviation results of transaction intentions, and mapping them into risk energy changes according to preset risk mapping rules, the originally qualitative or discrete intention deviation judgment results are transformed into risk increments or decrements with clear physical quantitative meanings, thereby achieving a refined quantitative expression of the risk status of transfer transactions. By continuously updating the current risk energy status based on the risk energy change and constructing a risk energy change trajectory in chronological order during the transfer transaction execution process, risk assessment is no longer limited to static judgments at a single moment, but can truly reflect the dynamic evolution characteristics of risk as it gradually accumulates, fluctuates, or is mitigated during transaction execution. This effectively avoids the problems of unstable risk judgments, oversensitivity to short-term anomalies, or delayed responses to continuous anomalies caused by relying solely on thresholds or instantaneous scores in existing technologies. Simultaneously, the risk energy change trajectory provides a temporally contextual basis for subsequent risk status determination and risk control level determination, enabling risk control strategies to make decisions based on the overall risk evolution trend rather than isolated events, thereby significantly improving the accuracy, continuity, and interpretability of real-time transfer transaction risk control.

[0016] 3. By introducing a set of risk control permission conditions after the risk control level is determined, and linking the execution of risk control actions with the transaction subject's permissions, transaction scenario conditions, and system operating status, an executability verification mechanism is added on top of the risk level assessment. This prevents risk control actions from being triggered directly when the execution conditions are not met. When the risk control permission assessment result meets the permission conditions of the corresponding risk control level, the system can promptly execute risk control actions that match the risk level, thereby effectively curbing high-risk transfer behavior and improving fund security. When the permission conditions are not met, the execution method of the risk control action can be adjusted or delayed, allowing the system to ensure security while maintaining transaction continuity and system stability, avoiding false blocking or user experience degradation caused by forced control. Furthermore, by pre-setting multiple risk control action types and supporting selection or combination execution according to risk control level, the risk control strategy has greater flexibility and adaptability. This effectively solves the problems of rigid execution, lack of permission constraints, and insufficient adaptability to complex transaction scenarios in existing technologies, making the risk control process of real-time transfer transactions more refined, controllable, and with good engineering feasibility. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the specific steps of an embodiment of a risk control method for real-time transfer transactions according to the present invention.

[0018] Figure 2 This is a schematic diagram of the module connections of an embodiment of the risk control system for real-time transfer transactions according to the present invention. Detailed Implementation

[0019] The following examples and... Figures 1-2 The present invention will be described in further detail, but the embodiments of the present invention are not limited thereto.

[0020] This invention discloses a risk control method for real-time transfer transactions, specifically including the following steps: Step S1: When accepting a transfer transaction request, collect transaction-related information corresponding to the transfer transaction request and establish the initial transaction intent fingerprint of the transfer transaction request; Step S2: During the execution of the transfer transaction, obtain the transfer execution behavior information, update the initial transaction intent fingerprint, and generate the corresponding transaction intent evolution trajectory. Step S3: Perform consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation results; Step S4: Based on the deviation result of the transaction intention, generate the corresponding risk energy change amount according to the preset risk mapping rules, and construct the corresponding risk energy change trajectory; Step S5: Based on the risk energy change trajectory, determine the risk status of the transfer transaction request and determine the risk control level corresponding to the transfer transaction request; Step S6: Set risk control permission conditions, further determine whether the risk control permission conditions corresponding to the risk control level are met, and execute risk control actions matching the risk control level according to the determination result.

[0021] In practical applications, upon receiving a transfer request, the system collects relevant transaction information to establish an initial transaction intent fingerprint. This allows for a structured representation of the initiator's transfer purpose, behavioral characteristics, and transaction context before the formal execution of the transfer. This provides a unified benchmark for subsequent risk analysis, avoiding the information insufficiency issues that arise from relying solely on single transaction parameters for risk assessment. The system can comprehensively characterize the transfer request from the perspective of transaction intent, laying the foundation for dynamic change analysis during subsequent transaction execution. During transaction execution, the system continuously acquires transfer execution behavior information and dynamically updates the initial transaction intent fingerprint, generating a corresponding transaction intent evolution trajectory. This transforms the originally static transaction intent into a dynamic evolutionary process that changes over time. The system can accurately reflect changes in intent caused by user operations, environmental changes, or interactive behaviors during transaction execution, avoiding fixing the transaction intent to a single moment's judgment. This enhances the system's ability to perceive the gradual evolution of abnormal behavior in risk identification. By analyzing the consistency between transfer execution information and the evolution trajectory of transaction intent, corresponding transaction intent deviation results are generated. This allows the system to determine whether the actual execution of a transfer transaction is consistent with its intent evolution process. It effectively identifies situations such as intent reversal, behavioral anomalies, or execution path deviations that occur during transaction execution, thus avoiding risk assessment based solely on outcome-based indicators or instantaneous anomalies. This improves the ability to identify hidden risks and progressively anomalous behaviors. Based on the transaction intent deviation results, a risk energy change is generated according to preset risk mapping rules, and a risk energy change trajectory is constructed. By introducing the risk energy change and its trajectory, the system can describe the accumulation, amplification, or decline trend of risk during transaction execution, avoiding simply treating risk as a binary or single-point score. This makes the risk assessment results closer to the objective laws of the gradual manifestation and dynamic changes of risk in real transaction scenarios. The system assesses the risk status of transfer transaction requests based on the trajectory of risk energy changes and determines the corresponding risk control level. By analyzing the trend, magnitude, and cumulative characteristics of the risk evolution trajectory, it differentiates between different risk stages. This allows the system to adopt differentiated control strategies based on the severity of the risk status, avoiding over-interception or risk omission due to overly simplistic risk assessment. By setting risk control permission conditions and further determining whether these conditions are met, the system executes matching risk control actions based on the assessment results. This introduces a permission constraint mechanism between risk control decisions and actual execution, ensuring that the execution of risk control actions is not only based on risk level assessment but also on comprehensive decisions considering factors such as the transaction entity's permissions, transaction scenario limitations, and system operating status. This improves the executability and stability of risk control strategies in real-world system environments, preventing disruptions to normal business operations due to inappropriate mandatory control measures.

[0022] The steps of collecting transaction-related information corresponding to the transfer request and establishing the initial transaction intent fingerprint of the transfer request upon receiving the request are as follows: Step S11: When accepting a transfer transaction request, collect transaction-related information corresponding to the transfer transaction request. The transaction-related information includes transfer amount information, recipient information, transfer initiation time, and transfer operation environment information. Step S12: Perform structured parsing on the transaction-related information, mapping different types of transaction-related information to preset information dimension sets to form a standard transaction feature set; Step S13: Based on the standard transaction feature set, extract the intent feature parameters used to characterize the expected behavior of the transfer transaction, and combine the intent feature parameters in a structured manner according to the preset intent feature combination rules to generate the initial transaction intent fingerprint corresponding to the prime number transfer transaction request.

[0023] In practical applications, when receiving transfer requests, the system collects transaction-related information such as transfer amount, recipient information, transfer initiation time, and transfer operation environment information. This allows the system to not only focus on a single transfer amount or account information but also simultaneously acquire the time characteristics and operational environment characteristics of the transaction. This provides a complete data foundation for subsequent analysis of the rationality and intent characteristics of transaction behavior, avoiding the problem of incomplete characterization of transaction intent due to insufficient information collection dimensions. By structurally parsing the collected transaction-related information of different types and mapping them to preset information dimension sets, a standard transaction feature set is formed. This eliminates the differences in format, scale, and semantics of the original transaction data, enabling various types of transaction information to be expressed and processed within a unified feature space. This helps transform originally scattered and heterogeneous transaction data into comparable and combinable standardized features, providing a unified input interface for subsequent intent feature extraction and rule calculation, thereby improving the stability and consistency of the transaction intent modeling process. Based on the standard transaction feature set, intention feature parameters are extracted to characterize the expected behavior of transfer transactions. The intention feature parameters are then structured and combined according to the preset intention feature combination rules to generate the initial transaction intention fingerprint corresponding to the transfer transaction request. Through the form of the initial transaction intention fingerprint, the transaction intention can be improved from a single feature judgment to a multi-dimensional feature collaborative characterization, providing a clear starting state for the intention evolution analysis in the subsequent transaction execution process. This allows risk control to revolve around the core logic of whether the intention has changed.

[0024] The steps involved in acquiring transfer execution behavior information, updating the initial transaction intent fingerprint, and generating the corresponding transaction intent evolution trajectory during the execution of a transfer transaction are as follows: Step S21: During the execution of the transfer transaction, obtain transfer execution behavior information, which includes transfer operation sequence change information, transfer confirmation behavior information, transfer execution delay information, and transfer amount adjustment information; Step S22: Process the transfer execution behavior information into events, converting the transfer execution behavior information into a sequence of execution behavior events ordered by time. Step S23: Based on the execution behavior event sequence, extract behavioral evolution feature parameters that reflect the changing characteristics of transfer execution behavior. The behavioral evolution feature parameters include behavioral occurrence sequence features, behavioral type change features, and behavioral rhythm change features. The behavioral evolution feature parameters are associated and mapped with the initial transaction intent fingerprint. According to the preset intent update rule, the corresponding intent feature parameters in the initial transaction intent fingerprint are adjusted to generate the corresponding intermediate transaction intent state. Step S24: According to the time sequence of the transfer execution behavior, the intermediate transaction intention state is continuously recorded to form a transaction intention evolution trajectory that represents the change of transaction intention as the transfer execution process progresses.

[0025] In practical applications, during the execution of transfer transactions, by acquiring transfer execution behavior information, including information on changes in the transfer operation sequence, transfer confirmation behavior, transfer execution delay, and transfer amount adjustment, the system can perceive potential operational adjustments, confirmation anomalies, or changes in execution rhythm that may occur from the initiation to completion of the transaction. This avoids relying solely on information from the transaction initiation stage for static judgment of transaction behavior, providing real and continuous data support for identifying potential anomalies or risks during execution. By processing transfer execution behavior information into event-driven sequences of execution behavior events ordered chronologically, the originally scattered and continuous operational behaviors are transformed into a structured set of time events. This allows the system to manage and analyze different types of execution behaviors in a unified event format, clearly depicting the order and interrelationships of various behaviors over time, providing standardized input for subsequent extraction of behavioral evolution features and time-series analysis. Based on the sequence of execution events, behavioral evolution feature parameters reflecting changes in transfer execution behavior are extracted, including behavioral occurrence timing features, behavioral type change features, and behavioral rhythm change features. From the event level, higher-level features that characterize behavioral evolution trends and change patterns are further abstracted. The system can identify whether there are abnormal delays, frequent changes, or rhythmic disturbances in execution behavior, thus providing a quantitative basis for judging whether transaction behavior deviates from its expected execution pattern. The behavioral evolution feature parameters are mapped to the initial transaction intent fingerprint, and the corresponding intent feature parameters in the initial transaction intent fingerprint are adjusted according to a preset intent update rule to generate intermediate transaction intent states. Behavioral changes during the execution phase are fed back into the transaction intent model, making the transaction intent no longer a fixed, static description, but dynamically correctable as execution behavior changes. This truly reflects the gradual evolution or deviation of intent during transaction execution, providing a dynamic intent basis for subsequent consistency analysis and risk assessment. The system continuously records the intermediate transaction intent states according to the chronological order of the transfer execution, forming an evolutionary trajectory of transaction intent that represents the changes in transaction intent as the transfer execution process progresses. This allows the system to analyze the direction and degree of change in transaction intent from both an overall and trend perspective, rather than focusing solely on the intent state at a single moment. This provides a basis for subsequent analysis of the intent. Figure 1 Consistency analysis, deviation determination, and risk quantification provide reliable historical evolution data.

[0026] The steps for performing consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation results are as follows: Step S31: Based on the sequence of execution events corresponding to the transfer execution behavior information, extract the transaction intention state feature sequence corresponding to the sequence of execution events in the time dimension from the transaction intention evolution trajectory; Step S32: Based on the sequence of execution behavior events and the sequence of transaction intent states, calculate the degree of matching between the transfer execution behavior events and the corresponding transaction intent states according to preset consistency analysis rules, and obtain the corresponding intent state. Figure 1 Consistency measurement results; Step S33, regarding the meaning Figure 1 Consistency measurement results are used for time-series trend analysis to identify changes in consistency between transfer execution behavior and the evolution trajectory of transaction intent, and to generate corresponding intent... Figure 1 Characteristic parameters of consistency change; Step S34, based on the stated intention Figure 1 The consistency change characteristic parameters are used to determine whether the current transfer execution behavior deviates from the intention evolution direction corresponding to the transaction intention evolution trajectory, and generate the corresponding transaction intention deviation result.

[0027] In practical applications, by using the sequence of execution behavior events as a time benchmark, the system extracts the corresponding sequence of transaction intention state features at the same time dimension from the trajectory of transaction intention evolution. This achieves a one-to-one mapping between behavior events and intention states, thus avoiding the information loss problem caused by coarse-grained comparisons at the overall level. The system slices the originally continuously evolving transaction intention trajectory into discrete intention states corresponding to specific execution behaviors, enabling consistency analysis to focus on specific behavioral nodes rather than remaining at the macro-statistical level. Through preset consistency analysis rules, the system calculates the degree of conformity between execution behavior events and corresponding transaction intention state features across multiple dimensions such as behavior type, occurrence sequence, and operational intensity, outputting a clear intention... Figure 1 Consistency measurement results enable the system to perform horizontal comparisons of intent matching for different transfer behaviors at different time points, providing standardized input for subsequent trend analysis and risk mapping, and avoiding subjective judgments based on human experience or static rules. By analyzing intent across continuous time dimensions... Figure 1 Trend analysis is performed on consistency measurement results to identify whether consistency remains stable, gradually declines, or undergoes abrupt changes, thereby generating a meaning index that reflects the direction and magnitude of consistency changes. Figure 1 Consistency change characteristic parameters upgrade risk identification from single-point anomaly judgment to process evolution judgment, effectively distinguishing between occasional minor inconsistencies and persistent, cumulative intention deviations. This provides stronger time sensitivity and robustness for identifying complex risk scenarios such as potential fraud, abnormal manipulation, or intentional hijacking. Through analysis of intention... Figure 1The trend, intensity, and duration of deviation reflected by the consistency change characteristic parameters determine whether the transfer execution behavior is still within the range of reasonable intention evolution or has entered a deviation state. Based on this, a clear transaction intention deviation result is generated, providing a direct and clear input basis for subsequent risk energy mapping and risk control level determination. This makes risk assessment no longer based solely on static characteristics or outcome indicators, but on a deep understanding of the dynamic deviation degree of transaction intention, fundamentally improving the foresight and accuracy of real-time transfer transaction risk identification.

[0028] Based on the above intention Figure 1 The steps of determining whether the current transfer execution behavior deviates from the intent evolution direction corresponding to the transaction intent evolution trajectory, based on the consistency change characteristic parameters, and generating the corresponding transaction intent deviation result, are as follows: Step S341, based on the stated intention Figure 1 Consistency change characteristic parameters are used to construct an intention deviation assessment parameter set, which includes consistency change direction parameters, consistency change magnitude parameters, and consistency change rate parameters. Step S342: Obtain the intention evolution direction feature parameters corresponding to the transaction intention evolution trajectory. The intention evolution direction feature parameters are used to characterize the subjective evolution direction of the transaction intention state as time changes. Step S343: Compare the consistency change direction parameter with the intention evolution direction feature parameter to determine whether the consistency change direction of the transfer execution behavior is consistent with the intention evolution direction of the transaction intention evolution trajectory. Step S344: If the direction of consistency change is inconsistent with the direction of intention evolution, the deviation intensity of the transfer execution behavior relative to the transaction intention evolution trajectory is comprehensively evaluated by combining the consistency change magnitude parameter and the consistency change rate parameter, and the deviation intensity evaluation result is obtained, and the corresponding transaction intention deviation result is generated.

[0029] In practical applications, by introducing parameters for the direction of consistent change, the magnitude of consistent change, and the rate of consistent change, the intention can be analyzed from three dimensions: directionality, degree, and dynamism. Figure 1The system characterizes consistency changes, transforming previously scattered consistency information into an evaluation parameter system with clear semantics. This provides clear logical input boundaries for subsequent deviation judgments, avoiding coarse judgments based solely on a single indicator or empirical threshold, thereby improving the systematic nature and stability of transaction intent deviation identification. By analyzing the evolution trajectory of transaction intent, directional feature parameters are formed, making the expected evolution path of transaction intent explicit and parameterized. This avoids using static initial intent as a reference for deviation judgments, accurately adapting to reasonable intent changes caused by normal operations during the execution of transfer transactions, improving the adaptability and accuracy of deviation judgments. By comparing the consistency change direction parameters with the intent evolution direction feature parameters, the system can quickly identify whether transfer execution behavior shows changes opposite to or deviating from the expected intent evolution trend. This allows the system to prioritize capturing reverse evolution behaviors with inherent risk significance, providing trigger conditions for subsequent deviation intensity assessments and avoiding misjudgments of normal behaviors that only exhibit amplitude fluctuations but maintain a consistent direction. By simultaneously considering the magnitude of the deviation and its speed of occurrence, it is possible to distinguish between different risk forms such as slight, gradual deviation and severe, sudden deviation. This allows the results of deviations from trading intentions to not only reflect whether a deviation has occurred, but also to what extent. This provides a hierarchical input basis for subsequent calculations of risk energy changes and determination of risk control levels, thereby enhancing the refinement and foresight of the entire risk control plan.

[0030] The steps of generating corresponding risk energy changes and constructing corresponding risk energy change trajectories based on the deviation results of the stated trading intentions, according to preset risk mapping rules, are as follows: Step S41: Based on the deviation result of the transaction intention, analyze the deviation type parameter and deviation intensity parameter of the deviation status of the transfer execution behavior; Step S42: Based on the deviation type parameter and deviation intensity parameter, according to the preset risk mapping rule, the deviation result of the transaction intention is mapped to the corresponding risk energy change, which is used to characterize the incremental or decremental change of the current transfer transaction risk status. Step S43: Update the current risk energy status corresponding to the transfer transaction based on the risk energy change, and generate the corresponding risk energy status value; Step S44: According to the time sequence of the transfer transaction execution process, continuously record the risk energy state value generated at each moment to construct a risk energy change trajectory that represents the dynamic change of risk state as the transfer execution process progresses.

[0031] In practical applications, by distinguishing different types of deviation (such as directional deviation, rhythmic deviation, or amplitude deviation) and quantifying their intensity, the determination of whether a deviation exists is transformed into a structured risk description with distinctions in category and intensity. This avoids generalizing deviation behaviors of different risk natures, thus laying the foundation for building a differentiated and adjustable risk energy change mechanism. By mapping different types and intensities of intentional deviations to positive or negative risk energy changes, the system can measure risk changes with a unified energy dimension. Introducing risk energy change as an intermediate quantitative carrier transforms risk assessment from discrete rule-based judgment to a continuous and cumulative dynamic process, providing a calculable and scalable quantitative basis for subsequent risk state updates and trajectory construction. By making incremental or decremental adjustments based on existing risk energy states, the risk state can evolve in real time with changes in transfer execution behavior. This avoids isolated assessments of the risk state at each moment, instead reflecting the evolution trend of risk throughout the entire transfer execution process through a state accumulation mechanism, thereby enhancing the continuity, memory, and stability of risk judgment. By forming a time series of risk energy states, the system can intuitively reflect the process of risk accumulation, mitigation, or abrupt change, providing a temporal context for subsequent risk state determination, risk control level classification, and risk control action triggering. This avoids making decisions based solely on risk values ​​at a single moment, thereby significantly improving the foresight and rationality of risk control strategies.

[0032] The steps for determining the risk status of the transfer transaction request and identifying the corresponding risk control level based on the risk energy change trajectory are as follows: Step S51: Based on the risk energy change trajectory, extract risk trajectory feature parameters that characterize the evolution of risk status. The risk trajectory feature parameters include risk energy change trend parameters, risk energy fluctuation amplitude parameters, and risk energy accumulation degree parameters. Step S52: Match the risk trajectory feature parameters with the preset risk status determination rules to determine the current risk status type of the transfer transaction request; Step S53: Based on the risk status type, obtain the risk status identifier parameter corresponding to the risk status type; Step S54: Based on the risk status identifier parameter, determine the risk control level corresponding to the transfer transaction request according to the preset risk control level mapping rule. The risk control level is used to indicate the level of risk control strategy implemented for the transfer transaction request.

[0033] In practical applications, by introducing parameters for risk energy change trends, risk energy fluctuation amplitudes, and risk energy accumulation, the system can characterize the risk evolution process from both macro-trend and micro-fluctuation perspectives. This transforms the original risk energy time series into a high-level feature expression usable for rule-based judgment, providing a stable and robust basis for subsequent risk state type identification and avoiding misjudgments due to instantaneous fluctuations. Through a rule matching mechanism, the system can classify risk states into different levels or types based on the trend, volatility, and cumulative characteristics of risk evolution, such as stable states, cumulative rising states, or abrupt risk states. This prevents risk judgment from directly mapping to control actions, instead first forming a semantically meaningful risk state description, thereby improving the interpretability and scalability of risk control logic. Through risk state identifier parameters, the system can represent different risk state types in a unified data format, facilitating transmission and invocation between different system modules. This enables structured output of risk state judgment results, allowing subsequent risk control level mapping and permission judgment processes to be processed based on standardized identifiers, thereby reducing system coupling and improving the modularity of the overall risk control architecture. By mapping different risk status types to different risk control levels, the system can adopt differentiated control strategies based on the degree of risk, such as allowing passage, enhancing verification, or blocking. The continuously evolving risk status is ultimately transformed into a hierarchical decision result that can directly drive the execution of risk control strategies, thereby achieving an effective connection between risk assessment and risk control and improving the accuracy and response efficiency of real-time transfer transaction risk control.

[0034] The steps include setting risk control permission conditions, further determining whether the risk control permission conditions corresponding to the risk control level are met, and executing risk control actions matching the risk control level based on the determination result. Specifically, this involves: Step S61: Based on the risk control level, obtain the risk control permission condition set corresponding to the risk control level. The risk control permission condition set includes transaction entity permission conditions, transaction scenario permission conditions, and system operation status permission conditions. Step S62: Based on the risk control permission condition set, perform permission matching and determination on the current transaction environment status, transaction subject status and system status of the transfer transaction request, and generate the corresponding risk control permission determination result; Step S63: When the risk control authority determination result meets the risk control authority conditions corresponding to the risk control level, execute the risk control action that matches the risk control level. Step S64: When the risk control authority determination result does not meet the risk control authority conditions corresponding to the risk control level, adjust the execution method of the risk control action or delay the execution of the risk control action. Step S65, wherein the risk control action includes one or more of the following: delayed processing of transfer transactions, secondary verification of transfer transactions, execution of transfer transaction restrictions, and blocking of transfer transactions.

[0035] In practical applications, by dividing permission conditions into transaction subject permission conditions, transaction scenario permission conditions, and system operation status permission conditions, risk control decisions not only rely on the risk level itself but also incorporate comprehensive constraints based on the transaction subject attributes, the transaction environment, and the current system operational capabilities. This avoids forcibly implementing risk control actions when the conditions for execution are not met, thereby improving the rationality and feasibility of risk control strategies. By comparing multi-dimensional status information with permission conditions, the system can accurately determine whether the current transaction meets the preconditions for executing specific risk control actions. An additional permission verification mechanism is added before the risk control action is executed to prevent miscontrol or control failures caused by unmet environmental conditions or the subject lacking the corresponding permissions. By directly triggering control measures appropriate to the risk level, the system can intervene in high-risk transactions in a timely manner when permission conditions are met, effectively transforming risk assessment results into specific control behaviors and ensuring that risk control strategies achieve maximum effectiveness under compliant and executable conditions. By dynamically adjusting the control methods or execution timing, the system can adopt more prudent processing strategies when permission conditions are limited. This avoids impacting system stability or user experience by forcibly executing risk control actions without the necessary permissions, while also reserving space for subsequent control once the conditions are met. By providing multiple control action types, the system can perform combined or progressive control based on risk levels and permission constraints, making risk control strategies more flexible and configurable. This avoids over-intervention or under-control issues caused by a single control method, thereby achieving a balance between security and transaction efficiency.

[0036] A risk control system for real-time transfer transactions, comprising the application of a risk control method for real-time transfer transactions as described above, including: The intent fingerprint construction module, upon receiving a transfer transaction request, collects transaction-related information corresponding to the transfer transaction request and establishes an initial transaction intent fingerprint for the transfer transaction request; The intent trajectory generation module acquires transfer execution behavior information during the execution of a transfer transaction, updates the initial transaction intent fingerprint, and generates the corresponding transaction intent evolution trajectory. The intent deviation analysis module performs consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation results. The risk energy change module generates the corresponding risk energy change amount according to the deviation result of the transaction intention and constructs the corresponding risk energy change trajectory according to the preset risk mapping rules. The control level determination module determines the risk status of the transfer transaction request based on the risk energy change trajectory, and determines the risk control level corresponding to the transfer transaction request. The control permission determination module sets risk control permission conditions, further determines whether the risk control permission conditions corresponding to the risk control level are met, and executes risk control actions matching the risk control level based on the determination result.

[0037] In practical applications, an intent fingerprint construction module is used to collect transaction-related information and construct an initial transaction intent fingerprint during the transaction request acceptance stage, thus characterizing the expected intent of the transaction from the source. An intent trajectory generation module continuously acquires transaction execution behavior information and dynamically updates the initial transaction intent fingerprint during the transaction execution process, forming a transaction intent evolution trajectory reflecting changes in the transaction intent as the execution process progresses. An intent deviation analysis module performs consistency analysis between the transaction execution behavior information and the transaction intent evolution trajectory, identifying deviations between the transaction execution behavior and the expected transaction intent, thereby achieving accurate identification of abnormal execution behavior. A risk energy change module further analyzes the transaction... The intention deviation result is converted into a risk energy change quantity according to the preset risk mapping rules, and a risk energy change trajectory is constructed to achieve a continuous quantitative characterization of the risk status of the transfer transaction. Through the control level determination module, the current risk status of the transfer transaction is determined based on the risk energy change trajectory, and the corresponding risk control level is determined, so that the risk assessment result has an executable level output. Through the control permission determination module, risk control permission conditions are introduced on the basis of risk control level to determine the feasibility of risk control actions, and risk control actions that match the risk control level are executed, adjusted or delayed according to the determination result, so as to ensure transaction security while taking into account system stability and transaction experience.

[0038] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A risk control method for real-time transfer transactions, characterized in that, Includes the following steps: When a transfer transaction request is received, transaction-related information corresponding to the transfer transaction request is collected, and an initial transaction intent fingerprint of the transfer transaction request is established; During the execution of a transfer transaction, transfer execution behavior information is obtained, the initial transaction intent fingerprint is updated, and a corresponding transaction intent evolution trajectory is generated. By performing a consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory, the corresponding transaction intent deviation results are obtained; Based on the deviation of the transaction intention, the corresponding risk energy change is generated according to the preset risk mapping rules, and the corresponding risk energy change trajectory is constructed. Based on the risk energy change trajectory, the risk status of the transfer transaction request is determined, and the risk control level corresponding to the transfer transaction request is determined. Set risk control permission conditions, further determine whether the risk control permission conditions corresponding to the risk control level are met, and execute risk control actions matching the risk control level based on the determination result.

2. The risk control method for real-time transfer transactions according to claim 1, characterized in that, The step of collecting transaction-related information corresponding to the transfer transaction request and establishing the initial transaction intent fingerprint of the transfer transaction request when accepting the transfer transaction request specifically includes: When accepting a transfer transaction request, the transaction-related information corresponding to the transfer transaction request is collected. The transaction-related information includes transfer amount information, recipient information, transfer initiation time, and transfer operation environment information. The transaction-related information is structured and parsed, and different types of transaction-related information are mapped to a preset set of information dimensions to form a standard set of transaction features; Based on the standard transaction feature set, intent feature parameters that characterize the expected behavior of the transfer transaction are extracted. According to the preset intent feature combination rules, the intent feature parameters are structured and combined to generate the initial transaction intent fingerprint corresponding to the prime number transfer transaction request.

3. The risk control method for real-time transfer transactions according to claim 2, characterized in that, The steps of obtaining transfer execution behavior information, updating the initial transaction intent fingerprint, and generating a corresponding transaction intent evolution trajectory during the execution of a transfer transaction are as follows: During the execution of a transfer transaction, transfer execution behavior information is obtained, including transfer operation sequence change information, transfer confirmation behavior information, transfer execution delay information, and transfer amount adjustment information. The transfer execution behavior information is processed into events, converting it into a sequence of execution behavior events ordered by time. Based on the sequence of execution behavior events, behavioral evolution feature parameters reflecting the changing characteristics of transfer execution behavior are extracted. The behavioral evolution feature parameters include behavioral occurrence sequence features, behavioral type change features, and behavioral rhythm change features. The behavioral evolution feature parameters are associated and mapped with the initial transaction intent fingerprint. According to the preset intent update rule, the corresponding intent feature parameters in the initial transaction intent fingerprint are adjusted to generate the corresponding intermediate transaction intent state. The intermediate transaction intent states are continuously recorded according to the chronological order of the transfer execution, forming a transaction intent evolution trajectory that represents the changes in transaction intent as the transfer execution process progresses.

4. The risk control method for real-time transfer transactions according to claim 1, characterized in that, The step of performing consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation result is as follows: Based on the sequence of execution events corresponding to the transfer execution behavior information, a sequence of transaction intent state features corresponding to the sequence of execution events in the time dimension is extracted from the transaction intent evolution trajectory; Based on the sequence of execution behavior events and the sequence of transaction intent states, the degree of matching between the transfer execution behavior events and the corresponding transaction intent states is calculated according to the preset consistency analysis rules, and the corresponding intent consistency measurement results are obtained. Perform time-series trend analysis on the intent consistency measurement results to identify the consistency changes between transfer execution behavior and transaction intent evolution trajectory, and generate corresponding intent consistency change feature parameters; Based on the intent consistency change feature parameters, it is determined whether the current transfer execution behavior deviates from the intent evolution direction corresponding to the transaction intent evolution trajectory, and a corresponding transaction intent deviation result is generated.

5. The risk control method for real-time transfer transactions according to claim 4, characterized in that, The step of determining whether the current transfer execution behavior deviates from the intent evolution direction corresponding to the transaction intent evolution trajectory based on the intent consistency change feature parameters, and generating the corresponding transaction intent deviation result, specifically includes: Based on the aforementioned intent consistency change characteristic parameters, an intent deviation evaluation parameter set is constructed, which includes consistency change direction parameters, consistency change magnitude parameters, and consistency change rate parameters. Obtain the intention evolution direction feature parameters corresponding to the transaction intention evolution trajectory. The intention evolution direction feature parameters are used to characterize the subjective evolution direction of the transaction intention state as time changes. The consistency change direction parameter is compared with the intention evolution direction feature parameter to determine whether the consistency change direction of the transfer execution behavior is consistent with the intention evolution direction of the transaction intention evolution trajectory. If the direction of consistency change is inconsistent with the direction of intent evolution, the deviation strength of the transfer execution behavior relative to the trajectory of transaction intent evolution is comprehensively evaluated by combining the consistency change magnitude parameter and the consistency change rate parameter, and the deviation strength evaluation result is obtained to generate the corresponding transaction intent deviation result.

6. The risk control method for real-time transfer transactions according to claim 1, characterized in that, The step of generating a corresponding risk energy change based on the deviation result of the transaction intention according to a preset risk mapping rule, and constructing a corresponding risk energy change trajectory, specifically includes: Based on the deviation results of the transaction intent, analyze the deviation type parameters and deviation intensity parameters of the deviation status of the transfer execution behavior; Based on the deviation type parameter and deviation intensity parameter, according to the preset risk mapping rule, the deviation result of the transaction intention is mapped to the corresponding risk energy change, which is used to characterize the incremental or decremental change of the current transfer transaction risk status. Based on the change in risk energy, the current risk energy status corresponding to the transfer transaction is updated, and the corresponding risk energy status value is generated. According to the time sequence of the transfer transaction execution process, the risk energy state value generated at each moment is continuously recorded to construct a risk energy change trajectory that represents the dynamic change of risk state as the transfer execution process progresses.

7. The risk control method for real-time transfer transactions according to claim 1, characterized in that, The step of determining the risk status of the transfer transaction request based on the risk energy change trajectory and determining the risk control level corresponding to the transfer transaction request specifically includes: Based on the risk energy change trajectory, risk trajectory feature parameters that characterize the evolution of risk status are extracted. The risk trajectory feature parameters include risk energy change trend parameters, risk energy fluctuation amplitude parameters, and risk energy accumulation degree parameters. The risk trajectory feature parameters are matched with preset risk status determination rules to determine the current risk status type of the transfer transaction request; Based on the risk status type, obtain the risk status identifier parameter corresponding to the risk status type; Based on the risk status identifier parameter, the risk control level corresponding to the transfer transaction request is determined according to the preset risk control level mapping rule. The risk control level is used to indicate the level of risk control strategy implemented for the transfer transaction request.

8. The risk control method for real-time transfer transactions according to claim 1, characterized in that, The steps of setting risk control permission conditions, further determining whether the risk control permission conditions corresponding to the risk control level are met, and executing risk control actions matching the risk control level based on the determination result are as follows: Based on the risk control level, obtain the risk control permission condition set corresponding to the risk control level. The risk control permission condition set includes transaction entity permission conditions, transaction scenario permission conditions, and system operation status permission conditions. Based on the risk control permission condition set, the current transaction environment status, transaction subject status and system status of the transfer transaction request are matched and determined to generate the corresponding risk control permission determination result. When the risk control authority determination result meets the risk control authority conditions corresponding to the risk control level, a risk control action matching the risk control level is executed. When the risk control authority determination result does not meet the risk control authority conditions corresponding to the risk control level, the execution method of the risk control action is adjusted or the execution of the risk control action is delayed. The risk control actions include one or more of the following: delayed processing of transfer transactions, secondary verification of transfer transactions, restriction of transfer transaction execution, and blocking of transfer transactions.

9. A risk control system for real-time transfer transactions, characterized in that, The risk control method for real-time transfer transactions as described in any one of claims 1-8 includes: The intent fingerprint construction module, upon receiving a transfer transaction request, collects transaction-related information corresponding to the transfer transaction request and establishes an initial transaction intent fingerprint for the transfer transaction request; The intent trajectory generation module acquires transfer execution behavior information during the execution of a transfer transaction, updates the initial transaction intent fingerprint, and generates the corresponding transaction intent evolution trajectory. The intent deviation analysis module performs consistency analysis between the transfer execution behavior information and the transaction intent evolution trajectory to obtain the corresponding transaction intent deviation results. The risk energy change module generates the corresponding risk energy change amount according to the deviation result of the transaction intention and constructs the corresponding risk energy change trajectory according to the preset risk mapping rules. The control level determination module determines the risk status of the transfer transaction request based on the risk energy change trajectory, and determines the risk control level corresponding to the transfer transaction request. The control permission determination module sets risk control permission conditions, further determines whether the risk control permission conditions corresponding to the risk control level are met, and executes risk control actions matching the risk control level based on the determination result.

Images