Reward risk control device based on big data feedback dynamic regulation and control

Abstract

The invention discloses a reward risk control device based on big data feedback dynamic regulation and control, and particularly relates to the technical field of data mining, which comprises a data acquisition module, a risk control decision engine module, a risk assessment module and an anomaly detection module, the data acquisition module collects user transaction data and behavior data, preprocesses the collected data, and stores the processed data in a data warehouse; the risk control decision engine module is used for extracting transaction data from a data warehouse and constructing risk control rules including threshold setting, user portraits and behavior models; the risk assessment module is used for performing risk control query on a transaction request twice when a user initiates payment, and performing risk scoring on a primary request and a recheck request respectively; and the anomaly detection module monitors the transaction data in real time, detects abnormal behaviors by using a machine learning algorithm, and generates a comprehensive risk assessment report in combination with risk control rules and anomaly detection results in the payment process.

Description

Technical Field

[0001] The present invention relates to the field of data mining technology, and more specifically, to a reward risk control device based on dynamic regulation of big data feedback. Background Art

[0002] With the rapid development of modern financial services and e-commerce, fund security and risk control have become critical issues for businesses and platforms. Existing fund risk control systems generally employ a pre-payment interception approach. This involves invoking risk control services in real time to determine the feasibility of a transaction when a user initiates a payment. While this approach can mitigate risk to a certain extent, it also has significant shortcomings.

[0003] On the one hand, pre-payment interception requires extremely high levels of immediacy and reliability from risk control services. Risk control services must analyze vast amounts of data and make decisions in a remarkably short timeframe. Any delay or service unavailability could result in transactions being approved, exposing the platform to potential financial losses. For example, if the risk control system fails to respond promptly under high-concurrency conditions or experiences a system failure, non-compliant transactions could remain unblocked.

[0004] On the other hand, due to the complexity of the payment chain, the availability of risk control services is often affected by multiple factors, including network latency and system load. In such cases, platforms often choose to approve potentially risky transactions to ensure a better user experience, further increasing the risk of capital loss. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a reward risk control device based on dynamic regulation of big data feedback to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a reward risk control device based on dynamic regulation of big data feedback, comprising a data acquisition module, a risk control decision engine module, a risk assessment module, and an anomaly detection module;

[0007] Data collection module: collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in the data warehouse;

[0008] Risk control decision engine module: extracts transaction data from the data warehouse and builds risk control rules, including threshold settings, user profiles, and behavior models;

[0009] Risk Assessment Module: When a user initiates a payment, two risk control queries are performed on the transaction request, with risk scores assigned for the initial and review requests respectively;

[0010] Anomaly Detection Module: Monitors transaction data in real time, uses machine learning algorithms to detect abnormal behavior, and generates a comprehensive risk assessment report during the payment process, combining risk control rules with anomaly detection results.

[0011] In a preferred embodiment, the data collection module collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in a data warehouse. The specific steps are as follows:

[0012] Step A1: Obtain user transaction data from the ride-hailing platform, including transaction amount, transaction time, merchant information, and payment method. Use application analytics tools to track user behavior, including page visit records, click paths, and dwell time. Pre-process the collected data to remove invalid, duplicate, and erroneous data records.

[0013] Step A2, data upload: perform extraction, conversion, and loading operations, prepare the processed data, and upload them in batches to the data warehouse for storage.

[0014] In a preferred embodiment, when a user initiates a payment, the risk assessment module performs two risk control queries on the transaction request, respectively providing risk scores for the initial request and the review request. The specific steps are as follows:

[0015] Step C1, initial risk scoring: Receive the payment request initiated by the user, extract the user's transaction data characteristics, and use the established risk control rules to perform an initial risk scoring on the transaction request to obtain the risk assessment result: R1=α1·P(y=1|X)+α2·Indicator(A>T high )+α3·Indicator(A<T low ), based on the initial risk score, when R1>Threshold, it is marked as high risk and triggers a review request. When R1≤Threshold, the transaction is allowed to continue, where R1 is the initial risk score, Threshold is the risk threshold, and X is the user's transaction data feature;

[0016] Step C2, review risk score: In the review request stage, update the user's transaction data characteristics, use the hash algorithm to route the request, and check whether there is a cached result. If there is a cached result, use the cached result directly; the cached result represents the risk score value calculated based on the user's historical transaction data and characteristics, which reduces the need for recalculation and improves the response speed. If it does not exist, use the same risk score formula to obtain the review risk score: R2 = α1·P(y=1|X')+α2·Indicator(A>T high )+α3·Indicator(A<T low), when R2>Threshold, it is marked as high risk and the transaction is rejected. When R2≤Threshold, the transaction is approved, where R2 is the review risk score, Threshold is the risk threshold, and X' is the updated user transaction data feature.

[0017] In a preferred embodiment, the anomaly detection module monitors transaction data in real time, uses machine learning algorithms to detect abnormal behavior, and generates a comprehensive risk assessment report during the payment process by combining the risk control decision engine's scores with the anomaly detection results. The specific steps are as follows:

[0018] Step D1, anomaly detection: During the transaction process, the amount A of each transaction is monitored in real time. i and compared with the threshold, when A i <T low 、A i >T high , marked as abnormal transactions, the support vector machine model is trained using samples of normal and abnormal labels in historical transaction data, and the predicted label is Determine whether the transaction is abnormal based on the predicted label, where A i is the amount of the i-th transaction, is the predicted label, 1 is abnormal, 0 is normal, ω is the weight vector, b is the bias term, and sign is the activation function;

[0019] Step D2: Calculate a comprehensive risk assessment value based on the anomaly detection results and the risk control rule score to determine the risk of the transaction. The mathematical expression of the comprehensive risk assessment value is: Among them, R is the risk control score, which reflects the potential risk of the transaction. λ1 and λ2 are weight coefficients used to adjust the impact of risk control rule scores and anomaly detection results. is the anomaly detection result; the comprehensive risk decision results are as follows:

[0020] when Approve the transaction;

[0021] when Request for review;

[0022] When R 综合 >Threshold, reject the transaction, where Threshold is the risk threshold; based on the comprehensive risk assessment value, a comprehensive risk assessment report is generated, including user ID, transaction amount, transaction timestamp, risk control rule score, prediction results, and the comprehensive risk decision result: approve the transaction, review the request, or reject the transaction.

[0023] The beneficial effects of the present invention are: collecting user transaction data and behavior data, pre-processing the collected data, storing the processed data in a data warehouse, extracting transaction data from the data warehouse, and constructing risk control rules, including threshold setting, user portraits and behavior models. When the user initiates a payment, two risk control queries are performed on the transaction request, and risk scores are performed for the initial and review requests respectively. The transaction data is monitored in real time, and a machine learning algorithm is used to detect abnormal behavior. During the payment process, a comprehensive risk assessment report is generated by combining risk control rules with anomaly detection results. The two risk control query mechanisms can provide more accurate risk assessments and effectively reduce misjudgments and missed judgments. The application of hash algorithms and cache mechanisms improves the computing speed of risk control services and the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a system structure diagram of the present invention;

[0025] Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0027] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the described features. In the description of this application, "plurality" means two or more, unless otherwise specifically specified.

[0028] In the description of this application, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art will recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0029] Example 1

[0030] This embodiment provides Figure 1 The device is a reward risk control device based on dynamic regulation of big data feedback, which specifically includes a data acquisition module, a risk control decision engine module, a risk assessment module, and an anomaly detection module.

[0031] Data collection module: collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in the data warehouse;

[0032] Risk control decision engine module: extracts transaction data from the data warehouse and builds risk control rules, including threshold settings, user profiles, and behavior models;

[0033] Risk Assessment Module: When a user initiates a payment, two risk control queries are performed on the transaction request, with risk scores assigned for the initial and review requests respectively;

[0034] Anomaly Detection Module: Monitors transaction data in real time, uses machine learning algorithms to detect abnormal behavior, and generates a comprehensive risk assessment report during the payment process, combining risk control rules with anomaly detection results.

[0035] In this embodiment, the data collection module specifically needs to be explained. The data collection module collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in a data warehouse. Centralized data storage facilitates data management and maintenance, reduces data redundancy, and improves data consistency and reliability. The specific steps are as follows:

[0036] Step A1: Obtain user transaction data from the ride-hailing platform, including transaction amount, transaction time, merchant information, and payment method. Use application analytics tools to track user behavior, including page visit records, click paths, and dwell time. Pre-process the collected data to remove invalid, duplicate, and erroneous data records.

[0037] Step A2, data upload: perform extraction, conversion, and loading operations, prepare the processed data, and upload them in batches to the data warehouse for storage.

[0038] In this embodiment, the risk control decision engine module is specifically described. The risk control decision engine module extracts transaction data from the data warehouse and constructs risk control rules, including threshold settings, user profiles, and behavior models. This can more accurately assess the risk level of each user and improve risk control efficiency. The specific steps are as follows:

[0039] Step B1, data extraction: extract transaction data from the data warehouse, including transaction frequency and amount distribution, and represent the extracted transaction data set as D = {d1, d2, ..., d n}, where d n The nth transaction record includes the key fields transaction amount, transaction timestamp, user ID, and transaction status S. When S=0, it indicates a failed transaction, and when S=1, it indicates a successful transaction.

[0040] Step B2: Threshold setting: Use statistical methods to calculate the mean μ and standard deviation σ to set the transaction amount threshold T high =μ+k·σ and T low =μ-k·σ, to control the detection of abnormal transactions, where k is a constant that controls the sensitivity of the threshold, T high and T low are the upper and lower thresholds respectively;

[0041] Step B3, User Profile: Based on the collected user transaction data and behavior data, extract transaction frequency, average transaction amount, and transaction success rate characteristics and construct a user profile. The specific calculation formula is as follows:

[0042]

[0043] Among them, T r Indicates the transaction frequency, M indicates the number of transactions within T time, is the average transaction amount, N is the total number of user transactions, A i is the amount of the i-th transaction, S rate is the transaction success rate, S is the number of successful transactions, and the user portrait is

[0044] Step B4, Behavior Model: User Profile The transaction data is converted into a feature matrix X and a label y, wherein the feature matrix X is the transaction data feature of the user, and the label y is the transaction status, 1 indicates success and 0 indicates failure, y = [S1, S2, ..., S N ], where SN Represents the status of the Nth transaction and uses the logistic regression algorithm to build a behavioral model: Among them, P(y=1|X) is the probability of successful transaction under feature X, β0,β1,...,β z is the parameter vector of the model, X z is the zth feature representation;

[0045] Step B5: Combine the threshold, user profile, and behavior model to form a complete risk control rule score. The mathematical expression is R = α1·P(y=1|X)+α2·Indicator(A>T high )+α3·Indicator(A<T low ), where P(y=1|X) is the probability of a successful transaction under feature X, A is the transaction amount, Indicator() is the indicator function, which is 1 when the transaction amount exceeds the upper and lower limit thresholds, and 0 otherwise, and (α1, α2, α3) are weight coefficients.

[0046] In this embodiment, the risk assessment module specifically requires explanation. When a user initiates a payment, the risk assessment module performs two risk control queries on the transaction request, assigning risk scores for the initial request and the review request respectively. These two risk control queries enable a more comprehensive assessment of transaction risk and improve the accuracy of risk assessment. The specific steps are as follows:

[0047] Step C1, initial risk scoring: Receive the payment request initiated by the user, extract the user's transaction data characteristics, and use the established risk control rules to perform an initial risk scoring on the transaction request to obtain the risk assessment result: R1=α1·P(y=1|X)+α2·Indicator(A>T high )+α3·Indicator(A<T low ), based on the initial risk score, when R1>Threshold, it is marked as high risk and triggers a review request. When R1≤Threshold, the transaction is allowed to continue, where R1 is the initial risk score, Threshold is the risk threshold, and X is the user's transaction data feature;

[0048] Step C2, review risk score: In the review request stage, update the user's transaction data characteristics, use the hash algorithm to route the request, and check whether there is a cached result. If there is a cached result, use the cached result directly; the cached result represents the risk score value calculated based on the user's historical transaction data and characteristics, which reduces the need for recalculation and improves the response speed. If it does not exist, use the same risk score formula to obtain the review risk score: R2 = α1·P(y=1|X')+α2·Indicator(A>T high)+α3·Indicator(A<T low ), when R2>Threshold, it is marked as high risk and the transaction is rejected. When R2≤Threshold, the transaction is approved, where R2 is the review risk score, Threshold is the risk threshold, and X' is the updated user transaction data feature.

[0049] In this embodiment, the anomaly detection module is specifically described. The anomaly detection module monitors transaction data in real time and uses machine learning algorithms to detect abnormal behavior. During the payment process, the module combines the risk control decision engine's scores with the anomaly detection results to generate a comprehensive risk assessment report. This can reduce manual intervention and resource waste, thereby lowering overall operating costs. The specific steps are as follows:

[0050] Step D1, anomaly detection: During the transaction process, the amount A of each transaction is monitored in real time. i and compared with the threshold, when A i <T low 、A i >T high , marked as abnormal transactions, the support vector machine model is trained using samples of normal and abnormal labels in historical transaction data, and the predicted label is Determine whether the transaction is abnormal based on the predicted label, where A i is the amount of the i-th transaction, is the predicted label, 1 is abnormal, 0 is normal, ω is the weight vector, b is the bias term, and sign is the activation function;

[0051] Step D2: Calculate a comprehensive risk assessment value based on the anomaly detection results and the risk control rule score to determine the risk of the transaction. The mathematical expression of the comprehensive risk assessment value is: Among them, R is the risk control score, which reflects the potential risk of the transaction. λ1 and λ2 are weight coefficients used to adjust the impact of risk control rule scores and anomaly detection results. is the anomaly detection result; the comprehensive risk decision results are as follows:

[0052] when Approve the transaction;

[0053] when Request for review;

[0054] When R 综合 >Threshold, reject the transaction, where Threshold is the risk threshold; based on the comprehensive risk assessment value, a comprehensive risk assessment report is generated, including user ID, transaction amount, transaction timestamp, risk control rule score, prediction results, and the comprehensive risk decision result: approve the transaction, review the request, or reject the transaction.

[0055] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0056] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0057] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0058] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0059] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0060] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0061] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A reward risk control device based on dynamic regulation of big data feedback, characterized by: It includes data collection module, risk control decision engine module, risk assessment module, and anomaly detection module; Data collection module: collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in the data warehouse; Risk control decision engine module: extracts transaction data from the data warehouse and builds risk control rules, including threshold settings, user profiles, and behavior models; Risk Assessment Module: When a user initiates a payment, two risk control queries are performed on the transaction request, with risk scores assigned for the initial and review requests respectively; Anomaly Detection Module: Monitors transaction data in real time, uses machine learning algorithms to detect abnormal behavior, and generates a comprehensive risk assessment report during the payment process, combining risk control rules with anomaly detection results.

2. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 1, characterized in that: The data collection module collects user transaction data and behavior data, pre-processes the collected data, and stores the processed data in the data warehouse. The specific steps are as follows: Step A1: Obtain user transaction data from the ride-hailing platform, including transaction amount, transaction time, merchant information, and payment method. Use application analytics tools to track user behavior, including page visit records, click paths, and dwell time. Pre-process the collected data to remove invalid, duplicate, and erroneous data records. Step A2, data upload: perform extraction, conversion, and loading operations, prepare the processed data, and upload them in batches to the data warehouse for storage.

3. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 1, characterized in that: The risk control decision engine module extracts transaction data from the data warehouse and constructs risk control rules, including threshold settings, user profiling, and behavior models. The specific steps are as follows: Step B1, data extraction: extract transaction data from the data warehouse, including transaction frequency and amount distribution, and represent the extracted transaction data set as D = {d1, d2, ..., d n }, where d n The nth transaction record includes the key fields transaction amount, transaction timestamp, user ID, and transaction status S. When S=0, it indicates a failed transaction, and when S=1, it indicates a successful transaction. Step B2: Threshold setting: Use statistical methods to calculate the mean μ and standard deviation σ to set the transaction amount threshold T high =μ+k·σ and T low =μ-k·σ, to control the detection of abnormal transactions, where k is a constant that controls the sensitivity of the threshold, T high and T low are the upper and lower thresholds respectively; The threshold, user profile and behavior model are combined to form a complete risk control rule score. The mathematical expression is R = α1·P(y=1|X)+α2·Indicator(A>T high )+α3·Indicator(A<T low ), where P(y=1|X) is the probability of a successful transaction under feature X, A is the transaction amount, Indicator() is the indicator function, which is 1 when the transaction amount exceeds the upper and lower limit thresholds, and 0 otherwise, and (α1, α2, α3) are weight coefficients.

4. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 3, characterized in that: The specific steps of user profiling include: extracting transaction frequency, average transaction amount and transaction success rate characteristics based on the collected user transaction data and behavior data and constructing a user profile. The specific calculation formula is as follows: Among them, T r Indicates the transaction frequency, M indicates the number of transactions within T time, is the average transaction amount, N is the total number of user transactions, A i is the amount of the i-th transaction, S rate is the transaction success rate, S is the number of successful transactions, and the user portrait is 5. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 3, characterized in that: The specific steps of the behavior model include: The transaction data is converted into a feature matrix X and a label y, wherein the feature matrix X is the transaction data feature of the user, and the label y is the transaction status, 1 indicates success and 0 indicates failure, y = [S1, S2, ..., S N ], where S N Represents the status of the Nth transaction and uses the logistic regression algorithm to build a behavioral model: Among them, P(y=1|X) is the probability of successful transaction under feature X, β0,β1,...,β z is the parameter vector of the model, X z is the zth feature representation.

6. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 1, characterized in that: When a user initiates a payment, the risk assessment module performs two risk control queries on the transaction request, scoring the risk for the initial and review requests respectively. The specific steps are as follows: Step C1, initial risk scoring: Receive the payment request initiated by the user, extract the user's transaction data characteristics, and use the established risk control rules to perform an initial risk scoring on the transaction request to obtain the risk assessment result: R1=α1·P(y=1|X)+α2·Indicator(A>T high )+α3·Indicator(A<T low ), based on the initial risk score, when R1>Threshold, it is marked as high risk and triggers a review request. When R1≤Threshold, the transaction is allowed to continue, where R1 is the initial risk score, Threshold is the risk threshold, and X is the user's transaction data feature; Step C2, review risk score: In the review request stage, update the user's transaction data characteristics, use the hash algorithm to route the request, and check whether there is a cached result. If there is a cached result, use the cached result directly; the cached result represents the risk score value calculated based on the user's historical transaction data and characteristics. If it does not exist, use the same risk score formula to obtain the review risk score: R2 = α1·P(y=1|X')+α2·Indicator(A>T high )+α3·Indicator(A<T low ), when R2>Threshold, it is marked as high risk and the transaction is rejected. When R2≤Threshold, the transaction is approved, where R2 is the review risk score, Threshold is the risk threshold, and X' is the updated user transaction data feature.

7. The reward risk control device based on dynamic regulation and control based on big data feedback according to claim 1, characterized in that: The anomaly detection module monitors transaction data in real time and uses machine learning algorithms to detect abnormal behavior. During the payment process, it combines the risk control decision engine's scores with the anomaly detection results to generate a comprehensive risk assessment report. The specific steps are as follows: Step D1, anomaly detection: During the transaction process, the amount A of each transaction is monitored in real time. i and compared with the threshold, when A i <T low 、A i >T high , marked as abnormal transactions, the support vector machine model is trained using samples of normal and abnormal labels in historical transaction data, and the predicted label is Determine whether the transaction is abnormal based on the predicted label, where A i is the amount of the i-th transaction, is the predicted label, 1 is abnormal, 0 is normal, ω is the weight vector, b is the bias term, and sign is the activation function; Step D2: Calculate a comprehensive risk assessment value based on the anomaly detection results and the risk control rule score to determine the risk of the transaction. The mathematical expression of the comprehensive risk assessment value is: Among them, R is the risk control score, λ1 and λ2 are weight coefficients, is the anomaly detection result; the comprehensive risk decision results are as follows: When R 综合 ≤Threshold and Approve the transaction; When R 综合 ≤Threshold and Request for review; When R 综合 >Threshold, reject the transaction, where Threshold is the risk threshold; based on the comprehensive risk assessment value, a comprehensive risk assessment report is generated, including user ID, transaction amount, transaction timestamp, risk control rule score, prediction results, and the comprehensive risk decision result: approve the transaction, review the request, or reject the transaction.

Images