Mobile payment method, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of mobile payment risk control, and relates to a mobile payment method, electronic equipment and a storage medium. According to the invention, the reference payment quota is matched according to the risk score-reference quota mapping table, so that the transaction quota is dynamically adjusted according to the personal payment risk condition of the user, and the condition that the high-risk user is excessively credit-granted or the low-risk user is excessively strictly limited is avoided; the real-time comprehensive risk score of the current mobile payment order is analyzed by performing multi-dimensional fusion on the basic risk score, the risk score environment factor and the current personal risk score of the user, the linkage effect of the risk score environment factor and the current personal risk score of the user is fully considered, and the situation that the mobile payment risk assessment dimension is single is avoided; the missed judgment rate of the mobile payment transaction risk is reduced; by combining the risk control measure and the limit management, the limit optimal configuration under the risk controllability is realized, and the user experience and safety under the abnormal transaction scene are guaranteed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of mobile payment risk control, and relates to a mobile payment method, an electronic device and a storage medium. BACKGROUND

[0002] Mobile payment, also known as mobile phone payment, is a service mode allowing a user to use a mobile terminal to make an account payment for goods or services consumed. The entire mobile payment value chain includes mobile operators, payment service providers, application providers, device providers, system integrators, merchants and end users.

[0003] With the popularization of mobile payment, payment security risks present dynamic, scenario and regional characteristics. The existing mobile payment risk control system can meet the basic needs, but still has certain deficiencies: (1) the existing mobile payment risk control system relies on static user portraits, and it is difficult to capture regional risk fluctuations in real time; (2) the risk assessment dimension of the existing mobile payment risk control system is single, only based on user historical behavior or single transaction characteristics, ignoring the risk influence caused by the change of regional environment, and also ignoring the risk influence caused by the change of individual user transaction; (3) the risk control measures of the existing mobile payment risk control system are disconnected with the quota management, resulting in high misjudgment rate and response lag problem, especially in high-risk areas or abnormal transaction scenarios, it is difficult to balance user experience and security. SUMMARY

[0004] In view of this, in order to solve the problems proposed in the background art, the present application provides a mobile payment method, an electronic device and a storage medium.

[0005] The purpose of the application can be achieved by the following technical solutions: the first aspect of the application provides a mobile payment method, comprising: step one, acquiring user mobile payment transaction data of a plurality of target regions in a current monitoring time period, and constructing a regional payment behavior portrait and a user individual payment behavior portrait accordingly.

[0006] Step two, dynamically calculating a risk score environmental factor based on the regional payment behavior portrait, dynamically calculating a user current personal risk score based on the user individual payment behavior portrait, and matching a reference payment quota according to a risk score-reference quota mapping table.

[0007] Step three, capturing transaction detailed data of a current mobile payment order initiated by any user belonging to any target region in real time, and calculating a basic risk score of the current mobile payment order through a basic risk score rule.

[0008] Step 4: Perform a weighted fusion calculation on the basic risk score combined with the risk score environmental factor and the user's current personal risk score to generate a real-time comprehensive risk score for the current mobile payment order, and compare it with the preset risk threshold to determine whether it is a risk event.

[0009] Step 5: If the current mobile payment order is a risk event, determine its risk level and automatically trigger risk control measures corresponding to the risk level.

[0010] A second aspect of the present invention provides an electronic device, characterized in that it includes a processor and a memory for storing execution instructions of the processor, wherein the processor is configured to execute the mobile payment method described in the present invention.

[0011] A third aspect of the present invention provides a storage medium storing one or more programs, which can be executed by one or more processors to implement the steps in the mobile payment method of the present invention.

[0012] Compared with the existing technology, the beneficial effects of the present invention are as follows: 1. The present invention constructs regional payment behavior portraits and user individual payment behavior portraits based on the mobile payment transaction data of users in multiple target areas during the current monitoring period, and comprehensively depicts user payment habits and regional payment characteristics in multiple dimensions, avoiding the one-sidedness of a single data dimension and improving the efficiency of identifying abnormal transactions.

[0013] 2. The present invention matches the benchmark payment limit according to the risk score-benchmark limit mapping table, thereby dynamically adjusting the transaction limit based on the user's current personal payment risk situation, avoiding excessive credit for high-risk users or overly strict restrictions on low-risk users.

[0014] 3. The present invention performs weighted fusion calculation based on the basic risk score, risk score environmental factors and user's current personal risk score obtained through analysis to obtain a real-time comprehensive risk score for the current mobile payment order. Through multi-dimensional fusion analysis, it fully considers the linkage effect of risk score environmental factors and user's current personal risk score, avoids the single dimension of mobile payment risk assessment, and reduces the missed judgment rate of mobile payment transaction risks.

[0015] 4. The present invention determines whether the current mobile payment order is a risk event. If so, it determines its risk level and automatically triggers risk control measures corresponding to the risk level. It combines risk control measures with credit limit management to achieve optimal credit limit configuration under controllable risks, which is beneficial to ensuring user experience and security in abnormal transaction scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 It is a schematic diagram of the implementation of the method steps of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] Example 1

[0020] See Figure 1 As shown, the present invention provides a mobile payment method, comprising the following steps: Step 1, obtaining user mobile payment transaction data in multiple target areas within the current monitoring time period, and constructing regional payment behavior portraits and user individual payment behavior portraits based on the data.

[0021] The regional payment behavior portrait represents the characteristics of the regional payment pattern, and the user individual payment behavior portrait represents the characteristics of the user's individual payment habit.

[0022] In a specific example, the user mobile payment transaction data includes but is not limited to the transaction amount, transaction timestamp, transaction GPS coordinates, transaction device, payment method and operation behavior sequence of each transaction order of several users in multiple target areas during the current monitoring time period.

[0023] It should be noted that the specific method of obtaining the user mobile payment transaction data is: extracting it from the online shopping interaction platform of several users in multiple target areas. The same applies to the detailed transaction data below and will not be repeated here.

[0024] As a preferred feasible embodiment, the specific construction process of the regional payment behavior portrait and the user individual payment behavior portrait includes: extracting the transaction amount, transaction timestamp, transaction GPS coordinates, transaction device, payment method and operation behavior sequence of each transaction order of several users belonging to multiple target areas during the current monitoring time period from the user mobile payment transaction data, and performing statistics on them to determine the regular transaction amount range, active time period, commonly used GPS coordinates, commonly used devices, commonly used payment methods and typical operation behavior sequence of several users belonging to multiple target areas during the current monitoring time period, and recording them as the user individual payment behavior profiles of several users belonging to multiple target areas.

[0025] It should be further explained that the specific operations for determining the regular transaction amount range, active time period, common GPS coordinates, common devices, common payment methods and typical operation behavior sequence of several users in multiple target areas during the current monitoring period are as follows: (1) Filter the transaction amount of each transaction order of several users in multiple target areas during the current monitoring period, and use the quantile to set their regular transaction amount range, which can be set to [25% quantile, 75% quantile], covering most normal transaction amounts. Large transactions can be extended to [25% quantile-1.5×IQR, 75% quantile+1.5×IQR], where IQR is the transaction amount of the interquartile range.

[0026] (2) The transaction timestamps of each transaction order of several users in multiple target areas during the current monitoring period are counted by weekdays and weekends to distinguish the active patterns of daily and weekends. Then, the transaction timestamps are counted by hour to generate an hour-transaction times distribution graph to identify high-frequency trading periods. Thus, the continuous hour intervals of weekdays and weekends in which the transaction times account for more than 10% of the daily average are obtained. These intervals are recorded as the active time periods of weekdays and weekends for several users in multiple target areas during the current monitoring period, and are collectively referred to as the active time periods of several users in multiple target areas during the current monitoring period.

[0027] (3) Count the number of transaction orders of the multiple users in the target areas during the current monitoring period according to the GPS coordinates, select the GPS coordinate with the highest number of transaction orders, and record it as the commonly used GPS coordinates of the multiple users in the target areas during the current monitoring period. Similarly, the commonly used devices and commonly used payment methods of the multiple users in the target areas during the current monitoring period can be obtained.

[0028] (4) Using the Apriori algorithm, identify the operation behavior sequence with the highest frequency in the operation behavior sequences of each transaction order of several users belonging to multiple target areas during the current monitoring period, and record it as the typical operation behavior sequence of several users belonging to multiple target areas during the current monitoring period.

[0029] As a specific example, the individual payment behavior profile of a user in a target area can be the regular transaction amount range (such as [50 yuan, 200 yuan]); active time period (such as 18:00-21:00 on weekdays, 14:00-17:00 on weekends); commonly used GPS coordinates (such as "XX District XX Shopping Mall"); commonly used devices (such as "iPhone 15"); commonly used payment methods (such as WeChat Pay (60%)); typical operation behavior sequence (such as "select product → confirm quantity → fingerprint payment").

[0030] Based on the individual payment behavior portraits of several users in multiple target areas, the amount distribution map, time pattern map, payment method distribution map and operation behavior sequence distribution map of multiple target areas are determined, and recorded as the regional payment behavior profiles of the multiple target areas.

[0031] In a specific example, the amount distribution graph includes but is not limited to the transaction amount distribution graphs of the mean, median and mode. The time regularity graph includes but is not limited to the transaction order distribution graphs of peak hours and low-frequency hours.

[0032] It should be further explained that the specific operations of determining the amount distribution map, time pattern map, payment method distribution map and operation behavior sequence distribution map of multiple target areas are as follows: (1) extracting the regular transaction amount ranges of several users in multiple target areas during the current monitoring period, and calculating the average, median and mode of the regular transaction amount ranges of users in multiple target areas during the current monitoring period, and drawing a two-dimensional coordinate system with the area as the horizontal coordinate and the transaction amount as the vertical coordinate, in which the points are drawn and connected according to the average, median and mode of the regular transaction amount ranges of users in multiple target areas during the current monitoring period to obtain the amount distribution map of multiple target areas.

[0033] (2) Extract the active time periods of several users belonging to multiple target areas during the current monitoring period, and calculate the average number of active time periods of users belonging to multiple target areas during the current monitoring period. Draw a two-dimensional coordinate system with the area as the horizontal coordinate and the active time period as the vertical coordinate. Draw points and connect lines according to the average number of active time periods of users belonging to multiple target areas during the current monitoring period to obtain a time pattern diagram of multiple target areas.

[0034] (3) Extract the commonly used payment methods of several users in multiple target areas during the current monitoring period, and draw a two-dimensional coordinate system with the area as the horizontal coordinate and the number of users of the payment method as the vertical coordinate. In the system, the number of users of each commonly used payment method of the users in multiple target areas during the current monitoring period is used to draw points and connect lines to obtain a payment method distribution map for multiple target areas. Similarly, a distribution map of the operation behavior sequence for multiple target areas can be obtained.

[0035] Based on the mobile payment transaction data of users in multiple target areas during the current monitoring period, the present invention constructs regional payment behavior portraits and user individual payment behavior portraits, comprehensively depicting user payment habits and regional payment characteristics in multiple dimensions, avoiding the one-sidedness of a single data dimension and improving the efficiency of identifying abnormal transactions.

[0036] Step 2: Dynamically calculate the risk score environment factor based on the payment behavior profile of the region, dynamically calculate the user's current personal risk score based on the individual payment behavior profile of the user, and match the benchmark payment limit according to the risk score-benchmark limit mapping table.

[0037] The risk scoring environmental factors represent the current regional risk level.

[0038] As a preferred feasible embodiment, the specific calculation process of the risk scoring environment factor and the user's current personal risk score includes: calculating the risk transaction ratio, risk transaction amount ratio and abnormal device activity of the corresponding target area based on the regional payment behavior portraits of multiple target areas and the user's individual payment behavior portraits, and weightedly summing them according to the preset corresponding weights to obtain the risk scoring environment factors of multiple target areas.

[0039] It should be further explained that the specific calculation process of the risky transaction ratio includes: extracting the total number of transactions and the number of risky transactions in the corresponding target area from the regional payment behavior profiles of multiple target areas, denoted as M1 and M1′ respectively, and calculating according to the formula Get the risk transaction ratio R1.

[0040] It should be noted that the specific operation of extracting the total transaction count and risky transaction count for each target region from the regional payment behavior profiles of multiple target regions is to count the total number of transaction orders of users in the multiple target regions during the current monitoring period as the total transaction count for the multiple target regions. The risky transaction count is the number of transaction orders determined to be risk events among the transaction orders of users in the multiple target regions during the current monitoring period.

[0041] The specific calculation process of the risk transaction amount ratio includes: extracting the total transaction amount and the total risk transaction amount of the corresponding target area from the regional payment behavior portraits of multiple target areas, respectively denoted as Q1 and Q1′, and calculating according to the formula The proportion of risky transaction amount R2 reflects the severity of capital loss.

[0042] It should be noted that the total transaction amount is the sum of all transaction orders (including normal transactions and risky transactions) in the target region during the current monitoring period, and is used to measure the overall transaction volume in the region. The total risky transaction amount is the sum of all transaction orders in the target region identified as risk events during the current monitoring period, reflecting the potential scale of financial losses caused by risky transactions.

[0043] The specific calculation process of the abnormal device activity includes: extracting the total device session number and abnormal device session number of the corresponding target area from the individual payment behavior profiles of several users in multiple target areas, denoted as H1 and H1′ respectively, and calculating according to the formula Get the abnormal device activity R3, where The device risk score is the average value. The device risk score is calculated in real time based on features such as jailbreak / proxy IP / high-frequency location jumps.

[0044] It should be noted that the total number of device sessions refers to the total number of payment sessions initiated by all devices in the target area during the current monitoring period, reflecting the level of device usage activity (a single session can include multiple transactions). The number of abnormal device sessions refers to the number of payment sessions initiated by abnormal devices in the target area during the current monitoring period. Abnormal devices refer to devices with security risk characteristics (such as jailbroken devices, devices using proxy IPs, and devices with frequent location hopping).

[0045] As a specific example, (1) the preset corresponding weights of the risk transaction ratio, risk transaction amount ratio and abnormal device activity in the target area can be 0.35, 0.4 and 0.25 respectively. Among them, the preset corresponding weight of the risk transaction ratio is 0.35, which reflects the frequency of risk transactions in the target area. The weight is medium, which reflects the focus on the risk at the transaction quantity level. The preset corresponding weight of the risk transaction amount ratio is 0.4, which reflects the severity of the capital loss in the risk transaction. It has the highest weight and prioritizes the control of capital security (such as the risk of large-scale theft). The preset corresponding weight of the abnormal device activity is 0.25, which reflects the risk at the device level (such as jailbroken devices, high-frequency location jumps). The weight is low but cannot be ignored.

[0046] (2) The preset corresponding weights of the risk transaction ratio, risk transaction amount ratio and abnormal device activity in the target area can be 0.2, 0.3 and 0.3 respectively. Among them, the preset corresponding weight of the risk transaction ratio is 0.2, which reflects that low-frequency large-value risk transactions may not be reflected in the transaction ratio, so the weight is the lowest. The preset corresponding weight of the risk transaction amount ratio is 0.5, which reflects the focus on the risk of large-value capital loss (such as corporate-to-public transfer scenarios), so the weight is the highest. The preset corresponding weight of the abnormal device activity is 0.3, which reflects that device risk is used as an auxiliary indicator to prevent large-value fraud initiated by abnormal devices.

[0047] It should be noted that the preset corresponding weights of the risk transaction ratio, risk transaction amount ratio and abnormal device activity in the target area are not fixed and can be adjusted according to specific needs, which will not be elaborated here.

[0048] Based on the individual payment behavior portraits of several users in multiple target areas, the risk transaction ratio, risk transaction amount ratio and abnormal device activity of the corresponding users in the corresponding target areas are calculated, and they are matched with the risk transaction ratios, risk transaction amount ratios and abnormal device activity corresponding risk scores of the users stored in the database, so as to obtain the risk transaction ratios, risk transaction amount ratios and abnormal device activity corresponding risk scores of the several users in the multiple target areas, and sum them up to obtain the current personal risk scores of the users in the multiple target areas.

[0049] It should be further explained that the calculation method of the risk transaction ratio, risk transaction amount ratio and abnormal device activity of the corresponding users in the target area is the same as that of the target area, and will not be repeated here.

[0050] As a preferred feasible embodiment, the specific process of matching the benchmark payment limit according to the risk score-benchmark limit mapping table includes: matching the current personal risk scores of several users belonging to multiple target areas with the benchmark payment limits corresponding to the current personal risk score ranges of each user in the risk score-benchmark limit mapping table; if the current personal risk score of a user belonging to a certain target area falls within the current personal risk score range of the user, then the benchmark payment limit corresponding to the current personal risk score range of the user is used as the benchmark payment limit of the user belonging to the target area, thereby obtaining the benchmark payment limits of several users belonging to multiple target areas.

[0051] As a specific example, the risk score-benchmark limit mapping is represented as shown in the following Table 1.

[0052] Table 1 Example of risk score-benchmark limit mapping table

[0053]

[0054] It should be noted that risk is negatively correlated with credit limit: the higher the risk score (such as 81-100 points), the higher the probability of abnormal payment behavior or the risk of capital loss of the user, and therefore lower single and daily cumulative limits (such as 2,000 yuan per transaction and 5,000 yuan per day) are restricted to control risk exposure; the lower the risk score (such as 0-30 points), the more reliable the user's credit performance, and a higher payment limit is allowed (such as 20,000 yuan per transaction and 50,000 yuan per day), which improves payment convenience.

[0055] The present invention matches the benchmark payment limit according to the risk score-benchmark limit mapping table, realizes dynamic adjustment of the transaction limit according to the user's current personal payment risk situation, and avoids excessive credit for high-risk users or overly strict restrictions on low-risk users.

[0056] Step 3: Capture the transaction details of the current mobile payment order initiated by any user in any target area in real time, and calculate the basic risk score of the current mobile payment order using the basic risk scoring rules.

[0057] In a specific example, the transaction details include but are not limited to the transaction amount, transaction timestamp, transaction GPS coordinates and operation behavior sequence of the current mobile payment order initiated by the user in the target area.

[0058] As a preferred feasible embodiment, the specific content of the basic risk scoring rules includes: extracting the transaction amount, transaction timestamp, transaction GPS coordinates and operation behavior sequence of the current mobile payment order initiated by the user in the target area from the transaction details data, and comparing them with the regular transaction amount range, active time period, commonly used GPS coordinates and typical operation behavior sequence in the user individual payment behavior profile of the corresponding user in the target area, and calculating the deviation of the current mobile payment order initiated by the user in the target area.

[0059] It should be further explained that the specific process of comparing the transaction amount, transaction timestamp, transaction GPS coordinates, transaction device, payment method and operation behavior sequence with the conventional transaction amount range, active time period, common GPS coordinates, common devices, common payment methods and typical operation behavior sequence is as follows: (1) According to the transaction amount deviation calculation formula Get the transaction amount deviation of the current mobile payment order initiated by users in the target area DJE, JEZ, and JEK are respectively the transaction amounts of the current mobile payment orders initiated by users in the target area and the midpoint and width of the regular transaction amount range in the individual payment behavior profile of the corresponding users in the target area.

[0060] For example, assuming the regular transaction amount range is [50 yuan, 200 yuan], the midpoint is 125 yuan, the width is 150 yuan, if the current amount is 300 yuan, the deviation is

[0061]

[0062] (2) Based on the transaction timestamp deviation calculation formula Get the transaction timestamp deviation of the current mobile payment order initiated by the user in the target area DS, DSZ, and DSK are the transaction timestamp of the current mobile payment order initiated by the user in the target area and the midpoint and width of the active time period in the individual payment behavior profile of the corresponding user in the target area.

[0063] (3) Based on the calculation formula of transaction GPS coordinate deviation Get the transaction GPS coordinate deviation of the current mobile payment order initiated by the user in the target area Where d0 and Δd are respectively the distance and permitted distance between the transaction GPS coordinates of the current mobile payment order initiated by the user in the target area and the transaction GPS coordinates in the individual payment behavior profile of the corresponding user in the target area.

[0064] (4) According to the calculation formula of the deviation degree of the operation behavior sequence Get the deviation degree of the operation behavior sequence of the current mobile payment order initiated by the user in the target area Among them, n0 and n are respectively the number of abnormal steps existing after comparing the operation behavior sequence of the current mobile payment order initiated by the user in the target area with the typical operation behavior sequence in the user individual payment behavior profile of the corresponding user in the target area, and the total number of steps existing in the typical operation behavior sequence in the user individual payment behavior profile of the corresponding user in the target area.

[0065] For example, assuming that the typical operation behavior sequence contains 3 steps, the operation behavior sequence of the current mobile payment order contains 4 steps and 1 step is abnormal, the deviation degree is

[0066]

[0067] It should be further explained that the specific calculation method of the deviation degree of the current mobile payment orders initiated by users in the target area includes: according to the calculation formula Get the deviation of the current mobile payment orders initiated by users in the target area

[0068] The deviation of the current mobile payment order initiated by the user in the target area is matched with the basic risk score corresponding to each deviation range stored in the database to obtain the basic risk score of the current mobile payment order initiated by the user in the target area.

[0069] Step 4: Perform a weighted fusion calculation on the basic risk score combined with the risk score environmental factor and the user's current personal risk score to generate a real-time comprehensive risk score for the current mobile payment order, and compare it with the preset risk threshold to determine whether it is a risk event.

[0070] As a preferred feasible embodiment, the specific process of generating the real-time comprehensive risk score of the current mobile payment order and comparing it with the preset risk threshold to determine whether it is a risk event includes: performing weighted fusion calculation on the basic risk score of the current mobile payment order initiated by the user in the target area, the risk score environmental factor of the target area, and the current personal risk score of the user in the target area according to the real-time comprehensive risk score standard calculation formula to generate the real-time comprehensive risk score of the current mobile payment order initiated by the user in the target area.

[0071] It should be further explained that the calculation formula for the real-time comprehensive risk score standard is specifically: ξ = α × JC + β × HJ + γ × GR, where JC, HJ, and GR are the basic risk score of the current mobile payment order initiated by the user in the target area, the risk score environment factor of the target area, and the current personal risk score of the user in the target area, respectively; α, β, and γ are the corresponding weights of the basic risk score, the risk score environment factor, and the user's current personal risk score, respectively.

[0072] A specific example, (1) α=0.5, β=0.35, γ=0.15.

[0073] Prioritize whether the current transaction behavior is consistent with user habits (such as amount deviation, device anomalies, etc.), and ensure rapid identification of real-time fraudulent behavior through high weighting. For example, if a user suddenly uses an unfamiliar device to make a large transfer, even if their historical risk score is low, risk control will be triggered due to the high basic risk score. It also reflects the overall risk level of the target area (such as the concentration of high-risk merchants and the activeness of abnormal devices). If a region has recently experienced frequent credit card fraud incidents (high environmental factors), even if the deviation of a single transaction is moderate, the weighted comprehensive score may still exceed the threshold, thereby improving the comprehensiveness of risk assessment.

[0074] This example is applicable to high-risk areas or high-risk business scenarios (such as cross-border payments and transactions with unfamiliar merchants). It is necessary to strictly control the real-time behavior of each transaction to prevent new fraud methods from bypassing historical data monitoring.

[0075] (2)α=0.4, β=0.2, γ=0.4.

[0076] Trusting the payment behavior of long-term low-risk users allows for a certain degree of transaction deviation. For example, if a user's historical score is consistently below 30 points (low risk), even if a transaction is conducted using an uncommon device, the overall score may still be below the threshold due to the high weight of the individual score, thus reducing disruption to high-quality users. This also prevents lags caused by relying solely on historical data. For example, if an established user suddenly engages in high-frequency abnormal transactions (a sudden increase in their basic risk score), even if their individual score history is good, the overall score will still trigger risk control, preventing "account farming" fraud.

[0077] This example is suitable for low-risk areas or mature user groups (such as long-term corporate users and individual users with good historical credit), using user historical performance as the core evaluation basis to improve payment convenience.

[0078] It should be noted that the corresponding weights of the basic risk score, risk score environmental factors and the user's current personal risk score are not fixed and can be adjusted according to specific needs, which will not be elaborated here.

[0079] If the real-time comprehensive risk score is greater than or equal to the preset risk threshold, the current mobile payment order initiated by the user in the target area is determined to be a risk event, and step five is executed; otherwise, the current mobile payment order initiated by the user in the target area is determined to be a non-risk event, and step one is executed.

[0080] The present invention performs weighted fusion calculation based on the basic risk score, risk score environmental factors and user's current personal risk score obtained through analysis to obtain a real-time comprehensive risk score for the current mobile payment order. Through multi-dimensional fusion analysis, the linkage effect of risk score environmental factors and user's current personal risk score is fully considered, avoiding the single dimension of mobile payment risk assessment and reducing the missed judgment rate of mobile payment transaction risks.

[0081] Step 5: If the current mobile payment order is a risk event, determine its risk level and automatically trigger risk control measures corresponding to the risk level.

[0082] As a preferred feasible embodiment, the specific content of determining the risk level of the current mobile payment order and automatically triggering the risk control measures corresponding to the risk level includes: determining the risk level of the current mobile payment order based on the real-time comprehensive risk score of the current mobile payment order.

[0083] If the risk level is low, secondary verification such as SMS verification code, biometrics, and security questions and answers will be triggered.

[0084] Specifically, the SMS verification code verification includes: sending a one-time verification code to the user's registered mobile phone, with a validity period of 1 minute and a failure limit of 3 times.

[0085] The biometric identification includes: face recognition, calling the device camera for liveness detection; fingerprint verification, reading the device fingerprint sensor data.

[0086] The security question and answer verification includes: preset question verification, personal questions set by the user (such as "first pet's name"); dynamic question generation, verification based on the user's transaction history (such as "last transaction amount yesterday").

[0087] If the risk level is medium, the transaction will be delayed and submitted for manual review. If the manual review is passed, the transaction will continue to be completed. Otherwise, the transaction will be canceled.

[0088] Specifically, the manual review includes automatic transfer to the risk control specialist review queue. The review is based on: user historical behavior patterns, device fingerprint matching, and geographic location rationality.

[0089] If the risk level is high, the transaction will be blocked and an automatic refund will be initiated for the completed transaction.

[0090] If the risk level is extremely high, a temporary freeze, credit limit reset, and password reset will be triggered, and the restrictions will be lifted after user verification.

[0091] As a preferred feasible embodiment, the specific content of determining the risk level of the current mobile payment order and automatically triggering risk control measures corresponding to the risk level also includes: while triggering the risk control measures corresponding to the risk level, based on the risk level of the current mobile payment order and the current personal risk score of the user in the target area, intelligently determine whether it is necessary to adjust the payment amount of the user in the target area and perform the operation.

[0092] Specifically, if the risk level of the current mobile payment order is low risk and the current personal risk score of the user in the target area is [0, 30], it is determined that the payment limit of the user in the target area needs to be adjusted, and the limit is increased.

[0093] If the risk level of the current mobile payment order is medium risk and the current personal risk score of the user in the target area is [0, 60], it is determined that there is no need to adjust the payment limit of the user in the target area, and the limit is maintained.

[0094] If the risk level of the current mobile payment order is high risk and the current personal risk score of the user in the target area is [0, 80], it is determined that the payment limit of the user in the target area needs to be adjusted, and the limit is reduced.

[0095] If the risk level of the current mobile payment order is extremely high and the current personal risk score of the user in the target area is [0, 100], it is determined that there is no need to adjust the payment limit of the user in the target area, and the limit is frozen.

[0096] The present invention determines whether the current mobile payment order is a risk event. If so, it determines its risk level and automatically triggers risk control measures corresponding to the risk level. It combines risk control measures with credit limit management to achieve optimal credit limit configuration under controllable risks, which is beneficial to ensuring user experience and security in abnormal transaction scenarios.

[0097] Example 2

[0098] A second aspect of the present invention provides an electronic device, characterized in that it includes a processor and a memory for storing execution instructions of the processor, wherein the processor is configured to execute the mobile payment method described in the present invention.

[0099] Example 3

[0100] A third aspect of the present invention provides a storage medium storing one or more programs, which can be executed by one or more processors to implement the steps in the mobile payment method of the present invention.

[0101] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. A mobile payment method, characterized in that: include: Step 1: Obtain user mobile payment transaction data for multiple target regions during the current monitoring period, and construct regional payment behavior profiles and individual user payment behavior profiles based on the data; Step 2: Dynamically calculate the risk score environment factor based on the regional payment behavior profile, dynamically calculate the user's current personal risk score based on the user's individual payment behavior profile, and match the benchmark payment limit according to the risk score-benchmark limit mapping table; Step 3: Capture in real time the transaction details of the current mobile payment order initiated by any user in any target area, and calculate the basic risk score of the current mobile payment order using the basic risk scoring rules; Step 4: The basic risk score is combined with the risk score environment factor and the user's current personal risk score for weighted fusion calculation to generate a real-time comprehensive risk score for the current mobile payment order, and the score is compared with the preset risk threshold to determine whether it is a risk event; Step 5: If the current mobile payment order is a risk event, determine its risk level and automatically trigger risk control measures corresponding to the risk level.

2. The mobile payment method according to claim 1, wherein: The specific process of constructing the regional payment behavior profile and the user's individual payment behavior profile includes: Extract the transaction amount, transaction timestamp, transaction GPS coordinates, transaction device, payment method, and operation behavior sequence of each transaction order of several users in multiple target areas during the current monitoring period from the user mobile payment transaction data, conduct statistics on the data, and determine the regular transaction amount range, active time period, commonly used GPS coordinates, commonly used devices, commonly used payment methods, and typical operation behavior sequence of the several users in the multiple target areas during the current monitoring period. These profiles are recorded as individual payment behavior profiles of the several users in the multiple target areas. Based on the individual payment behavior portraits of several users in multiple target areas, the amount distribution map, time pattern map, payment method distribution map and operation behavior sequence distribution map of multiple target areas are determined, and recorded as the regional payment behavior profiles of the multiple target areas.

3. The mobile payment method according to claim 1, wherein: The specific calculation process of the risk score environmental factor and the user's current personal risk score includes: Based on the regional payment behavior profiles and individual user payment behavior profiles of multiple target regions, the risk transaction ratio, risk transaction amount ratio, and abnormal device activity of the corresponding target regions are calculated, and the risk scoring environmental factors of the multiple target regions are obtained by weighted summing them according to the preset corresponding weights; Based on the individual payment behavior portraits of several users in multiple target areas, the risk transaction ratio, risk transaction amount ratio and abnormal device activity of the corresponding users in the corresponding target areas are calculated, and they are matched with the risk transaction ratios, risk transaction amount ratios and abnormal device activity corresponding risk scores of the users stored in the database, so as to obtain the risk transaction ratios, risk transaction amount ratios and abnormal device activity corresponding risk scores of the several users in the multiple target areas, and sum them up to obtain the current personal risk scores of the users in the multiple target areas.

4. The mobile payment method according to claim 3, wherein: The specific process of matching the benchmark payment amount according to the risk score-benchmark amount mapping table includes: The current personal risk scores of several users in multiple target areas are matched with the benchmark payment limits corresponding to the current personal risk score ranges of each user in the risk score-benchmark limit mapping table. If the current personal risk score of a user in a certain target area falls within the current personal risk score range of the user, the benchmark payment limit corresponding to the current personal risk score range of the user is used as the benchmark payment limit of the user in the target area, thereby obtaining the benchmark payment limits of several users in multiple target areas.

5. The mobile payment method according to claim 1, wherein: The specific contents of the basic risk scoring rules include: Extract the transaction amount, transaction timestamp, transaction GPS coordinates, and operation behavior sequence of the current mobile payment order initiated by users in the target area from the transaction details data. Compare these with the regular transaction amount range, active time period, common GPS coordinates, and typical operation behavior sequence in the user individual payment behavior profile of the corresponding user in the target area, and calculate the deviation degree of the current mobile payment order initiated by users in the target area. The deviation of the current mobile payment order initiated by the user in the target area is matched with the basic risk score corresponding to each deviation range stored in the database to obtain the basic risk score of the current mobile payment order initiated by the user in the target area.

6. The mobile payment method according to claim 1, wherein: The specific process of generating the real-time comprehensive risk score of the current mobile payment order and comparing it with the preset risk threshold to determine whether it is a risk event includes: The basic risk score of the current mobile payment order initiated by the user in the target area, the risk score environmental factor of the target area, and the current personal risk score of the user in the target area are weighted and integrated according to the real-time comprehensive risk score standard calculation formula to generate the real-time comprehensive risk score of the current mobile payment order initiated by the user in the target area; If the real-time comprehensive risk score is greater than or equal to the preset risk threshold, the current mobile payment order initiated by the user in the target area is determined to be a risk event, and step five is executed; otherwise, the current mobile payment order initiated by the user in the target area is determined to be a non-risk event, and step one is executed.

7. The mobile payment method according to claim 1, wherein: The specific contents of determining the risk level of the current mobile payment order and automatically triggering the risk control measures corresponding to the risk level include: Determine the risk level of the current mobile payment order based on the real-time comprehensive risk score; If the risk level is low, secondary verification such as SMS verification code, biometrics, and security questions and answers will be triggered; If the risk level is medium, the transaction will be delayed and submitted for manual review. If the manual review is passed, the transaction will continue to be completed; otherwise, the transaction will be canceled. If the risk level is high, the transaction will be blocked and an automatic refund will be issued for the completed transaction; If the risk level is extremely high, a temporary freeze, credit limit reset, and password reset will be triggered, and the restrictions will be lifted after user verification.

8. The mobile payment method according to claim 7, wherein: The specific contents of determining the risk level of the current mobile payment order and automatically triggering the risk control measures corresponding to the risk level also include: While triggering risk control measures corresponding to the risk level, based on the risk level of the current mobile payment order and the current personal risk score of the user in the target area, it is intelligently determined whether the payment limit of the user in the target area needs to be adjusted and the operation is performed.

9. An electronic device, characterized in that: The invention comprises a processor and a memory for storing execution instructions of the processor, wherein the processor is configured to execute the mobile payment method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the mobile payment method according to any one of claims 1 to 8.

Images