Method and device for adjusting fraud score threshold

Abstract

The invention provides a fraud score threshold adjustment method and device, and the method comprises the steps: obtaining a fraud score through an anti-fraud model, and carrying out the risk decision based on the fraud score and a rejection threshold in the anti-fraud model; calculating the proportion of fraudulent customers in the sample to obtain the black concentration of the sample; taking the reference black concentration corresponding to the rejection threshold value and the sample black concentration belonging to the same distribution as an original hypothesis, if the original hypothesis is not established, when the reference black concentration corresponding to the rejection threshold value is greater than the sample black concentration, modifying the rejection threshold value based on the lower limit of the current scoring range raised by one level, otherwise, the rejection threshold is modified based on lowering the lower limit of the current scoring range by one level until the original hypothesis is established. In the scheme, the sample black concentration in the adjustment period is calculated, the reference black concentration and the sample black concentration belong to the same distribution as the original hypothesis, and if the original hypothesis is not established, the rejection threshold value is adjusted until the original hypothesis is established, so that the purposes of ensuring that the rejection rate of the anti-fraud model is stable and avoiding rejecting a large number of normal customers are achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and device for adjusting a fraud scoring threshold. Background Art

[0002] Anti-fraud identification and credit rating are usually required in the pre-loan approval process of auto finance. Although both rely on algorithm models, the actual groups they face are completely different. Generally speaking, more than 99.9% of loan applicants received by financial institutions are normal customers. By obtaining basic customer data, credit data, historical performance data, etc., statistical modeling is used to obtain a score card, and the customer's credit rating is sorted from high to low. The lower the credit score, the higher the probability of default. Therefore, financial institutions can dynamically adjust the credit score threshold as needed. For normal loan customers, customer quality is always stable, while the very small number of fraudulent customers, which account for less than 0.1% of the total, have serious sample imbalance problems. The credit rating model cannot identify and give a low score. Therefore, the core reason why the anti-fraud identification technology system is different from the credit rating is that the customer groups they face are naturally different.

[0003] After the anti-fraud model and credit rating model are launched, the Population Stability Index (PSI) is usually used to monitor whether the model score distribution and feature distribution are stable. When the model's indicators are abnormal, it means that the model's performance is declining, and the model needs to be retrained and replaced in time. In the automotive finance scenario, the fraud ratio usually does not exceed 1:1000, the sources of fraudulent customers are wide, and there are strong deviations in data characteristics, which leads to the instability of the anti-fraud model. Model overfitting often triggers certain feature values, causing the model to judge the customer's fraud risk to be high and frequently make risky decisions of rejection. Through post-loan AB testing, it can be found that the real fraud probability of a large number of suspected fraudulent customers hit in a short period of time will not increase significantly, which indirectly shows that the anti-fraud model often overfits, and the disadvantage of monitoring based on the PSI indicator is lag. Therefore, models with overfitting risks may often intercept customers excessively in a short period of time, causing business damage. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for adjusting a fraud scoring threshold, so as to ensure a stable rejection rate of an anti-fraud model and avoid rejecting a large number of normal customers.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] A first aspect of an embodiment of the present invention discloses a method for adjusting a fraud scoring threshold, the method comprising:

[0007] The anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, the current score range is one of the multiple score ranges included in the preset score mapping table, each score range corresponds to a reference black density, and the reference black density is the proportion of real fraudulent customers among customers with the fraud score within the corresponding score range in the past period;

[0008] Whenever the number of customers who have completed the risk decision reaches a preset number, the proportion of customers determined to be fraudulent customers among the preset number of customers is calculated to obtain the sample black concentration;

[0009] Based on the hypothesis testing method, taking the reference black density corresponding to the rejection threshold and the sample black density as belonging to the same distribution as the original hypothesis, and judging whether the original hypothesis is established;

[0010] If the original hypothesis is not true, then when the benchmark black concentration corresponding to the rejection threshold is greater than the sample black concentration, the current scoring range is increased by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the increase; when the benchmark black concentration corresponding to the rejection threshold is less than the sample black concentration, the current scoring range is reduced by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the reduction; return to execute the hypothesis testing method, take the benchmark black concentration corresponding to the rejection threshold and the sample black concentration belong to the same distribution as the original hypothesis, and judge whether the original hypothesis is true, until the original hypothesis is true.

[0011] Preferably, in the anti-fraud model, a corresponding general rejection amount is set for each score range. Accordingly, the anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model, including:

[0012] The anti-fraud model predicts the fraud scores of incoming customers;

[0013] determining whether the fraud score is greater than the rejection threshold;

[0014] If the fraud score is greater than the rejection threshold, querying from the multiple score ranges to obtain a target score range corresponding to the fraud score, and determining whether the ordinary rejection amount corresponding to the target score range is greater than 0;

[0015] If it is greater than 0, the customer is determined to be a fraudulent customer, and the ordinary rejection amount corresponding to the target score range is reduced by 1;

[0016] If it is not greater than 0, the customer is determined to be a normal customer;

[0017] If the fraud score is not greater than the rejection threshold, the customer is determined to be a normal customer.

[0018] Preferably, if in the anti-fraud model, a corresponding priority rejection amount is set for each score range, and the priority rejection amount is half of the general rejection amount, then before determining whether the general rejection amount corresponding to the target score range is greater than 0, the method further includes:

[0019] If the customer is a mass-sending risk customer, determining whether the priority rejection amount corresponding to the target score range is greater than 0;

[0020] If it is greater than 0, the customer is determined to be a fraudulent customer, and the priority rejection amount corresponding to the target score range is reduced by 1;

[0021] If it is not greater than 0, the step of determining whether the ordinary rejection amount corresponding to the target score range is greater than 0 is performed.

[0022] Preferably, after modifying the rejection threshold, the method further comprises:

[0023] The ordinary rejection amount is adjusted to the lower limit of the score range corresponding to the modified rejection threshold, and the priority rejection amount is adjusted to half of the ordinary rejection amount.

[0024] Preferably, after determining that the customer is a fraudulent customer, the method further includes:

[0025] A fraud tag is added to the loan information of the customer.

[0026] A second aspect of an embodiment of the present invention discloses a device for adjusting a fraud scoring threshold, the device comprising:

[0027] A risk decision unit, in which the anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and a rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, the current score range is one of the multiple score ranges included in a preset score mapping table, each score range corresponds to a reference black density, and the reference black density is the proportion of real fraudulent customers among customers with the fraud score within the corresponding score range in the past period;

[0028] A calculation unit, configured to calculate the proportion of fraudulent customers among the preset number of customers to obtain a sample black concentration whenever the number of customers who have completed the risk decision reaches a preset number;

[0029] A testing unit, configured to determine whether the original hypothesis is established by taking the reference black density and the sample black density corresponding to the rejection threshold as belonging to the same distribution based on a hypothesis testing method;

[0030] An adjustment unit is used for, if the original hypothesis is not established, then when the reference black concentration corresponding to the rejection threshold is greater than the sample black concentration, increasing the current scoring range by one level, and modifying the rejection threshold based on the lower limit of the current scoring range after the increase by one level; when the reference black concentration corresponding to the rejection threshold is less than the sample black concentration, decreasing the current scoring range by one level, and modifying the rejection threshold based on the lower limit of the current scoring range after the decrease by one level, returning to execute the hypothesis testing method, taking the reference black concentration corresponding to the rejection threshold and the sample black concentration belonging to the same distribution as the original hypothesis, and judging whether the original hypothesis is established, until the original hypothesis is established.

[0031] Preferably, in the anti-fraud model, a corresponding general rejection amount is set for each scoring range, and accordingly, the risk decision unit includes:

[0032] Scoring unit: the anti-fraud model predicts the fraud scores of incoming customers;

[0033] A first judging unit, configured to judge whether the fraud score is greater than the rejection threshold;

[0034] A second judgment unit is configured to query and obtain a target score range corresponding to the fraud score from the multiple score ranges if the fraud score is greater than the rejection threshold, and to judge whether the ordinary rejection amount corresponding to the target score range is greater than 0;

[0035] A determination unit is used to determine that the customer is a fraudulent customer if the score is greater than 0, and reduce the ordinary rejection amount corresponding to the target score range by 1; if the score is not greater than 0, determine that the customer is a normal customer; if the fraud score is not greater than the rejection threshold, determine that the customer is a normal customer.

[0036] Preferably, if in the anti-fraud model, a corresponding priority rejection amount is set for each scoring range, and the priority rejection amount is half of the ordinary rejection amount, then the second judgment unit is further used to:

[0037] Before determining whether the ordinary rejection amount corresponding to the target score range is greater than 0, if the customer is a mass risk customer, determine whether the priority rejection amount corresponding to the target score range is greater than 0; if it is greater than 0, determine that the customer is a fraudulent customer, and reduce the priority rejection amount corresponding to the target score range by 1; if it is not greater than 0, execute the step of determining whether the ordinary rejection amount corresponding to the target score range is greater than 0.

[0038] Preferably, the device further comprises:

[0039] The quota adjustment unit resets the common rejection quota corresponding to each of the scoring ranges after modifying the rejection threshold, and adjusts the priority rejection quota to half of the common rejection quota.

[0040] Preferably, the device further comprises:

[0041] The marking unit is used to add a fraud label to the loan information of the customer after determining that the customer is a fraud customer.

[0042] Based on the method and device for adjusting the fraud score threshold provided by the above-mentioned embodiment of the present invention, the anti-fraud model predicts the fraud score of the incoming customer, and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, and the current score range is one of the multiple score ranges contained in the preset score mapping table, each of the score ranges corresponds to a benchmark black density, and the benchmark black density is the proportion of real fraudulent customers among the customers whose fraud scores are within the corresponding score range in the past period; whenever the number of customers who have completed the risk decision reaches a preset number, the proportion of customers determined to be fraudulent customers among the preset number of customers is calculated to obtain the sample black density; based on the hypothesis testing method, the rejection threshold corresponding to the rejection threshold is converted into a sample black density. The reference black density and the sample black density belong to the same distribution as the original hypothesis, and judge whether the original hypothesis is established; if the original hypothesis is not established, when the reference black density corresponding to the rejection threshold is greater than the sample black density, the current scoring range is increased by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the increase by one level; when the reference black density corresponding to the rejection threshold is less than the sample black density, the current scoring range is reduced by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the reduction by one level, and the step of returning to execute the hypothesis testing method, taking the reference black density and the sample black density corresponding to the rejection threshold as the original hypothesis, and judging whether the original hypothesis is established, until the original hypothesis is established. In this scheme, a dynamic adjustment period is set, and the sample black density within the adjustment period is calculated. Based on the hypothesis testing method, the reference black density and the sample black density corresponding to the rejection threshold are taken as the original hypothesis that they belong to the same distribution. When the original hypothesis is not established, the rejection threshold is adjusted until the original hypothesis is established, so as to achieve the purpose of ensuring the stability of the rejection rate of the anti-fraud model and avoiding the rejection of a large number of normal customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0044] Figure 1 A flowchart of a method for adjusting a fraud scoring threshold value disclosed in an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a scoring mapping table disclosed in an embodiment of the present invention;

[0046] Figure 3A flowchart of another method for adjusting a fraud scoring threshold value disclosed in an embodiment of the present invention;

[0047] Figure 4 A structural diagram of a device for adjusting a fraud scoring threshold disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0049] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0050] like Figure 1 As shown, it is a flowchart of a method for adjusting a fraud scoring threshold disclosed in an embodiment of the present invention, so as to solve the problem that a large number of customers hit the anti-fraud model in a short period of time, a large number of normal customers are rejected, and the business is damaged. The method mainly includes the following steps:

[0051] Step S101: The anti-fraud model predicts a fraud score for the incoming customer and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model.

[0052] Among them, the rejection threshold is the lower limit of the current scoring range, the current scoring range is one of the multiple scoring ranges contained in the preset scoring mapping table, each scoring range corresponds to a benchmark black concentration, and the benchmark black concentration is the proportion of real fraudulent customers among customers with fraud scores within the corresponding scoring range in the past period.

[0053] Among them, real fraud customers refer to customers who have been confirmed to be fraud customers and have been labeled as fraudulent.

[0054] In step S101, the customer applies for a car loan through an App or mini program on a mobile device, and communicates with the server through the network. The server collects the customer's information and processes it into the input feature input model of the anti-fraud model. The anti-fraud model feedbacks whether the customer's application is approved. The anti-fraud model can be any classification model or anomaly detection model with good performance, such as random forest, gradient boosting decision tree, and support vector machine. The fraud score is derived from the anti-fraud model, and the score increases with the probability of fraud.

[0055] like Figure 2 , which is a schematic diagram of a scoring mapping table disclosed in an embodiment of the present invention.

[0056] Exemplarily, the process of establishing a scoring mapping table is as follows:

[0057] Obtain 10,000 historical customer fraud risk scores and fraud labels in the past year, sort and classify customers according to the fraud scores, with a score range of [0,200] points, and 10 grades of 20 points each. The number of fraudulent customers / customers in each grade is the corresponding benchmark black density of the group. Assuming that there are 100 samples with scores in [120,140) and 5 fraudulent customers, the corresponding benchmark black density is 0.05, 30 samples with scores in [140,160) and 9 fraudulent customers, the corresponding benchmark black density is 0.3, 30 samples with scores in [160,180) and 12 fraudulent customers, the corresponding benchmark black density is 0.4, and 10 samples with scores in [180,200] and 5 fraudulent customers, the corresponding benchmark black density is 0.5.

[0058] It can be understood that the fraud label is the real confirmed fraud customers, and the benchmark black concentration = the number of real fraud customers / the number of historical customers.

[0059] In an embodiment of the present invention, the lower limit of the scoring range is selected to set the rejection threshold in the anti-fraud model. When the anti-fraud model is just launched and the rejection threshold has not yet been dynamically adjusted, the lower limit of the scoring range in which the baseline black concentration is the same as the target black concentration is selected as the rejection threshold based on the target black concentration set by the business side.

[0060] Assuming the target black density is 0.3, check the score mapping table and get the score range of the base black density of 0.3: [140, 160), then set 140 points as the rejection threshold. If the target black density falls between two score ranges, for example 0.35, the lower limit of the score range corresponding to the lower limit base black density is the rejection threshold, and the corresponding group is [140, 160), and set 140 points as the rejection threshold.

[0061] It should be noted that any expert rules, including risk decision rules, can be preset in the anti-fraud model. After the anti-fraud model obtains the fraud score of the customer, it uses the risk decision rules to determine whether the customer is a fraudulent customer or a group fraudulent customer, and outputs the corresponding judgment result. For the specific risk decision process, please refer to Figure 3 Corresponding embodiments of the present invention.

[0062] Step S102: Whenever the number of customers who have completed risk decisions reaches a preset number, the proportion of customers determined to be fraudulent customers among the preset number of customers is calculated to obtain the sample black concentration.

[0063] Here, customers whose fraud scores are greater than the rejection threshold are determined to be fraudulent customers. Sample black density = number of determined fraudulent customers / number of sample customers.

[0064] In step S102, a preset number is used as a dynamic adjustment period. The preset number may be 5000, which means that after risk decisions are made on 5000 customers using the anti-fraud model, the sample black concentration is calculated, and based on the sample black concentration, it is determined whether the rejection threshold in the anti-fraud model needs to be adjusted.

[0065] In one embodiment, time can also be used as a period, such as a day, a week, etc., to calculate the proportion of customers determined as fraudulent customers within the time period through anti-fraud model risk decision-making to obtain the sample black concentration.

[0066] Step S103: Based on the hypothesis testing method, the reference black density corresponding to the rejection threshold and the sample black density belong to the same distribution as the original hypothesis, and whether the original hypothesis is established is determined. If yes, step S104 is executed; if not, step S105 is executed.

[0067] In the specific implementation process of step S103, a hypothesis test is used to compare the significance level α (0.05) corresponding to the binomial distribution under the current rejection threshold (samples greater than the rejection threshold are 1, otherwise 0). A two-sided test is used here. When the statistic p is less than α, it is considered that the H0 hypothesis of the same distribution is not established, and the threshold needs to be adjusted.

[0068] It should be noted that hypothesis testing: first, an original hypothesis (H0) is set, which means that the baseline black concentration and the sample black concentration are the same, and it is necessary to test whether this original hypothesis is true.

[0069] Binomial distribution: In this test, the sample data will be classified according to the current rejection threshold: samples greater than the threshold are recorded as 1, and samples less than or equal to the threshold are recorded as 0. The samples formed in this way can be described by the binomial distribution.

[0070] Significance level α: The significance level set here is 0.05, which means that in 5% of cases, the original hypothesis may be wrongly rejected.

[0071] Two-sided test: means focusing on the difference between the reference black concentration and the sample black concentration, whether the reference black concentration is higher or lower than the sample black concentration.

[0072] p-value: If the calculated p-value is less than the significance level α (0.05), it means there is enough evidence to reject the original hypothesis. This indicates that there is a significant difference between the baseline black concentration and the sample black concentration.

[0073] Step S104: Determine not to adjust the rejection threshold.

[0074] Step S105: When the reference black concentration corresponding to the rejection threshold is greater than the sample black concentration, the current scoring range is increased by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the increase; when the reference black concentration corresponding to the rejection threshold is less than the sample black concentration, the current scoring range is reduced by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the reduction, and the process returns to step S103 until the original hypothesis is established.

[0075] In step S105, if the reference black density is greater than the sample black density, it indicates that the reference black density corresponding to the current rejection threshold is too low and the rejection threshold needs to be increased; otherwise, it indicates that the sample black density is too high and the rejection threshold needs to be lowered.

[0076] Assume that the current sample black density is 0.19, 140 points is the rejection threshold, adjust the current score range up by one level, and use the lower limit of the score range after the increase (160 points) as the rejection threshold, and then perform the hypothesis test again. If H0 is established, the adjustment is terminated, otherwise, the adjustment continues. Similarly, if the current sample black density is 0.41, 160 points is the rejection threshold, and the rejection threshold is lowered to 140 points. If H0 is established, the adjustment is terminated, otherwise, the adjustment continues.

[0077] For example, in actual application scenarios, after the anti-fraud model is launched, the model rejection rate may deviate, and the rejection rate needs to be adjusted dynamically to ensure business stability. For example, the historical benchmark [140,160) segment black concentration is 0.3, [160,180) segment black concentration is 0.4, and [180,200] segment black concentration is 0.5. If the newly launched anti-fraud model is abnormal, its ranking is poor after running for a period of time. For example, the [140,160] segment black concentration is 0.4, the [160,180) segment black concentration is 0.3, and the [180,200] segment black concentration is 0.2, indicating that the model has a large deviation and needs to be updated to ensure its sorting ability. Before updating the model, the model rejection threshold is adjusted through the method for adjusting the fraud scoring threshold disclosed in the embodiment of the present invention to ensure the stability of the overall rejection rate. In addition, when the newly launched anti-fraud model has run for a period of time and its recognition ability declines, it enters an underfitting state and the rejection rate decreases. The method for adjusting the fraud scoring threshold disclosed in the embodiment of the present invention can automatically lower the rejection threshold, so that the model can ensure the stability of the overall rejection rate.

[0078] For example, in actual application scenarios, when dealers need to promote a certain car due to inventory pressure and other reasons, in addition to the original channel customers, they will also attract a large number of customers from other channels in a short period of time, resulting in a surge in the number of applications in a certain period of time. The overall risk feature distribution of new channel customers is quite different from that of existing channel customers, resulting in a large number of misjudgments in the anti-fraud model. Among them, the rules that will lead to an increase in the hit rate of risk feature rules include: short-term concentrated increase in dealer volume, highly consistent financial plans, abnormal customer location sources, etc. Specific input variables are developed for the above features, and the model will frequently reject customers who hit the above rule sets. When the dealer triggers a large number of rejections in a short period of time, the sample black concentration is greater than the deviation difference of the benchmark black concentration at this time. The method for adjusting the fraud scoring threshold disclosed in the embodiment of the present invention will increase the rejection threshold to match the benchmark black concentration, thereby ensuring that normal customers pass. For example, a suspected fraud application has the following common combination features: the same salesperson + the same model + the same financial plan + buying a car in a different place. If the model outputs a fraud score of 150 points for such customers, and the current rejection threshold is set to 140 points, the application will be rejected. If a dealer in a certain place sets up a promotion and conducts marketing to clear out the inventory of cars for promotion, a large number of customers will apply in a short period of time. At this time, it is unreasonable to make risk decisions based on the rejection threshold. By dynamically adjusting the rejection threshold to 160 points, they will not be judged as high-risk customers, thus avoiding a large number of normal customers being intercepted in a short period of time, which will affect the development of business.

[0079] Based on the above-mentioned method for adjusting a fraud scoring threshold disclosed in an embodiment of the present invention, in this scheme, a dynamic adjustment period is set, the sample black concentration within the adjustment period is calculated, and based on a hypothesis testing method, the benchmark black concentration corresponding to the rejection threshold and the sample black concentration are taken as the original hypothesis that they belong to the same distribution. When the original hypothesis is not true, the rejection threshold is adjusted until the original hypothesis is true, so as to achieve the purpose of ensuring the stability of the rejection rate of the anti-fraud model and avoiding the rejection of a large number of normal customers.

[0080] To explain in detail the specific implementation process of step S101 in the above embodiment of the present invention, Figure 3 , which is a flow chart of another method for adjusting a fraud scoring threshold value disclosed in an embodiment of the present invention.

[0081] The specific implementation process of step S101 includes the following steps:

[0082] Step S301: The anti-fraud model predicts the fraud score of the incoming customer.

[0083] Step S302: Determine whether the fraud score is greater than the rejection threshold. If yes, execute step S304; if no, execute step S303.

[0084] Step S303: Determine whether the customer is a normal customer.

[0085] Step S304: query and obtain a target score range corresponding to the fraud score from multiple score ranges.

[0086] Step S305: Determine whether the customer is a mass-sending risk customer. If yes, execute step S306; if no, execute step S308.

[0087] Step S306: Determine whether the priority rejection amount corresponding to the target score range is greater than 0. If so, execute step S307; if not, execute step S308.

[0088] Step S307: Determine that the customer is a fraudulent customer, and reduce the priority rejection amount corresponding to the target score range by 1.

[0089] Step S308: Determine whether the ordinary rejection amount corresponding to the target score range is greater than 0. If so, execute step S309; ​​if not, execute step S303.

[0090] Step S309: Determine that the customer is a fraudulent customer, and reduce the normal rejection amount corresponding to the target score range by 1.

[0091] In an embodiment of the present invention, a normal queue and a priority queue are provided in the anti-fraud model. A customer who is not identified as a mass-messaging risk customer by the identification rules for mass-messaging risk customers preset in the anti-fraud model enters the normal queue, and is determined whether the customer is a fraudulent customer based on the customer's fraud score and the margin of the normal rejection amount set for each score range in the anti-fraud model.

[0092] For customers identified as mass risk by the identification rules for mass risk customers preset in the anti-fraud model, they will first enter the priority queue, and based on the customer's fraud score and the margin of the priority rejection amount set for each score range in the anti-fraud model, it will be determined whether the customer is a fraudulent customer. If the margin of the priority rejection amount is 0, the customer will enter the ordinary queue.

[0093] The ordinary rejection amount corresponding to each scoring range is set according to the benchmark black density corresponding to the scoring range. For example, there are 30 people in the scoring range [140,160) and the benchmark black density is 0.3, then the ordinary rejection amount of the scoring range is 30*0.3=9 people; there are 30 people in the scoring range [160,180) and the benchmark black density is 0.4, then the ordinary rejection amount of the scoring range is 30*0.4=12 people; there are 10 people in the scoring range [180,200] and the benchmark black density is 0.5, then the ordinary rejection amount of the scoring range is 10*0.5=5 people.

[0094] Preferably, the priority rejection amount corresponding to any scoring range is half of the ordinary rejection amount corresponding to the scoring range. If the ordinary rejection amount is an odd number, the larger value is rounded up. For example, if the ordinary rejection amount is 5, then the priority rejection amount is 3.

[0095] In one embodiment, after the rejection threshold is modified, the ordinary rejection amount corresponding to each scoring range is reset, and the priority rejection amount is adjusted to half of the ordinary rejection amount.

[0096] In the embodiment of the present invention, when the customer fraud score is higher than the rejection threshold and there is a remaining rejection amount in the score range corresponding to the customer fraud score, the request may be rejected, otherwise it may be released.

[0097] For example, when a customer enters the general queue, assuming the rejection threshold is 140 points, when the customer's fraud score is greater than 140 points, the customer is checked to see which score range the customer falls into. If there is a remaining general rejection quota corresponding to the score range, the customer is rejected (i.e., determined to be a fraudulent customer). If not, the customer is released (i.e., determined to be a normal customer). If the customer enters the priority queue, the priority queue quota is used first to perform rejection. After the priority queue quota is used up, the customer enters the general queue.

[0098] For example, in actual application scenarios, for suspected mass fraud cases, it is necessary to temporarily lower the fraud threshold to ensure that the current suspected fraud customers will not be passed. The recent increase in fraud cases for a certain model indicates that there are loopholes in the risk control that have been discovered by the black industry. Usually, they will be attacked in a short period of time, inducing mass cases. The analysis and summary of common risk characteristics include: buying a car in a different place (the place of registration, the place of purchase, and the place of permanent residence are inconsistent), too many recent credit records (multiple institutional inquiries or car loan application records), abnormal sales data (dealers concentrate on increasing volume in a short period of time, no promotional activities, the same salesperson), and obvious traces of unified packaging (highly consistent financial plans and high consistency of customer information). Develop specific model input variables for the above characteristics to reject customers who hit the above rule set. When a customer is determined to be a mass risk, the priority rejection quota is used first to execute the rejection. After the priority rejection quota is used up, the customer enters the ordinary queue. If there is still an ordinary rejection quota in the score range corresponding to the fraud score, it will remain rejected.

[0099] Based on the above disclosed method for adjusting the fraud score threshold value in the embodiment of the present invention, in this solution, when the customer's fraud score is higher than the rejection threshold value and the customer is not a mass-spreading risk customer, if there is a surplus of the ordinary rejection amount in the corresponding score range, it can be rejected, otherwise it will be released. If the customer is determined to be a mass-spreading risk, he will be given priority to enter the priority queue and use the priority rejection amount in the score range. If there is no priority rejection amount, he will enter the ordinary queue. While ensuring that customers with relatively high fraud risks are rejected first, it prevents mass-spreading risk events from causing the consumption of the ordinary rejection amount, which in turn causes ordinary suspected fraud customers to be released.

[0100] Based on the above-mentioned method for adjusting the fraud scoring threshold disclosed in the embodiment of the present invention, Figure 4 , which is a structural diagram of a device for adjusting a fraud scoring threshold value disclosed in an embodiment of the present invention, includes: a risk decision unit 401 , a calculation unit 402 , a verification unit 403 and an adjustment unit 404 .

[0101] The risk decision unit 401 is used to predict the fraud score of the incoming customer based on the anti-fraud model, and make risk decisions on the customer based on the fraud score and the rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, and the current score range is one of the multiple score ranges included in the preset score mapping table. Each score range corresponds to a benchmark black density, and the benchmark black density is the proportion of real fraudulent customers among customers with fraud scores within the corresponding score range in the past period.

[0102] The calculation unit 402 is used to calculate the proportion of fraudulent customers among the preset number of customers whenever the number of customers who have completed risk decisions reaches a preset number, and obtain the sample black concentration.

[0103] The testing unit 403 is used to determine whether the original hypothesis is established by taking the reference black density and the sample black density corresponding to the rejection threshold as belonging to the same distribution based on a hypothesis testing method.

[0104] The adjustment unit 404 is used to, if the original hypothesis is not established, increase the current score range by one level when the reference black concentration corresponding to the rejection threshold is greater than the sample black concentration, and modify the rejection threshold based on the lower limit of the current score range after the increase by one level; when the reference black concentration corresponding to the rejection threshold is less than the sample black concentration, reduce the current score range by one level, and modify the rejection threshold based on the lower limit of the current score range after the decrease by one level, and return to execute the hypothesis testing method, take the reference black concentration corresponding to the rejection threshold and the sample black concentration belonging to the same distribution as the original hypothesis, and judge whether the original hypothesis is established, until the original hypothesis is established.

[0105] In one embodiment, in the anti-fraud model, a corresponding general rejection amount is set for each scoring range, and accordingly, the risk decision unit includes:

[0106] Scoring unit: The anti-fraud model predicts the fraud score of the incoming customer.

[0107] The first judgment unit is used to judge whether the fraud score is greater than a rejection threshold.

[0108] The second judgment unit is used to query and obtain a target score range corresponding to the fraud score from multiple score ranges if the fraud score is greater than the rejection threshold, and to judge whether the ordinary rejection amount corresponding to the target score range is greater than 0.

[0109] The determination unit is used to determine that the customer is a fraudulent customer if the score is greater than 0, and reduce the ordinary rejection amount corresponding to the target score range by 1; if the score is not greater than 0, the customer is determined to be a normal customer; if the fraud score is not greater than the rejection threshold, the customer is determined to be a normal customer.

[0110] In one embodiment, if in the anti-fraud model, a corresponding priority rejection amount is set for each scoring range, and the priority rejection amount is half of the ordinary rejection amount, the second judgment unit is further used to:

[0111] Before determining whether the ordinary rejection amount corresponding to the target score range is greater than 0, if the customer is a mass risk customer, determine whether the priority rejection amount corresponding to the target score range is greater than 0; if it is greater than 0, the customer is determined to be a fraudulent customer, and the priority rejection amount corresponding to the target score range is reduced by 1; if it is not greater than 0, execute the step of determining whether the ordinary rejection amount corresponding to the target score range is greater than 0.

[0112] In one embodiment, the adjusting device further comprises:

[0113] The quota adjustment unit, after modifying the rejection threshold, resets the ordinary rejection quota for each scoring range and adjusts the priority rejection quota to half of the ordinary rejection quota.

[0114] In one embodiment, the adjusting device further comprises:

[0115] The marking unit is used to add a fraud label to the customer's loan information after determining that the customer is a fraud customer.

[0116] Based on the above-mentioned method for adjusting a fraud scoring threshold disclosed in an embodiment of the present invention, in this scheme, a dynamic adjustment period is set, the sample black concentration within the adjustment period is calculated, and based on a hypothesis testing method, the benchmark black concentration corresponding to the rejection threshold and the sample black concentration are taken as the original hypothesis that they belong to the same distribution. When the original hypothesis is not true, the rejection threshold is adjusted until the original hypothesis is true, so as to achieve the purpose of ensuring the stability of the rejection rate of the anti-fraud model and avoiding the rejection of a large number of normal customers.

[0117] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0118] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0119] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for adjusting a fraud scoring threshold, characterized in that: The method comprises: The anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, the current score range is one of the multiple score ranges included in the preset score mapping table, each score range corresponds to a reference black density, and the reference black density is the proportion of real fraudulent customers among customers with the fraud score within the corresponding score range in the past period; Whenever the number of customers who have completed the risk decision reaches a preset number, the proportion of customers determined to be fraudulent customers among the preset number of customers is calculated to obtain the sample black concentration; Based on the hypothesis testing method, taking the reference black density corresponding to the rejection threshold and the sample black density as belonging to the same distribution as the original hypothesis, and judging whether the original hypothesis is established; If the original hypothesis is not true, then when the benchmark black concentration corresponding to the rejection threshold is greater than the sample black concentration, the current scoring range is increased by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the increase; when the benchmark black concentration corresponding to the rejection threshold is less than the sample black concentration, the current scoring range is reduced by one level, and the rejection threshold is modified based on the lower limit of the current scoring range after the reduction; return to execute the hypothesis testing method, take the benchmark black concentration corresponding to the rejection threshold and the sample black concentration belong to the same distribution as the original hypothesis, and judge whether the original hypothesis is true, until the original hypothesis is true.

2. The method according to claim 1, characterized in that: In the anti-fraud model, a corresponding general rejection amount is set for each score range. Accordingly, the anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and the rejection threshold set in the anti-fraud model, including: The anti-fraud model predicts the fraud scores of incoming customers; determining whether the fraud score is greater than the rejection threshold; If the fraud score is greater than the rejection threshold, querying from the multiple score ranges to obtain a target score range corresponding to the fraud score, and determining whether the ordinary rejection amount corresponding to the target score range is greater than 0; If it is greater than 0, the customer is determined to be a fraudulent customer, and the ordinary rejection amount corresponding to the target score range is reduced by 1; If it is not greater than 0, the customer is determined to be a normal customer; If the fraud score is not greater than the rejection threshold, the customer is determined to be a normal customer.

3. The method according to claim 2, characterized in that If in the anti-fraud model, a corresponding priority rejection amount is set for each score range, and the priority rejection amount is half of the general rejection amount, then before determining whether the general rejection amount corresponding to the target score range is greater than 0, the method further includes: If the customer is a mass-sending risk customer, determining whether the priority rejection amount corresponding to the target score range is greater than 0; If it is greater than 0, the customer is determined to be a fraudulent customer, and the priority rejection amount corresponding to the target score range is reduced by 1; If it is not greater than 0, the step of determining whether the ordinary rejection amount corresponding to the target score range is greater than 0 is performed.

4. The method according to claim 3, characterized in that After modifying the rejection threshold, the method further comprises: The ordinary rejection amount corresponding to each scoring range is reset, and the priority rejection amount is adjusted to half of the ordinary rejection amount.

5. The method according to claim 3, characterized in that: After determining that the customer is a fraudulent customer, the method further includes: A fraud tag is added to the loan information of the customer.

6. A device for adjusting a fraud scoring threshold, characterized in that: The device comprises: A risk decision unit, in which the anti-fraud model predicts a fraud score for the incoming customer, and makes a risk decision for the customer based on the fraud score and a rejection threshold set in the anti-fraud model; the rejection threshold is the lower limit of the current score range, the current score range is one of the multiple score ranges included in a preset score mapping table, each score range corresponds to a reference black density, and the reference black density is the proportion of real fraudulent customers among customers with the fraud score within the corresponding score range in the past period; A calculation unit, configured to calculate the proportion of fraudulent customers among the preset number of customers to obtain a sample black concentration whenever the number of customers who have completed the risk decision reaches a preset number; A testing unit, configured to determine whether the original hypothesis is established by taking the reference black density and the sample black density corresponding to the rejection threshold as belonging to the same distribution based on a hypothesis testing method; An adjustment unit is used for, if the original hypothesis is not established, then when the reference black concentration corresponding to the rejection threshold is greater than the sample black concentration, increasing the current scoring range by one level, and modifying the rejection threshold based on the lower limit of the current scoring range after the increase by one level; when the reference black concentration corresponding to the rejection threshold is less than the sample black concentration, decreasing the current scoring range by one level, and modifying the rejection threshold based on the lower limit of the current scoring range after the decrease by one level, returning to execute the hypothesis testing method, taking the reference black concentration corresponding to the rejection threshold and the sample black concentration belonging to the same distribution as the original hypothesis, and judging whether the original hypothesis is established, until the original hypothesis is established.

7. The device according to claim 6, characterized in that In the anti-fraud model, a corresponding general rejection amount is set for each scoring range. Accordingly, the risk decision unit includes: Scoring unit: the anti-fraud model predicts the fraud scores of incoming customers; A first judging unit, configured to judge whether the fraud score is greater than the rejection threshold; A second judgment unit is configured to query and obtain a target score range corresponding to the fraud score from the multiple score ranges if the fraud score is greater than the rejection threshold, and to judge whether the ordinary rejection amount corresponding to the target score range is greater than 0; A determination unit is used to determine that the customer is a fraudulent customer if the score is greater than 0, and reduce the ordinary rejection amount corresponding to the target score range by 1; if the score is not greater than 0, determine that the customer is a normal customer; if the fraud score is not greater than the rejection threshold, determine that the customer is a normal customer.

8. The device according to claim 7, characterized in that If in the anti-fraud model, a corresponding priority rejection amount is set for each scoring range, and the priority rejection amount is half of the ordinary rejection amount, the second judgment unit is further used to: Before determining whether the ordinary rejection amount corresponding to the target score range is greater than 0, if the customer is a mass-sending risk customer, determine whether the priority rejection amount corresponding to the target score range is greater than 0; if it is greater than 0, determine that the customer is a fraudulent customer, and reduce the priority rejection amount corresponding to the target score range by 1; If it is not greater than 0, the step of determining whether the ordinary rejection amount corresponding to the target score range is greater than 0 is performed.

9. The device according to claim 8, characterized in that The device also includes: The quota adjustment unit resets the common rejection quota corresponding to each of the scoring ranges after modifying the rejection threshold, and adjusts the priority rejection quota to half of the common rejection quota.

10. The device according to claim 8, characterized in that The device also includes: The marking unit is used to add a fraud label to the loan information of the customer after determining that the customer is a fraud customer.

Images