Object risk level determination method and device, equipment, storage medium and product
By acquiring risk rules and data, dividing the target groups, and determining the accuracy of the hit and the risk level, the problem of inaccurate risk level determination in traditional methods is solved, and a more scientific and accurate risk assessment is achieved.
Patent Information
- Application Number
- CN202511253665.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-03
- Publication Date
- 2025-10-28
AI Technical Summary
Traditional methods that rely on expert experience to determine a client's risk level suffer from human bias and difficulty in linearly accumulating risk levels when multiple low-risk rules are met, leading to inaccurate risk level determination.
By acquiring risk data from multiple risk rules and objects, object groups are divided, the hit accuracy of each rule within a group is determined, and the higher of the object's highest risk level and the group's risk level is selected as the target risk level.
This enables more accurate determination of customer risk levels, reduces human bias, and improves the scientific rigor and accuracy of risk level determination.
Smart Images
Figure CN120852041A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of big data technology, and in particular to a method, apparatus, device, storage medium and program product for determining the risk level of an object. Background Technology
[0002] In order to respond to the requirements of "early identification, early exposure, early warning, and early handling" of transaction risks, financial institutions need to identify and warn about the risks of their customers.
[0003] In the past, risk levels for customer alerts were typically determined by expert experience, with different levels representing different levels of risk for each customer. However, this method has the following drawbacks: first, it is susceptible to bias from human experience; second, when a customer triggers multiple alert rules, their risk level is often difficult to determine, especially when multiple low-risk rules are triggered, as the risk level cannot be linearly accumulated. Therefore, the traditional method of relying on expert experience to determine customer risk levels cannot accurately identify a customer's risk level. Summary of the Invention
[0004] This application provides a method, apparatus, device, storage medium, and program product for determining the risk level of an object. It can accurately determine the customer's risk level through the process of determining and comparing group risk levels and the highest risk level.
[0005] In a first aspect, embodiments of this application provide a method for determining the risk level of an object, including:
[0006] Obtain risk data for multiple risk rules and multiple objects. For each object, determine at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to the matched risk rule; each risk rule is configured with a risk level.
[0007] Multiple objects are divided into multiple object groups. For each object group, the accuracy of each risk rule in each object group is determined based on the risk data of the objects in the object group and the hit risk rules.
[0008] For each object, based on the accuracy of the object's hit on at least one risk rule in each object group, determine the object's warning value in its respective object group and the group risk level corresponding to the warning value.
[0009] Choose the higher risk level from the group risk level and the highest risk level corresponding to the object as the target risk level for the object.
[0010] In one possible implementation, for each object, determining at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule, includes:
[0011] For each object, the object's risk data is compared with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule.
[0012] From the risk levels corresponding to at least one risk rule, select the highest risk level to obtain the highest risk level corresponding to the risk rule that the object's risk data matches.
[0013] In one possible implementation, multiple objects are divided into multiple object groups, including:
[0014] Get the object size type of each of the multiple objects;
[0015] Objects of the same size type are grouped together to obtain multiple object groups.
[0016] In one possible implementation, for each object group, the accuracy of each risk rule in each object group is determined based on the risk data of the objects in the object group and the hit risk rules, including:
[0017] For each object group, determine the first number of objects with abnormal records in the object group based on the risk data of the objects in the object group;
[0018] Based on the fact that each object in the object group has at least one risk rule matched, count the second number of objects matched by each risk rule in the object group.
[0019] For each risk rule, the ratio of the second quantity corresponding to the risk rule in each object group to the first quantity corresponding to the object group is used as the hit accuracy of the risk rule in each object group.
[0020] In one possible implementation, for each object, based on the accuracy of the hit of at least one risk rule hit in each object group, the warning value of the object in its respective object group and the group risk level corresponding to the warning value are determined, including:
[0021] For each object, based on the accuracy of the hit of at least one risk rule hit by the object in each object group, the accuracy of the hit of at least one risk rule hit by the object in the object group to which the object belongs is accumulated to obtain the warning value of the object in the object group to which the object belongs;
[0022] Based on the proportion of objects with abnormal records among multiple objects, and the hit accuracy of each risk rule in each object group, determine the low threshold and high threshold of the warning value;
[0023] Based on the low and high thresholds of the warning value, a risk level assessment range is generated.
[0024] Based on the warning value and risk level assessment range of the object, the group risk level corresponding to the warning value of the object is determined.
[0025] In one possible implementation, a risk level assessment range is generated based on a low warning threshold and a high warning threshold, including:
[0026] The range of values below the low threshold of the warning value will be considered as the low-risk level range.
[0027] The range of values greater than or equal to the lower threshold of the warning value and less than the upper threshold of the warning value is defined as the range of medium-risk levels.
[0028] The range of values greater than or equal to the high threshold of the warning value will be defined as the high-risk range.
[0029] The risk level assessment range is generated by combining the ranges of low risk, medium risk, and high risk.
[0030] Secondly, embodiments of this application provide an object risk level determination device, comprising:
[0031] The risk rule matching module is used to acquire risk data from multiple risk rules and multiple objects. For each object, it determines at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to the matched risk rule. Each risk rule is configured with a risk level.
[0032] The accuracy determination module is used to divide multiple objects into multiple object groups. For each object group, based on the risk data of the objects in the object group and the hit risk rules, the accuracy of each risk rule in each object group is determined.
[0033] The group risk level determination module is used to determine, for each object, the warning value of the object in its respective object group and the group risk level corresponding to the warning value, based on the accuracy of the object in hitting at least one risk rule in each object group.
[0034] The target risk level determination module is used to select the higher risk level from the group risk level and the highest risk level corresponding to the object as the target risk level of the object.
[0035] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;
[0036] The memory stores the instructions that the computer executes;
[0037] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.
[0038] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.
[0039] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.
[0040] The object risk level determination method, apparatus, device, storage medium, and program product provided in this application first acquire multiple risk rules and risk data of multiple objects. For each object, at least one risk rule matched by the object's risk data is determined, along with the highest risk level corresponding to the matched risk rule. That is, the highest risk level matched by the object under multiple risk rules can be determined through analysis between risk rules and risk data, for reference when subsequently determining the target risk level. Each risk rule is configured with a risk level. Further, multiple objects are divided into multiple object groups. For each object group, group analysis can be performed. Based on the risk data of objects in the object group and the matched risk rules, the accuracy of each risk rule in each object group is determined. Even further, for each object, based on the accuracy of the matched risk rule in each object group, the warning value of the object in its object group and the group risk level corresponding to the warning value can be determined for reference when subsequently determining the target risk level. Based on this, by combining the analysis of the hit rate between risk data and risk rules, and the grouping analysis of risk rule hit rates, the highest risk level and group risk level corresponding to the object are determined, and the higher level is selected as the target risk level for the object. Compared with the method of determining risk level based on human experience, the above-mentioned object risk level determination method can accurately determine the customer's early warning risk level. Attached Figure Description
[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0042] Figure 1 A schematic diagram illustrating the scenario for the object risk level determination method provided in this application;
[0043] Figure 2 A flowchart illustrating the method for determining the risk level of the object provided in this application;
[0044] Figure 3 A flowchart illustrating another method for determining the risk level of an object provided in this application;
[0045] Figure 4 A schematic diagram of the object risk level determination device provided in this application;
[0046] Figure 5 A structural schematic diagram of the device for determining the risk level of the object provided in this application.
[0047] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation
[0048] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0049] The object risk level determination method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, terminal 102 can communicate with server 104 via a network, or directly or indirectly via wired communication. Taking the object risk level determination method applied to server 104 as an example, the financial institution's server 104 can obtain risk data for multiple objects from multiple terminals 102 within the financial institution. Furthermore, server 104 can obtain multiple risk rules in response to configuration by the financial institution's management personnel. For each object, server 104 can determine at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule. Each risk rule is configured with a risk level. Further still, server 104 can divide multiple objects into multiple object groups. For each object group, based on the risk data of the objects in the object group and the matched risk rules, it determines the accuracy of each risk rule's hit in each object group. For each object, server 104 can determine the object's warning value in its respective object group and the group risk level corresponding to the warning value, based on the accuracy of the object's hit value for at least one risk rule in each object group. Based on this, server 104 can select the higher risk level as the target risk level for the object, considering both the highest risk level and the group risk level. Terminal 102 can be, but is not limited to, devices deployed by financial institutions in various regions, such as personal computers, laptops, smartphones, tablets, and IoT devices. Server 104 can be a physical server or a cloud server.
[0050] In one embodiment, a method for determining the risk level of an object is provided. This embodiment uses the application of this object risk level determination method to server 104 as an example for illustration. It can be understood that this object risk level determination method can also be applied to terminal 102, and can also be applied to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. Figure 2 As shown, the method for determining the risk level of this object includes:
[0051] Step 202: Obtain risk data for multiple risk rules and multiple objects. For each object, determine at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to the matched risk rule; each risk rule is configured with a risk level.
[0052] Specifically, risk data may include: (1) risk characteristics required by risk rules, such as the financial data, behavioral data, transaction data, credit information, etc. of the object; (2) the size of the object itself and the industry to which it belongs, used to determine the object size type of the object, as the basis for object grouping; (3) abnormal records of the object's defaults and overdue payments in the historical period.
[0053] Specifically, the risk rules can be defined as follows: when a risk data representation object exhibits a specific behavior, a certain risk rule is determined to be matched, and this risk rule is configured with a risk level (low, medium, high). For example, if the risk data representation object has inconsistent financial data, it can be determined that the object's risk data matches risk rule X, corresponding to a risk level of "high". The specific risk rules and the risk levels configured for the risk rules can be flexibly configured according to actual needs, and this embodiment does not limit them.
[0054] Optionally, the server can first obtain multiple preset risk rules and pre-stored risk data of multiple objects for subsequent big data analysis. For each object, the object's risk data is compared with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule.
[0055] For example, a hit list can be generated by aggregating the risk data of all objects, including at least one risk rule that each object hits, and the risk level of each hit risk rule. The hit list may specifically include the following:
[0056] Object A hits risk rule r01, risk level - low;
[0057] Object B hits risk rule r02, risk level - medium;
[0058] Object B hits risk rule r03, risk level - low;
[0059] Object C hits risk rule r03, risk level - high;
[0060] ………………
[0061] Step 204: Divide multiple objects into multiple object groups. For each object group, determine the accuracy of each risk rule in each object group based on the risk data of the objects in the object group and the hit risk rules.
[0062] Optionally, the server can group multiple objects according to preset grouping rules, resulting in multiple object groups. For each object group, the server can calculate the actual risk of the objects in the group based on the risk data of each object, and also calculate the predicted risk of each risk rule applied to each object group based on the risk rules matched by each object. Furthermore, the server can calculate and determine the accuracy of each risk rule in each object group by considering the actual risk of the objects in each group and the predicted risk of each risk rule applied to each object group.
[0063] Specifically, for object group 1, the number of objects in object group 1 that actually have abnormal records (which can be determined based on the risk data of each object in object group 1) can be used as the actual risk situation of the objects in object group 1. The number of objects in object group 1 that match risk rule r01 can be used as the predicted risk situation after risk rule r01 is applied in object group 1.
[0064] For example, taking a scenario with two object groups and risk rules r01 to r02, step 204 can be used to calculate the accuracy of each risk rule in each object group, as follows: (1) Accuracy of risk rule r01 in object group 1: r011; Accuracy of risk rule r01 in object group 2: r012. (2) Accuracy of risk rule r02 in object group 1: r021; Accuracy of risk rule r02 in object group 2: r022. As shown in Table 1:
[0065] Table 1. Examples of the accuracy of risk rules in each object group.
[0066] object group Risk Rules Accuracy 1 r01 <![CDATA[r011]]> 2 r01 <![CDATA[r012]]> 1 r02 <![CDATA[r021]]> 2 r02 <![CDATA[r022]]>
[0067] Step 206: For each object, based on the accuracy of the hit of at least one risk rule hit in each object group, determine the warning value of the object in its object group and the group risk level corresponding to the warning value.
[0068] Optionally, for each object, from the hit accuracy of at least one risk rule hit by the object in each object group, the hit accuracy of at least one risk rule hit by the object in the object group to which the object belongs is determined, thereby accumulating the hit accuracy of at least one risk rule hit by the object in the object group to which the object belongs, to obtain the warning value of the object in the object group. Further, the server can determine the group risk level corresponding to the warning value based on the risk level assessment interval used to evaluate the risk level represented by the warning value.
[0069] For example, taking object A as an example, if there are 3 object groups, and object A belongs to object group 1, and object A hits risk rule r01 and risk rule r02, and the accuracy of risk rule r01 in object group 1 is r011, the accuracy of risk rule r01 in object group 2 is r012, the accuracy of risk rule r01 in object group 3 is r013, the accuracy of risk rule r02 in object group 1 is r021, the accuracy of risk rule r02 in object group 2 is r022, and the accuracy of risk rule r02 in object group 3 is r023, then the warning value of object A = the sum of the accuracy of the risk rules hit by object A in its own object group 1 = r011 + r021.
[0070] Step 208: Select the higher risk level from the group risk level and the highest risk level corresponding to the object as the target risk level of the object.
[0071] The highest risk level is determined based on the matching between risk rules and the risk data of the target object. This rule-matching method directly determines the prediction accuracy of the risk rules applied to the target object. The group risk level is determined by conducting group analysis on multiple targets, analyzing the hit rate of risk rules across different target groups. This is a customer segmentation analysis method that determines the prediction accuracy of the risk rules applied to the target object within its target group based on the target object's group and the hit accuracy of the risk rules within that group. Comparing the highest risk level and the group risk level together helps to accurately determine the target object's risk level.
[0072] Optionally, the server can select the higher risk level from the group risk level corresponding to the object and the highest risk level as the target risk level for the object, and perform a risk assessment on the object based on the target risk level. Compared with the traditional method of relying on manual determination of the object's risk level, the above steps 202 to 208 can determine the object's risk level more systematically, scientifically, effectively, and accurately.
[0073] For example, the server can take different risk management measures based on the target risk level of the object. If the target risk level is low, it will continue to monitor; if the target risk level is medium or high, it may take measures such as contacting the object to increase collateral or reducing business operations.
[0074] The above-described method for determining the risk level of an object first acquires risk data for multiple risk rules and multiple objects. For each object, it determines at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to that risk rule. That is, by analyzing the relationship between risk rules and risk data, it first determines the highest risk level the object matches under multiple risk rules, which is then used as a reference when determining the target risk level. Each risk rule is configured with a risk level. Further, the multiple objects are divided into multiple object groups. For each object group, group analysis can be performed. Based on the risk data of the objects in the object group and the matched risk rules, the accuracy of each risk rule's hit rate in each object group is determined. Even further, for each object, based on the accuracy of the object's hit rate of at least one risk rule in each object group, the warning value of the object in its object group and the corresponding group risk level are determined, which are then used as a reference when determining the target risk level. Based on this, combining the hit analysis between risk data and risk rules, and the group analysis of risk rule hit rates, the highest risk level and group risk level corresponding to the object are determined, and the higher level is selected as the object's target risk level. Compared to relying on human experience to determine risk levels, the above-mentioned method for determining the risk level of an object can accurately determine the risk level of a customer.
[0075] In one embodiment, for each object, determining at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule, includes:
[0076] For each object, the object's risk data is compared with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule.
[0077] From the risk levels corresponding to at least one risk rule, select the highest risk level to obtain the highest risk level corresponding to the risk rule that the object's risk data matches.
[0078] Each risk rule is assigned a risk level; the higher the risk level, the greater the transaction risk of the represented object.
[0079] Optionally, for each object, the server can compare the object's risk data with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule. Further, the server can select the highest risk level from the risk levels corresponding to the at least one matched risk rule of the object's risk data to obtain the highest risk level corresponding to the risk rule matched by the object's risk data.
[0080] For example, if the risk data of object A matches the following risk rules: risk rule r01 (risk level - low), risk rule r02 (risk level - medium), risk rule r03 (risk level - medium), and risk rule r04 (risk level - high), then the server can determine that "risk level - high" is the highest risk level corresponding to the multiple risk rules that match the risk data of object A.
[0081] In this embodiment, for each object, the collision between the object's risk data and multiple risk rules can be used to determine at least one risk rule that the object's risk data matches, as well as the highest risk level of the match at least one risk rule, so as to accurately determine the target risk level of the object in the future.
[0082] In one embodiment, dividing multiple objects into multiple object groups includes:
[0083] Get the object size type of each of the multiple objects;
[0084] Objects of the same size type are grouped together to obtain multiple object groups.
[0085] Among them, the object size type can be specifically divided into: large, medium and small in a certain industry.
[0086] Optionally, the server can obtain the object size type of each of the multiple objects from the stored data, and group the multiple objects according to their respective object size types, grouping those belonging to the same object size type into the same group, thus obtaining multiple object groups.
[0087] For example, the server can divide multiple objects into three object groups according to their size type (large, medium, and small), that is, to achieve the grouping of multiple objects. Among them, object group 1 contains objects of the large size type, object group 2 contains objects of the medium size type, and object group 3 contains objects of the small size type.
[0088] In the above embodiments, multiple objects can be grouped / clustered according to their object size type, so that grouped backtesting (clustered prediction) can be performed based on the divided object groups, and the hit accuracy of each risk rule in each object group can be calculated separately.
[0089] In one possible implementation, for each object group, based on the risk data of the objects in the object group and the hit risk rules, the hit accuracy of each risk rule in each object group is determined, including:
[0090] For each object group, determine the first number of objects with abnormal records in the object group based on the risk data of the objects in the object group;
[0091] Based on the fact that each object in the object group has at least one risk rule matched, count the second number of objects matched by each risk rule in the object group.
[0092] For each risk rule, the ratio of the second quantity corresponding to the risk rule in each object group to the first quantity corresponding to the object group is used as the hit accuracy of the risk rule in each object group.
[0093] Optionally, for each object group, the server can calculate the actual risk situation (the actual number of objects with abnormal records) of the objects in the object group based on the risk data of the objects in the object group; that is, determine the first number of objects with abnormal records in the object group. Simultaneously, for each object group, the server can also calculate the predicted risk situation after applying each risk rule to each object group, based on at least one risk rule matched by each object in the object group; that is, calculate the second number of objects matched by each risk rule in each object group. Further, for each risk rule, the server can use the ratio of the second number corresponding to the risk rule in each object group to the first number of the object group itself as the hit accuracy of the risk rule in each object group.
[0094] For example, taking object groups (large, medium, and small) and risk rules (r01, r02) as an example, for each object group, the server can count the first number of objects with abnormal records in the object group based on the risk data of multiple objects in the object group, and count the second number of objects in each object group (large, medium, and small) that each risk rule (r01, r02) hits based on at least one risk rule that each object in the object group hits. For each risk rule, the ratio of the second number of objects hit by the risk rule to the first number of objects hit by the risk rule is used as the hit accuracy of the risk rule in each object group. For example, (1) the hit precision of risk rule r01 in the object group "Large" = (the second number of objects hit by r01 in "Large") / (the first number of objects with abnormal records in "Large"); (2) the hit precision of risk rule r01 in the object group "Medium" = (the second number of objects hit by r01 in "Medium") / (the first number of objects with abnormal records in "Medium"); (3) the hit precision of risk rule r02 in the object group "Large" = (the second number of objects hit by r02 in "Large") / (the first number of objects with abnormal records in "Large"). And so on, determine the hit precision of each risk rule r01 and r02 in the object groups "Large", "Medium", and "Small".
[0095] In one embodiment, an example table of hit accuracy for risk rules in large, medium, and small object groups is shown in Table 2:
[0096] Table 2 shows examples of the hit accuracy of risk rules for large, medium, and small object groups.
[0097] object group Risk Rules Accuracy Large r01 0.00538 medium r01 0.00234 Small r01 0.01140 Large r02 0.00157 medium r02 0.02315 Small r02 0.00892
[0098] In the above embodiments, backtesting can be performed by dividing the target groups (customer groups) and calculating the hit accuracy of each risk rule in each target group.
[0099] In one embodiment, for each object, based on the accuracy of the hit of at least one risk rule hit in each object group, the warning value of the object in its respective object group and the group risk level corresponding to the warning value are determined, including:
[0100] For each object, based on the accuracy of the hit of at least one risk rule hit by the object in each object group, the accuracy of the hit of at least one risk rule hit by the object in the object group to which the object belongs is accumulated to obtain the warning value of the object in the object group to which the object belongs;
[0101] Based on the proportion of objects with abnormal records among multiple objects, and the hit accuracy of each risk rule in each object group, determine the low threshold and high threshold of the warning value;
[0102] Based on the low and high thresholds of the warning value, a risk level assessment range is generated.
[0103] Based on the warning value and risk level assessment range of the object, the group risk level corresponding to the warning value of the object is determined.
[0104] Optionally, for each object, the server can determine the accuracy of at least one risk rule hit by the object in the object group within each object group from the accuracy of the hit of at least one risk rule hit by the object. By accumulating the accuracy of the hit of at least one risk rule hit by the object in the object group, the server obtains the warning value for the object in its object group. For example, given object groups (large, medium, small) and risk rules (r01, r02), if object A belongs to the "large" object group, and object A's risk data hits risk rules r01 and r02, with risk rule r01 having an accuracy of 0.00538 in "large" and risk rule r02 having an accuracy of 0.00157, then the warning value for object A = the accuracy of risk rule r01 hit by object A in "large" (0.00538) + the accuracy of risk rule r02 hit by object A in "large" (0.00157).
[0105] Optionally, the server can determine the low and high thresholds for alert values based on the proportion of objects with anomalous records among multiple objects and the hit accuracy of each risk rule in each object group. Specifically, the hit accuracy of each risk rule in each object group can be aggregated and divided into multiple tiers from largest to smallest, such as 20 tiers from 0 to the maximum value. These tiers are then sorted from smallest to largest, and the improvement in hit accuracy within each tier is calculated sequentially. Improvement = Hit accuracy within tier / Proportion of objects with anomalous records among multiple objects. Furthermore, the hit accuracy that first increases the improvement to greater than 1 can be used as the low threshold for alert values, and the hit accuracy that first increases the improvement to greater than 10 can be used as the high threshold for alert values.
[0106] In one embodiment, the range of risk levels below the lower threshold of the warning value can be defined as the low-risk range; the range of risk levels above or equal to the lower threshold and below the higher threshold of the warning value can be defined as the medium-risk range; and the range of risk levels above or equal to the higher threshold of the warning value can be defined as the high-risk range. The low-risk, medium-risk, and high-risk ranges are then combined to generate a risk level assessment range for evaluating the risk level of the warning value.
[0107] Optionally, the group risk level corresponding to the object's warning value can be determined based on the risk level assessment range in which the object's warning value falls. When the object's warning value falls within the low-risk range, the corresponding group risk level is "low"; when the object's warning value falls within the medium-risk range, the corresponding group risk level is "medium"; and when the object's warning value falls within the high-risk range, the corresponding group risk level is "high". The risk level of "high" is higher than "medium", and the risk level of "medium" is higher than "low".
[0108] In the above embodiments, the group risk level corresponding to the warning value of an object can be determined based on the risk level assessment range in which the warning value of the object falls, and can be used as a reference when assessing the target risk level of the object.
[0109] In one specific embodiment, such as Figure 3 The diagram shows a flowchart of another method for determining the risk level of an object, which mainly includes the following steps:
[0110] (1) Data collection. This includes: acquiring risk data for multiple risk rules and multiple objects.
[0111] (2) Risk rule collision. This includes: for each object, colliding the object's risk data with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule.
[0112] (3) Predict the highest risk level. This includes: selecting the highest risk level from the risk levels corresponding to at least one risk rule, and obtaining the highest risk level corresponding to the risk rule that the object's risk data matches.
[0113] (4) Customer segmentation prediction. This includes dividing multiple objects into multiple object groups, and for each object group, determining the accuracy of each risk rule in each object group based on the risk data of the objects in the object group and the hit risk rules.
[0114] (5) Predict the group risk level. For each object, based on the accuracy of the object in hitting at least one risk rule in each object group, determine the warning value of the object in its object group and the group risk level corresponding to the warning value.
[0115] (6) Determine the target risk level. For each object, select the higher level from the group risk level and the highest risk level corresponding to the object as the target risk level for the object.
[0116] (7) Take risk management measures. This includes taking corresponding risk management measures based on the target risk level of the object to improve the efficiency and accuracy of risk screening and enhance the level of risk prevention and control.
[0117] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0118] Based on the same inventive concept, this application also provides an object risk level determination device for implementing the object risk level determination method described above. The solution provided by this object risk level determination device is similar to the solution described in the state determination method above. Therefore, the specific limitations in one or more device embodiments provided below can be found in the limitations of the object risk level determination method described above, and will not be repeated here.
[0119] In one embodiment, such as Figure 4 As shown, an object risk level determination device is provided, comprising:
[0120] The risk rule matching module 402 is used to acquire risk data of multiple risk rules and multiple objects. For each object, it determines at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to the matched risk rule. Each risk rule is configured with a risk level.
[0121] The accuracy determination module 404 is used to divide multiple objects into multiple object groups, and for each object group, determine the accuracy of each risk rule in each object group based on the risk data of the objects in the object group and the hit risk rules.
[0122] The group risk level determination module 406 is used to determine, for each object, the warning value of the object in its respective object group and the group risk level corresponding to the warning value, based on the accuracy of the object in hitting at least one risk rule in each object group.
[0123] The target risk level determination module 408 is used to select the higher level from the group risk level and the highest risk level corresponding to the object as the target risk level of the object.
[0124] The aforementioned object risk level determination device first acquires multiple risk rules and risk data for multiple objects. For each object, it determines at least one risk rule that the object's risk data matches, as well as the highest risk level corresponding to that risk rule. That is, by analyzing the relationship between risk rules and risk data, it first determines the highest risk level the object matches under multiple risk rules, which is used as a reference when subsequently determining the target risk level. Each risk rule is configured with a risk level. Further, the multiple objects are divided into multiple object groups. For each object group, group analysis can be performed. Based on the risk data of the objects in the object group and the matched risk rules, the accuracy of each risk rule's hit rate in each object group is determined. Even further, for each object, based on the accuracy of the hit rate of at least one risk rule matched by the object in each object group, the object's warning value and the corresponding group risk level are determined, used as a reference when subsequently determining the target risk level. Based on this, combining the hit analysis between risk data and risk rules, and the group analysis of risk rule hit rates, the highest risk level and group risk level corresponding to the object are determined, and the higher level is selected as the object's target risk level. Compared to relying on human experience to determine risk levels, the above-mentioned method for determining the risk level of an object can accurately determine the risk level of a customer.
[0125] In one possible implementation, the risk rule matching module is specifically used to: for each object, match the object's risk data with multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule; select the highest risk level from the risk levels corresponding to each of the at least one risk rule to obtain the highest risk level corresponding to the risk rule that the object's risk data matches.
[0126] In one possible implementation, the hit accuracy determination module is specifically used to: obtain the object size type of each of the multiple objects; and group objects of the same object size type into the same group to obtain multiple object groups.
[0127] In one possible implementation, the hit accuracy determination module can also be used to: for each object group, determine a first number of objects with abnormal records in the object group based on the risk data of the objects in the object group; based on at least one risk rule that each object in the object group hits, count a second number of objects that each risk rule hits in the object group; and for each risk rule, use the ratio of the second number corresponding to the risk rule in each object group to the first number corresponding to the object group as the hit accuracy of the risk rule in each object group.
[0128] In one possible implementation, the group risk level determination module is specifically used to: for each object, based on the accuracy of the hit of at least one risk rule hit by the object in each object group, accumulate the accuracy of the hit of at least one risk rule hit by the object in the object group to which the object belongs, to obtain the warning value of the object in the object group to which it belongs; determine the low threshold and high threshold of the warning value based on the proportion of objects with abnormal records among multiple objects and the accuracy of the hit of each risk rule in each object group; generate a risk level assessment interval based on the low threshold and high threshold of the warning value; and determine the group risk level corresponding to the warning value of the object based on the warning value of the object and the risk level assessment interval.
[0129] In one possible implementation, the group risk level determination module can also be used to: define the range of intervals less than the lower threshold of the warning value as the low-risk level range; define the range of intervals greater than or equal to the lower threshold of the warning value and less than the higher threshold of the warning value as the medium-risk level range; define the range of intervals greater than or equal to the higher threshold of the warning value as the high-risk level range; and aggregate the low-risk level range, medium-risk level range, and high-risk level range to generate a risk level assessment range.
[0130] Each module in the above-mentioned device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.
[0131] Figure 5 A structural diagram of the equipment used to determine the risk level of the object provided in this application. (For example...) Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus.
[0132] In the specific implementation process, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to execute the above-mentioned object risk level determination method.
[0133] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.
[0134] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.
[0135] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.
[0136] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.
[0137] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.
[0138] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.
[0139] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.
[0140] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.
[0141] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.
[0142] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0143] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0144] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0145] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.
[0146] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.
Claims
1. A method for determining the risk level of an object, characterized in that, include: Obtain risk data for multiple risk rules and multiple objects; for each object, determine at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule; Each of the aforementioned risk rules is configured with a risk level; The multiple objects are divided into multiple object groups. For each object group, the hit accuracy of each risk rule in each object group is determined based on the risk data of the objects in the object group and the hit risk rules. For each object, based on the accuracy of the hit of at least one risk rule hit by the object in each object group, the warning value of the object in its object group and the group risk level corresponding to the warning value are determined. From the group risk level corresponding to the object and the highest risk level, select the higher level as the target risk level for the object.
2. The method according to claim 1, characterized in that, For each of the objects, determining at least one risk rule that the object's risk data matches, and the highest risk level corresponding to the matched risk rule, includes: For each object, the risk data of the object is compared with the multiple risk rules to determine at least one risk rule that the object's risk data matches, and the risk level corresponding to each of the at least one matched risk rule; From the risk levels corresponding to each of the at least one risk rule, the highest risk level is selected to obtain the highest risk level corresponding to the risk rule that the risk data of the object hits.
3. The method according to claim 1, characterized in that, The step of dividing the plurality of objects into multiple object groups includes: Obtain the object size type of each of the multiple objects; Objects of the same size type are grouped together to obtain multiple object groups.
4. The method according to claim 1, characterized in that, For each object group, determining the accuracy of each risk rule in each object group based on the risk data of the objects in the object group and the hit risk rules includes: For each object group, a first number of objects with abnormal records in the object group is determined based on the risk data of the objects in the object group; Based on at least one risk rule that is matched by each object in the object group, count the second number of objects matched by each risk rule in the object group. For each of the aforementioned risk rules, the ratio of the second quantity of the risk rule corresponding to each object group to the first quantity corresponding to the object group is used as the hit accuracy of the risk rule in each object group.
5. The method according to claim 1, characterized in that, For each object, determining the warning value of the object in its respective object group and the group risk level corresponding to the warning value based on the accuracy of the hit of at least one risk rule hit by the object in each object group includes: For each object, based on the accuracy of the hit of at least one risk rule hit by the object in each object group, the accuracy of the hit of at least one risk rule hit by the object in the object group to which the object belongs is accumulated to obtain the warning value of the object in the object group to which the object belongs; Based on the proportion of objects with abnormal records among the multiple objects, and the hit accuracy of each risk rule in each object group, determine the low threshold and high threshold of the warning value; Based on the low threshold and the high threshold of the warning value, a risk level assessment range is generated; Based on the warning value of the object and the risk level assessment range, the group risk level corresponding to the warning value of the object is determined.
6. The method according to claim 5, characterized in that, The step of generating a risk level assessment range based on the low threshold and the high threshold of the warning value includes: The range of values below the warning threshold is defined as the low-risk level range. The range of values greater than or equal to the lower threshold of the warning value and less than the higher threshold of the warning value is defined as the range of medium-risk levels. The range of values greater than or equal to the high threshold of the warning value shall be defined as the high-risk level range. The low-risk level range, the medium-risk level range, and the high-risk level range are combined to generate a risk level assessment range.
7. A device for determining the risk level of an object, characterized in that, include: The risk rule matching module is used to acquire risk data of multiple risk rules and multiple objects, and for each object, determine at least one risk rule that the risk data of the object matches, as well as the highest risk level corresponding to the matched risk rule; Each of the aforementioned risk rules is configured with a risk level; The accuracy determination module is used to divide the multiple objects into multiple object groups, and for each object group, determine the accuracy of each risk rule in each object group based on the risk data of the objects in the object group and the hit risk rules. The group risk level determination module is used to determine, for each object, the warning value of the object in its respective object group and the group risk level corresponding to the warning value, based on the accuracy of the object in hitting at least one risk rule in each object group. The target risk level determination module is used to select the higher risk level from the group risk level and the highest risk level corresponding to the object as the target risk level of the object.
8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.
10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.