Multi-head loan risk assessment method, electronic equipment and storage medium

Abstract

The invention provides a multi-head loan risk assessment method, electronic equipment and a storage medium, and the method comprises the steps: obtaining a customer group type where a target object is located and multi-head loan query information of the target object in a historical time window, obtaining risk assessment information of the target object according to the customer group type and the multi-head loan query information, and according to the risk assessment information, carrying out loan risk assessment on the target object. Therefore, the debit and credit risk assessment is performed in combination with the customer group type and the multi-head debit and credit query information, and the risk assessment accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to a multi-loan risk assessment method, an electronic device and a storage medium. BACKGROUND

[0002] In the credit field, multi-loan, i.e. a borrower applying for loans at multiple institutions within a short period of time, is the core inducement leading to the outbreak of credit risk.

[0003] In related technologies, a traditional risk control model is usually based on multi-loan features, i.e. counting the number of credit inquiries or the number of loan applications within a fixed time window (such as 30 days) to perform risk assessment.

[0004] However, the above risk assessment method does not consider the differences in loan behavior characteristics of different user groups, resulting in low risk assessment accuracy. SUMMARY

[0005] Therefore, the embodiments of the present application provide a multi-loan risk assessment method, an electronic device and a storage medium to combine customer group types and multi-loan inquiry information to perform loan risk assessment, thereby improving the accuracy of risk assessment.

[0006] In a first aspect, the embodiments of the present application provide a multi-loan risk assessment method, comprising: obtaining a customer group type to which a target object belongs and multi-loan inquiry information of the target object within a historical time window; obtaining risk assessment information of the target object according to the customer group type and the multi-loan inquiry information; performing loan risk assessment on the target object according to the risk assessment information.

[0007] In an optional implementation, the multi-loan inquiry information includes historical inquiry times of each inquiry of at least two loan inquiry channels, and the obtaining of the risk assessment information of the target object according to the customer group type and the multi-loan inquiry information comprises: obtaining sub-time sequence decay loan features of each inquiry according to a customer group risk weight corresponding to the customer group type, the historical inquiry times of each inquiry and a preset time risk weight; obtaining time sequence decay loan features according to the sub-time sequence decay loan features of each inquiry; obtaining the risk assessment information according to the time sequence decay loan features.

[0008] In an optional implementation, the obtaining of the risk assessment information according to the time sequence decay loan features comprises: Based on the number of queries through the at least two loan query channels and the channel risk weight corresponding to each loan query channel, the loan risk factor of the target object is obtained; The risk assessment information is obtained based on the time-decaying lending characteristics and the lending risk factors.

[0009] In an optional implementation, obtaining the lending risk factor of the target object based on the number of queries from the at least two lending query channels and the channel risk weight corresponding to each lending query channel includes: Based on the number of queries from the at least two loan query channels, obtain the total number of queries within the historical time window; Based on the number of queries for each loan query channel and the total number of queries, the query contribution of each loan query channel is obtained. The lending risk factor is obtained based on the query contribution and the channel risk weight.

[0010] In an optional implementation, obtaining the risk assessment information based on the time-decaying lending characteristics and the lending risk factor includes: Based on the time-decaying lending characteristics, a pre-trained risk assessment model is used to obtain the time-decay lending risk score of the target object; The risk assessment information is obtained based on the time-series lending risk score and the lending risk factors.

[0011] In an optional implementation, the method further includes: Update the risk weight of the customer group according to the industry cycle corresponding to the customer group type; The step of obtaining the sub-time-series decaying lending characteristics of each query based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight includes: Based on the updated customer group risk weights, the historical query times of each query, and the time risk weights corresponding to the historical query times of each query, the sub-time-series decay lending characteristics of each query are obtained.

[0012] In an optional implementation, the method further includes: Dynamically update the channel risk weights corresponding to each loan inquiry channel; The step of obtaining the lending risk factor of the target object based on the number of queries from the at least two lending query channels and the channel risk weight corresponding to each lending query channel includes: The lending risk factor is obtained based on the number of queries from the at least two lending query channels and the updated channel risk weights corresponding to each lending query channel.

[0013] In an optional implementation, the method further includes: Categorize multiple historical lending objects to generate a customer group type library; The process of obtaining the customer group type of the target object includes: Determine the customer group type of the target object from the customer group type library.

[0014] Secondly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method described in any of the first aspects.

[0015] Thirdly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method described in any of the first aspects.

[0016] This application provides a method, electronic device, and storage medium for assessing multiple borrowing risks. The method includes: acquiring the customer group type of the target object and its multiple borrowing query information within a historical time window; acquiring risk assessment information for the target object based on the customer group type and the multiple borrowing query information; and conducting a borrowing risk assessment on the target object based on the risk assessment information. This approach, combining customer group type and multiple borrowing query information to conduct borrowing risk assessment, improves the accuracy of risk assessment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure One ; Figure 2 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Two ; Figure 3A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Three ; Figure 4 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Four ; Figure 5 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Five ; Figure 6 A schematic diagram of the structure of the multiple borrowing risk assessment device provided in the embodiments of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] Traditional risk control models assess lending risk by statistically analyzing features such as the number of credit inquiries or loan applications within a fixed time window. This approach fails to consider the differences in borrowing behavior among different user groups. In other words, it applies the same standards and thresholds to all users. For example, applying for more than 5 loans in the past 30 days is considered high risk. However, this ignores the differences in borrowing behavior among different user groups. For self-employed individuals, frequent borrowing for business turnover is common, so applying for more than 5 loans in the past 30 days does not necessarily indicate high risk. But for low-income groups, multiple borrowing may truly reflect their financial vulnerability, and applying for more than 5 loans in the past 30 days would be considered high risk.

[0021] To address the aforementioned issues, this application improves the accuracy of risk assessment by combining customer group types and multiple loan inquiry information.

[0022] Figure 1 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure One In this embodiment, the executing entity can be an electronic device, such as a computer device or a terminal device.

[0023] like Figure 1 As shown, the method may include: S101. Obtain the customer group type of the target object and the multiple loan query information of the target object within the historical time window.

[0024] The target group is the borrower, and the target group is segmented to obtain the customer group type of the target group, such as business owners, self-employed individuals, white-collar workers, blue-collar workers, etc.

[0025] Historical time windows can be time windows prior to the current moment, such as within 30 days or within 50 days.

[0026] Multiple loan inquiry information of a target object within a historical time window refers to the loan inquiry information of a target object from at least two lending institutions within a historical time window. Specifically, if a target object applies for a loan from a lending institution once, it will trigger the lending institution to conduct a loan inquiry on the target object, which is also known as a credit inquiry. The lending institution here can also be understood as the loan inquiry channel, such as banks, consumer finance companies, and guarantee platforms.

[0027] The loan inquiry information may include the historical inquiry time of each inquiry by the target object on various loan inquiry channels. For example, user A has made 3 inquiries in the past 30 days: once on loan inquiry channel 1 7 days ago, once on loan inquiry channel 2 15 days ago, and once on loan inquiry channel 2 30 days ago.

[0028] In an alternative implementation, the method may further include: Categorize multiple historical lending objects to generate a customer group type library; Step S101 above, obtaining the customer group type of the target object, may include: Determine the target object's customer group type from the customer group type library.

[0029] Among these, multiple historical borrowers are those who have completed loans through at least two loan inquiry channels. The process involves collecting multiple loan inquiry information from at least two channels and conducting full data analysis to identify these multiple historical borrowers. These borrowers are then categorized, for example, into n customer group types, forming a customer group type database. The target customer's customer group type is then matched from this database for refined customer group management. The n customer group types can include business owners, self-employed individuals, white-collar workers, blue-collar workers, etc.

[0030] S102. Obtain risk assessment information for the target group based on customer type and multiple loan inquiry information.

[0031] S103. Based on the risk assessment information, conduct a lending risk assessment on the target entity.

[0032] By considering the target customer group type and combining it with multiple loan inquiry information, risk assessment information is obtained to evaluate the target customer's lending risk. This risk assessment information is used to assess the target customer's lending risk. In this way, when assessing lending risk for different customer groups, the differences in borrowing behavior characteristics of different user groups can be taken into account, improving the accuracy of risk assessment, and the timeliness of risk assessment is improved through a real-time feedback mechanism.

[0033] Figure 2 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Two ,like Figure 2 As shown, in an optional implementation, the multiple loan query information includes: the historical query time of each query from at least two loan query channels.

[0034] Step S102 above, which involves obtaining risk assessment information for the target based on customer type and multiple loan inquiry information, may include: S201. Based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight, obtain the sub-time-series decay lending characteristics of each query.

[0035] Different customer groups have different risk weights. For example, the risk weight for customer group 1 is λ=0.15, and the risk weight for customer group 2 is λ=0.08. The preset time risk weight is the weight of the time elapsed since the query on the lending risk. Customer group 1 can be blue-collar workers, and customer group 2 can be white-collar workers.

[0036] The historical query time for each query can be the time since the query. The sub-time series decaying borrowing and lending characteristics are negatively correlated with the time since the query. The longer the time since the query, the smaller the sub-time series decaying borrowing and lending characteristics, and the shorter the time since the query, the larger the sub-time series decaying borrowing and lending characteristics.

[0037] In some embodiments, a time decay function is defined: ,in, To query the present time, for example: User A has 3 queries within 30 days (7 days ago, 15 days ago, and 30 days ago), with a preset time risk weight of 0.8, and User A belongs to customer group 1, then the sub-time-series decaying borrowing and lending characteristics of each query can be represented as: ≈0.28 ≈0.09 ≈0.01.

[0038] In some embodiments, different historical query times can correspond to different time risk weights. Since historical query times and sub-time series decay lending characteristics are negatively correlated, the longer the time since the beginning of the year, the greater the time risk weight, and the shorter the time since the beginning of the year, the smaller the time risk weight.

[0039] In some embodiments, step S201 above, which obtains the sub-time-series decaying lending characteristics of each query based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight, may include: Based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the time risk weight corresponding to the historical query time of each query, the sub-time-series decay lending characteristics of each query are obtained.

[0040] S202. Obtain the time-series decaying borrowing features based on the sub-time-series decaying borrowing features of each query.

[0041] The time-series decaying borrowing feature can include the sub-time-series decaying borrowing feature of each query and the total sub-time-series decaying borrowing feature.

[0042] Continuing with the example above, user A's total sub-time-decayed lending feature can be the sum of the sub-time-decayed lending features of each query, expressed as: .

[0043] S203. Obtain risk assessment information based on the time-decaying lending characteristics.

[0044] In some embodiments, the time-decaying lending features are used as input to a pre-trained risk assessment model. After processing by the risk assessment model, the model output is the time-decay lending risk score of the target object. The risk assessment information includes the time-decay lending risk score. The higher the time-decay lending risk score, the higher the lending risk; the lower the time-decay lending risk score, the lower the lending risk.

[0045] The risk assessment model is trained using historical time-series decaying lending features labeled with historical time-series lending risk scores. These historical time-series decaying lending features can include sub-time-series decaying lending features of each historical query and total historical sub-time-series decaying lending features.

[0046] In this embodiment, generally speaking, the more multiple queries there are, the higher the lending risk. The traditional fixed window statistics are replaced with time decay weighted summation, which pays more attention to recent high-risk queries, improves time sensitivity, and thus improves the accuracy of lending risk assessment.

[0047] Figure 3 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Three ,like Figure 3As shown, in an optional implementation, step S203 above, obtaining risk assessment information based on time-decaying lending characteristics, may include: S301. Based on the number of queries from at least two loan query channels and the channel risk weight corresponding to each loan query channel, obtain the loan risk factor of the target object.

[0048] Different loan inquiry channels correspond to different channel risk weights. Taking banks, consumer finance companies, and guarantee platforms as examples, banks have higher application thresholds. If a user starts frequently applying for bank loans, it may indicate that their cash flow is already tight, but they are still seeking low-cost funds. Consumer finance companies have more lenient risk control than banks and offer higher interest rates. Frequent applications to these institutions send a stronger risk signal. Guarantee platforms offer high interest rates and low thresholds. If a user starts being frequently inquired about by these institutions, it often means that they have already been rejected by banks, which is a strong signal of high risk. Therefore, the channel risk weight corresponding to microloan platforms is > the channel risk weight corresponding to consumer finance companies > the channel risk weight corresponding to banks.

[0049] For example, the channel risk weight W for banks is 0.8, for consumer finance companies it is 1.0, and for guarantee platforms it is 1.2.

[0050] The system obtains the number of queries for the target object on various lending inquiry channels. Based on the number of queries on each lending inquiry channel and the channel risk weight corresponding to each lending inquiry channel, the system calculates the lending risk factor for the target object. The lending risk factor can be a weighted sum of the number of queries on each lending inquiry channel and the channel risk weight corresponding to each lending inquiry channel, thereby enabling the system to assess the lending risk of the target object from the perspective of lending inquiry channels.

[0051] In an optional implementation, step S301 above, which involves obtaining the lending risk factor of the target object based on the number of queries from at least two lending query channels and the channel risk weight corresponding to each lending query channel, may include: Based on the number of queries from at least two lending inquiry channels, obtain the total number of queries within the historical time window; Based on the number of queries for each loan query channel and the total number of queries, obtain the query contribution of each loan query channel; Based on the query contribution and channel risk weight, obtain the lending risk factor.

[0052] The total number of queries in the historical window is the sum of the number of queries from at least two lending query channels. The ratio of the number of queries from each lending query channel to the total number of queries is taken as the query contribution of each lending query channel. The lending risk factor is obtained by weighting the query contribution and channel risk weight.

[0053] In some embodiments, a risk factor function is defined: ,in This refers to the channel risk weight corresponding to each loan inquiry channel. The higher the risk of the loan inquiry channel, the higher the channel risk weight. This represents the contribution of each loan inquiry channel. For example, if the total number of inquiries is 100 and the number of bank inquiries is 40, the bank's contribution is 40%, totaling 100%.

[0054] For example, a bank has a channel risk weight W=0.8 and a query contribution C=20%; a consumer finance company has a channel risk weight W=1.0 and a query contribution C=50%; and a guarantee platform has a channel risk weight W=1.2 and a query contribution C=30%. Therefore, the lending risk factor (RF) = 0.8 × 0.2 + 1 × 0.5 + 1.2 × 0.3 = 1.02. Specifically, when the contribution of low-risk queries increases, the risk factor decreases; conversely, when the contribution of high-risk queries increases, the risk factor increases.

[0055] S302. Obtain risk assessment information based on the time-decaying lending characteristics and lending risk factors.

[0056] By combining time-decaying lending characteristics and lending risk factors to obtain risk assessment information, risk assessment of target objects can be comprehensively conducted from the perspectives of customer groups, time, and lending inquiry channels, thereby improving the accuracy of lending risk assessment.

[0057] In an optional implementation, step S302 above, obtaining risk assessment information based on time-decaying lending characteristics and lending risk factors, may include: Based on the characteristics of time-decaying lending, a pre-trained risk assessment model is used to obtain the time-decay lending risk score of the target object; Risk assessment information is obtained based on time-series lending risk scores and lending risk factors.

[0058] The risk assessment model was trained using historical time-series decaying lending features labeled with historical time-series lending risk scores.

[0059] The time-decaying lending characteristics are used as input to the risk assessment model, and the model output is the time-decay lending risk score of the target object. Then, risk assessment information is obtained based on the time-decay lending risk score and lending risk factors.

[0060] In some embodiments, risk assessment information may be a time-series lending risk score and the sum of lending risk factors.

[0061] In other embodiments, the risk assessment information can be the product of a time-series lending risk score and a lending risk factor, expressed as:

[0062] in, This indicates risk assessment information. This indicates the time-series lending risk score. This indicates the risk factor for lending.

[0063] In this embodiment, the accuracy of loan assessment is improved by quantifying the risk weights of different loan inquiry channels.

[0064] Figure 4 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Four ,like Figure 4 As shown, in an optional implementation, the method may further include: S401. Update the customer group risk weights according to the industry cycle corresponding to the customer group type.

[0065] Among them, the industry cycle corresponding to the customer group type is the cycle of the industry in which the customer group type is located, including the industry upswing, the industry stabilization period, and the industry downturn.

[0066] If the industry is currently in a downturn, the customer group risk weight λ is reduced to update the customer group risk weight. For example, the customer group risk weight λ is reduced by 10% to extend the risk-sensitive period. If the industry is currently in an upturn or stable period corresponding to the customer group type, the customer group risk weight λ is kept fixed.

[0067] Step S201 above, based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight, obtains the sub-time-series decaying lending characteristics of each query, which may include: S402. Based on the updated customer group risk weights, the historical query times of each query, and the time risk weights corresponding to the historical query times of each query, obtain the sub-time-series decaying lending characteristics of each query.

[0068] Each query's historical query time corresponds to a time risk weight. Different historical query times can correspond to different time risk weights. Since historical query time and sub-time series decay lending characteristics are negatively correlated, the longer the time since the query, the greater the time risk weight, and the shorter the time since the query, the smaller the time risk weight.

[0069] Based on the updated customer group risk weights, the historical query times of each query, and the time risk weights corresponding to the historical query times of each query, the sub-time-series decaying lending characteristics of each query are calculated using the aforementioned time decay function.

[0070] In this embodiment, by dynamically adjusting the risk weight of customer groups based on the industry cycle corresponding to the customer group type, the problem of assessment lag caused by traditional fixed parameters is solved, and the loan risk assessment is more reasonable.

[0071] Figure 5 A flowchart illustrating the multiple borrowing risk assessment method provided in this application embodiment. Figure Five ,like Figure 5 As shown, in an optional implementation, the method may further include: S501. Dynamically update the channel risk weights corresponding to each loan inquiry channel.

[0072] The channel risk weights for each loan inquiry channel are dynamically updated according to a preset cycle, such as monthly, thereby reassessing the channel risk weights for each loan inquiry channel.

[0073] In some embodiments, the channel risk weight corresponding to each loan inquiry channel is dynamically updated based on the status of the loan risk inquiry channel. The status of the loan risk inquiry channel may include the marketing strategy of the loan inquiry channel. For example, if the marketing strategy is changed to expand to lower-tier customer groups, the corresponding channel risk weight will be reduced.

[0074] In other embodiments, the channel risk weights corresponding to each lending inquiry channel are dynamically updated according to the regulatory strategy for lending risk inquiry channels. For example, if the regulatory strategy is a tightening strategy, the corresponding channel risk weights are increased.

[0075] Of course, the above is only an example description. In actual situations, there are, but are not limited to, the two methods mentioned above for dynamically updating channel risk weights.

[0076] Step S301 above, which obtains the lending risk factor of the target object based on the number of queries from at least two lending query channels and the channel risk weight corresponding to each lending query channel, may include: S502. Obtain the lending risk factor based on the number of queries from at least two lending query channels and the updated channel risk weights corresponding to each lending query channel.

[0077] In this embodiment, a dynamic analysis and update mechanism is used to quantify the updated channel risk weights corresponding to each lending query channel to calculate the lending risk factor, which improves the accuracy of the lending risk factor calculation. Furthermore, since traditional risk control models use static parameters and have long iteration and deployment cycles, they cannot capture changes in behavioral patterns brought about by different economic cycles and industry policy changes. This solution uses parameter updates to dynamically adjust the risk assessment model in real time, achieving real-time model updates to a certain extent. At the same time, it can play a significant role in tightening risk in environments with rapidly changing risks.

[0078] Based on the same inventive concept, this application also provides a multiple borrowing risk assessment device corresponding to the multiple borrowing risk assessment method. Since the principle of the device in this application is similar to the multiple borrowing risk assessment method described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0079] Figure 6 This is a schematic diagram of the structure of a multi-borrowing risk assessment device provided in an embodiment of this application. The device can be located in an electronic device.

[0080] like Figure 6 As shown, the device may include: The acquisition module 601 is used to acquire the customer group type of the target object and the multiple loan query information of the target object within the historical time window; The acquisition module 601 is also used to acquire risk assessment information of the target object based on the customer group type and multiple loan query information; The processing module 602 is used to conduct a lending risk assessment on the target object based on the risk assessment information.

[0081] In an optional implementation, the multiple lending query information includes: the historical query time of each query from at least two lending query channels. The acquisition module 601 is specifically used for: Based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight, the sub-time series decay lending characteristics of each query are obtained. Based on the sub-time-series decaying borrowing features of each query, obtain the time-series decaying borrowing features; Risk assessment information is obtained based on the time-decaying lending characteristics.

[0082] In an optional implementation, the acquisition module 601 is specifically used for: Based on the number of queries from at least two loan query channels and the channel risk weight corresponding to each loan query channel, obtain the loan risk factor of the target object; Risk assessment information is obtained based on the time-decaying lending characteristics and lending risk factors.

[0083] In an optional implementation, the acquisition module 601 is specifically used for: Based on the number of queries from at least two lending inquiry channels, obtain the total number of queries within the historical time window; Based on the number of queries for each loan query channel and the total number of queries, obtain the query contribution of each loan query channel; Based on the query contribution and channel risk weight, obtain the lending risk factor.

[0084] In an optional implementation, the acquisition module 601 is specifically used for: Based on the characteristics of time-decaying lending, a pre-trained risk assessment model is used to obtain the time-decay lending risk score of the target object; Risk assessment information is obtained based on time-series lending risk scores and lending risk factors.

[0085] In an optional implementation, the processing module 602 is further configured to: Update customer group risk weights based on the industry cycle corresponding to the customer group type. The acquisition module is specifically used to obtain the sub-time-series decaying lending characteristics of each query based on the updated customer group risk weight, the historical query time of each query, and the time risk weight corresponding to the historical query time of each query.

[0086] In an optional implementation, the processing module 602 is further configured to: Dynamically update the channel risk weights corresponding to each loan inquiry channel; The acquisition module 601 is specifically used to obtain the lending risk factor based on the number of queries from at least two lending query channels and the updated channel risk weights corresponding to each lending query channel.

[0087] In an optional implementation, the processing module 602 is further configured to: Categorize multiple historical lending objects to generate a customer group type library; The acquisition module is specifically used to determine the customer group type of the target object from the customer group type library.

[0088] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0089] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 7 As shown, the device may include a processor 701, a memory 702, and a bus 703. The memory 702 stores machine-readable instructions that can be executed by the processor 701. When the electronic device is running, the processor 701 communicates with the memory 702 through the bus 703, and the processor 701 executes the machine-readable instructions to perform the above-described method.

[0090] This application also provides a computer-readable storage medium storing a computer program, which is executed by a processor to perform the above-described method.

[0091] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0092] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0093] 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.

[0094] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0095] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 described in the various embodiments of this application. 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.

[0096] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0097] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for assessing the risk of multiple borrowing, characterized in that, include: Obtain the customer group type of the target object and the multiple loan query information of the target object within the historical time window; Based on the customer group type and the multiple borrowing query information, obtain the risk assessment information of the target object; Based on the risk assessment information, a lending risk assessment is conducted on the target entity.

2. The method according to claim 1, characterized in that, The multiple borrowing inquiry information includes: the historical query time of each query from at least two borrowing inquiry channels; the step of obtaining the risk assessment information of the target object based on the customer group type and the multiple borrowing inquiry information includes: Based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight, the sub-time-series decay lending characteristics of each query are obtained. Based on the sub-time-decaying borrowing features of each query, obtain the time-decaying borrowing features; The risk assessment information is obtained based on the time-decaying lending characteristics.

3. The method according to claim 2, characterized in that, The step of obtaining the risk assessment information based on the time-series decay lending characteristics includes: Based on the number of queries through the at least two loan query channels and the channel risk weight corresponding to each loan query channel, the loan risk factor of the target object is obtained; The risk assessment information is obtained based on the time-decaying lending characteristics and the lending risk factors.

4. The method according to claim 3, characterized in that, The step of obtaining the lending risk factor of the target object based on the number of queries from the at least two lending query channels and the channel risk weight corresponding to each lending query channel includes: Based on the number of queries from the at least two loan query channels, obtain the total number of queries within the historical time window; Based on the number of queries for each loan query channel and the total number of queries, the query contribution of each loan query channel is obtained. The lending risk factor is obtained based on the query contribution and the channel risk weight.

5. The method according to claim 3, characterized in that, The step of obtaining the risk assessment information based on the time-decaying lending characteristics and the lending risk factors includes: Based on the time-decaying lending characteristics, a pre-trained risk assessment model is used to obtain the time-decay lending risk score of the target object; The risk assessment information is obtained based on the time-series lending risk score and the lending risk factors.

6. The method according to claim 2, characterized in that, The method further includes: Update the risk weight of the customer group according to the industry cycle corresponding to the customer group type; The step of obtaining the sub-time-series decaying lending characteristics of each query based on the customer group risk weight corresponding to the customer group type, the historical query time of each query, and the preset time risk weight includes: Based on the updated customer group risk weights, the historical query times of each query, and the time risk weights corresponding to the historical query times of each query, the sub-time-series decay lending characteristics of each query are obtained.

7. The method according to claim 3, characterized in that, The method further includes: Dynamically update the channel risk weights corresponding to each loan inquiry channel; The step of obtaining the lending risk factor of the target object based on the number of queries from the at least two lending query channels and the channel risk weight corresponding to each lending query channel includes: The lending risk factor is obtained based on the number of queries from the at least two lending query channels and the updated channel risk weights corresponding to each lending query channel.

8. The method according to claim 1, characterized in that, The method further includes: Categorize multiple historical lending objects to generate a customer group type library; The process of obtaining the customer group type of the target object includes: Determine the customer group type of the target object from the customer group type library.

9. An electronic device, characterized in that, include: The electronic device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the method according to any one of claims 1 to 8.

Images