A method, system, computer device and storage medium for verifying doubtful loans

Abstract

This article provides a method, system, computer device, and storage medium for verifying suspicious loans, which relates to the field of loan verification. Among them, the method for verifying suspicious loans includes: establishing an association relationship between a risk identification library and a loan risk early warning model in advance; each risk identification library includes multiple risk identification factors, and the verification directions of each risk identification factor are different; according to the association relationship between the risk identification library and the loan risk early warning model, determine the risk identification factors associated with the early warning task; the early warning task includes multiple suspicious loans output by the same loan risk early warning model; use the risk identification factors associated with the early warning task to verify the early warning task. This article can automatically determine the verification direction of the early warning task and use relevant risk identification factors for verification, reducing the difficulty of manual verification, and the verification direction is targeted, which can improve the verification efficiency.

Description

Technical Field

[0001] This article relates to the field of loan verification, and particularly to a method, system, computer device, and storage medium for verifying suspicious loans. Background Art

[0002] In the prior art, more than 100 personal loan risk warning models have been established according to dimensions such as loan authenticity and credit risk, loan purpose, operational risk, and post-loan assistance. There are extremely many suspicious personal loan information warned. To ensure the safety of lending, it is necessary to verify the suspicious information. In the prior art, it is mainly verified by manually collecting a large amount of information and analyzing and comparing, which has the problems of consuming a large amount of manpower and time and extremely low verification efficiency.

[0003] In addition, the suspicious personal loan information warned by each loan warning model is numerous and the risk levels are different, and the verification business directions are also different. In the prior art, the verification data dimension is small, the verification means is single, and the pertinence and universality of verification are lacking.

[0004] The existing risk prediction model can only predict risk doubts and classify the risk doubts, but this classification is only a general classification. When conducting verification, it is still manually based on experience to determine the verification direction of the suspicious information output by the risk prediction model, and the verification accuracy and efficiency are relatively low. Summary of the Invention

[0005] This article is used to solve the problems in the prior art that the verification of personal suspicious loans wastes manpower, has low verification efficiency, has a single verification means, and is insufficient to support the universality and pertinence of verification.

[0006] To solve the above technical problems, a first aspect of this article provides a method for verifying suspicious loans, including:

[0007] Pre-establish the association relationship between the risk identification class library and the loan risk warning model; wherein, each risk identification class library includes multiple risk identification factors, and the verification directions of each risk identification factor are different;

[0008] According to the association relationship between the risk identification class library and the loan risk warning model, determine the risk identification factors associated with the warning task; wherein, the warning task includes multiple suspicious loans output by the same loan risk warning model;

[0009] Use the risk identification factors associated with the warning task to verify the warning task.

[0010] In an embodiment of this article, pre-establishing the association relationship between the risk identification class library and the loan risk warning model includes:

[0011] For each historical loan under each loan product, conduct a verification according to each risk identification factor, and determine the risk identification factor combination of each historical loan based on the verification results;

[0012] Based on the historical suspicious loans output by the loan risk early warning model and the risk identification factor combination of each historical loan, determine the risk identification factor combination of the historical suspicious loans;

[0013] According to the risk identification factor combination of the historical suspicious loans, count the number of risk identification factors under the same risk early warning model triggering the same risk identification library;

[0014] If the number of risk identification factors under a risk early warning model triggering a risk identification library accounts for a first predetermined value of the total number of risk identification factors under this risk identification library, then associate this risk identification library with this risk early warning model.

[0015] In an embodiment of this article, establishing the association relationship between the risk identification library and the loan risk early warning model in advance further includes:

[0016] According to the number of risk identification factors under the same risk early warning model triggering the same risk identification library, determine the importance degree of the risk identification library to the risk early warning model;

[0017] According to the importance degree of the risk identification library to the risk early warning model, determine the association order between the risk identification library and the risk early warning model.

[0018] In an embodiment of this article, establishing the association relationship between the risk identification library and the loan risk early warning model in advance further includes:

[0019] Analyze the similarity between the new loan risk early warning model and the existing loan risk early warning models;

[0020] Use the risk identification libraries related to the existing loan risk early warning models with similarity greater than a second predetermined value as the risk identification libraries associated with the new loan risk early warning model.

[0021] In an embodiment of this article, the risk identification factors include: loan screening rules and early warning task screening rules;

[0022] The loan screening rules are used to screen out the suspicious loans to be identified from the early warning tasks;

[0023] The early warning task screening rules are used to verify the suspicious loans to be identified.

[0024] In an embodiment of this article, using the risk identification factors associated with the early warning task to verify the early warning task includes:

[0025] The early warning tasks screen out the suspicious loans to be identified corresponding to each risk identification factor according to the loan screening rules of the risk identification factors associated with them.

[0026] Use the early warning task screening rules in the risk identification factors to verify the suspicious loans to be identified.

[0027] In one embodiment of this article, the method for verifying suspicious loans further includes:

[0028] Determine the relationship coefficient between the early warning task and the risk identification factor according to the suspicious loans that determine risk problems in the historical early warning tasks and the risk identification factors associated with the early warning tasks.

[0029] Sort multiple early warning tasks according to the relationship coefficients between multiple early warning tasks and risk identification factors.

[0030] Verify the early warning tasks according to the sorting results.

[0031] In one embodiment of this article, determining the relationship coefficient between the early warning task and the risk identification factor according to the suspicious loans that determine risk problems in the historical early warning tasks and the risk identification factors associated with the early warning tasks includes:

[0032] According to the suspicious loans that determine risk problems in the historical early warning tasks and the risk identification factors associated with the early warning tasks, through logistic regression algorithm analysis, determine the relationship coefficient between each suspicious loan in the early warning task and the risk identification factor.

[0033] According to the relationship coefficients between each suspicious loan in the early warning task and the risk identification factor, through decision tree algorithm, determine the relationship coefficient between the early warning task and the risk identification factor.

[0034] In one embodiment of this article, the method for verifying suspicious loans further includes:

[0035] Sort the suspicious loans in the early warning task according to the relationship coefficients between each suspicious loan in the early warning task and the risk identification factor.

[0036] When verifying the early warning task, verify the suspicious loans according to the sorting results of the suspicious loans in the early warning task.

[0037] In one embodiment of this article, sorting the suspicious loans in the early warning task according to the relationship coefficients between each suspicious loan in the early warning task and the risk identification factor includes:

[0038] Sort the suspicious loans in the early warning task in descending order according to the relationship coefficients between each suspicious loan in the early warning task and the risk identification factor.

[0039] In one embodiment of the present disclosure, sorting multiple early warning tasks according to the correlation coefficients between the multiple early warning tasks and risk identification factors includes:

[0040] Sorting the multiple early warning tasks in descending order according to the correlation coefficients between the multiple early warning tasks and risk identification factors.

[0041] In one embodiment of the present disclosure, the early warning tasks are generated in the following manner:

[0042] Grouping multiple suspicious loans output by the same loan risk early warning model according to a preset grouping rule;

[0043] Each grouping result constitutes an early warning task.

[0044] A second aspect of the present disclosure provides a suspicious loan verification system, including:

[0045] A preprocessing module for pre-establishing an association relationship between a risk identification library and a loan risk early warning model; wherein each risk identification library includes multiple risk identification factors, and the verification directions of each risk identification factor are different;

[0046] A risk identification factor determination module for determining risk identification factors associated with an early warning task according to the association relationship between the risk identification library and the loan risk early warning model; wherein the early warning task includes multiple suspicious loans output by the same loan risk early warning model;

[0047] A verification module for verifying the early warning task by using the risk identification factors associated with the early warning task.

[0048] In a third aspect of the present disclosure, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the suspicious loan verification method described in any one of the foregoing is implemented.

[0049] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores an executable computer program. When the computer program is executed by a processor, the suspicious loan verification method described in any one of the foregoing embodiments is implemented.

[0050] The suspicious loan verification method, system, computer device, and storage medium provided by the present disclosure can automatically determine the verification direction of an early warning task and perform verification by using relevant risk identification factors by pre-establishing an association relationship between a risk identification library and a loan risk early warning model, reduce the difficulty of manual verification, and the verification direction is targeted, which can improve the verification efficiency.

[0051] In order to make the above and other purposes, features and advantages of this article more obvious and understandable, the following provides preferred embodiments and, in conjunction with the accompanying drawings, detailed descriptions are as follows. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of this article or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of this article. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0053] Figure 1 Shows the first flowchart of the doubtful loan verification method in the embodiments of this article;

[0054] Figure 2 Shows the first flowchart of the process for establishing the association relationship between the risk identification library and the loan risk early warning model in the embodiments of this article;

[0055] Figure 3 Shows the second flowchart of the process for establishing the association relationship between the risk identification library and the loan risk early warning model in the embodiments of this article;

[0056] Figure 4 Shows the third flowchart of the process for establishing the association relationship between the risk identification library and the loan risk early warning model in the embodiments of this article;

[0057] Figure 5 Shows the flowchart of the early warning task verification process in the embodiments of this article;

[0058] Figure 6 Shows the second flowchart of the doubtful loan verification method in the embodiments of this article;

[0059] Figure 7 Shows the flowchart of the process for determining the relationship coefficient between the early warning task and the risk identification factor in the embodiments of this article;

[0060] Figure 8 Shows the structural diagram of the doubtful loan verification system in the embodiments of this article;

[0061] Figure 9 Shows the structural diagram of the computer device in the embodiments of this article.

[0062] Description of the Reference Numerals in the Drawings:

[0063] 810, Preprocessing Module;

[0064] 820, Risk Identification Factor Determination Module;

[0065] 830, Verification Module;

[0066] 902, Computer device;

[0067] 904, Processor;

[0068] 906, Memory;

[0069] 908, Driving mechanism;

[0070] 910, Input / output module;

[0071] 912, Input device;

[0072] 914, Output device;

[0073] 916, Rendering device;

[0074] 918, Graphical user interface;

[0075] 920, Network interface;

[0076] 922, Communication link;

[0077] 924, Communication bus. Detailed implementation manners

[0078] Next, the technical solutions in the embodiments of the present disclosure will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present disclosure.

[0079] In the prior art, there are problems such as waste of manpower, low verification efficiency, single verification means, and insufficient support for the universality and pertinence of verification in the verification of personal doubtful loans.

[0080] To solve the above technical problems, in an embodiment of the present disclosure, a method for verifying doubtful loans is provided. The method for verifying doubtful loans can run on intelligent terminals, including smart phones, tablet computers, desktop computers, etc., and can also be a separate application program, a small program embedded in other programs, or in the form of a web page, etc. The present disclosure does not limit the specific implementation manners. Specifically, as Figure 1 shown, the method for verifying doubtful loans includes;

[0081] Step 110, establishing an association relationship between a risk identification library and a loan risk early warning model in advance; wherein, each risk identification library includes multiple risk identification factors, and the loan verification directions of each risk identification factor are different. The risk identification factor takes the doubtful loans output by the loan risk early warning model as the analysis object and is a risk prompt model established on the basis of the loan risk early warning model.

[0082] The loans in this article refer to the loans that have been reviewed and granted by the bank, including but not limited to personal loans, collective enterprise loans, private enterprise loans, etc. This article does not specifically limit the types of loans.

[0083] The risk identification library is the classification of risk identifications, which groups risk identifications with similar characteristics into one category. The risk identification library can be established based on historical verification data and risk classification information. It can also be established by summarizing the specific verification methods and specific risk matters used by experts during manual verification and confirmation.

[0084] Taking the banking system as an example, using big data analysis technology, deeply mine loan and customer data in dimensions such as customer qualification identification, work unit identification, borrower income identification, abnormal characteristics of the seller of second-hand housing loans, abnormal characteristics of collateral, overvaluation of collateral value, abnormal sources of down payment funds, fund flow of second-hand housing loans, abnormal loan repayment, and associated relationships, and establish a risk identification library. Each risk identification library is further divided into risk identification factors, as shown in Table 1.

[0085] Table 1

[0086]

[0087]

[0088] Table 1 is only a specific embodiment of this article and is not used to limit the risk identification library and risk identification factors. When those skilled in the art implement specifically, other risk identification libraries and risk identification factors can also be set according to requirements.

[0089] There are multiple loan risk early warning models. Each loan risk early warning model includes establishing screening rules for loans according to regulatory requirements or business rules. The input content of the loan risk early warning model is loan information, including but not limited to loan information itself, basic information of loan customers, transfer information, and credit information. The output content of the loan risk early warning model is suspicious loans that meet the screening rules. Among them, suspicious loans are loans with risks, such as loans whose loan purposes do not conform to bank regulations.

[0090] The associated relationship between the risk identification library and the loan risk early warning model is shown in Table 2 below.

[0091] Table 2

[0092]

[0093] Step 120, according to the associated relationship between the risk identification library and the loan risk early warning model, determine the risk identification factors associated with the early warning task; among them, the early warning task includes multiple suspicious loans output by the same loan risk early warning model.

[0094] In this step, the early warning tasks can be generated in the following way: group multiple suspicious loans output by the same loan risk early warning model according to the pre-set grouping rules; each grouping result constitutes an early warning task. Among them, the pre-set grouping rules include, but are not limited to, grouping by the branch where the loan belongs, the counterparty, etc. This article does not make specific limitations on the grouping rules.

[0095] In this step, through the association relationship between the risk identification library and the loan risk early warning model, the verification direction of the suspicious loans in the early warning task can be accurately determined. Specifically, the early warning task is related to the risk early warning model, and the risk early warning model is related to the risk identification library. Therefore, the risk identification factors associated with the early warning task can be determined.

[0096] Step 130, verify the early warning task by using the risk identification factors associated with the early warning task.

[0097] In this step, at least one early warning task screening rule indicating the loan verification direction is defined in each risk identification factor. The suspicious loans to be identified are verified through the risk identification factors, and it is determined whether there is a real risk for the suspicious loans according to the verification results. In specific implementation, in order to ensure the accuracy of the verification results, a manual verification process will also be added.

[0098] In this embodiment, by pre-establishing the association relationship between the risk identification library and the loan risk early warning model, the verification direction of the early warning task can be automatically determined and verified by using the relevant risk identification factors, reducing the difficulty of manual verification, and the verification direction is targeted, which can improve the verification efficiency.

[0099] In an embodiment of this article, as Figure 2 shown, the above step 110 of pre-establishing the association relationship between the risk identification library and the loan risk early warning model includes:

[0100] Step 210, verify each historical loan under each loan product according to the early warning task screening rule of each risk identification factor, and determine the risk identification factor combination of each historical loan according to the verification results.

[0101] In this step, the verification results include risky, risk-free, and undetermined. The risk identification factors with risky verification results are grouped together to obtain the risk identification factor combination of the historical loan.

[0102] Step 220, determine the risk identification factor combination of the historical suspicious loans according to the historical suspicious loans output by the loan risk early warning model and the risk identification factor combination of each historical loan.

[0103] In this step, the historical suspicious loans refer to the loans identified by the loan risk early warning model. The historical loans refer to the loans that have existed historically and have not been identified by the loan risk early warning model.

[0104] First, match the historical suspicious loans with each historical loan, and use the combination of risk identification factors corresponding to the matched historical loans as the combination of risk identification factors for the historical suspicious loans.

[0105] Step 230: According to the combination of risk identification factors of the historical suspicious loans, count the number of risk identification factors triggered by the same risk early warning model under the same risk identification library. If the number of risk identification factors triggered by a risk early warning model under a risk identification library accounts for the first predetermined value of the total number of risk identification factors in this risk identification library, then associate this risk identification library with this risk early warning model. Otherwise, remove the risk identification factors that do not meet the conditions.

[0106] In this step, the historical suspicious loans are identified by the risk early warning model. Therefore, there is an associated relationship between the historical suspicious loans and the risk early warning model, as shown in Table 3. In addition, there is an associated relationship between the historical suspicious loans and the combination of risk identification factors, as shown in Table 4. Based on the same risk early warning model and the same risk identification library as the statistical basis, summarizing Table 3 and Table 4 can determine the number of risk identification factors triggered by the same risk early warning model under the same risk identification library.

[0107] Table 3

[0108] Risk early warning model Historical doubtful loans Risk early warning model 1 Doubtful loan 1, Doubtful loan 2, Doubtful loan 3… …… ……

[0109] Table 4

[0110]

[0111] In an embodiment of this article, for the case where a loan risk early warning model is associated with multiple risk identification libraries, as Figure 3 shown, the process of pre - establishing the associated relationship between the risk identification library and the loan risk early warning model in step 110, in addition to including the above - mentioned steps 210 to 230, also includes:

[0112] Step 310: Determine the importance degree of the risk identification library to the risk early warning model according to the number of risk identification factors triggered by the same risk early warning model under the same risk identification library;

[0113] Step 320: Determine the associated order between the risk identification library and the risk early warning model according to the importance degree of the risk identification library to the risk early warning model.

[0114] Furthermore, step 130 retrieves the risk identification factors in the corresponding risk identification library according to the associated order between the risk identification library and the risk early warning model to verify the early warning task.

[0115] In this embodiment, through steps 310 and 320, the importance of the risk identification library for the risk early warning model can be determined, thereby improving the efficiency of risk identification.

[0116] In one embodiment of this article, as Figure 4 shown, step 110 pre - establishes the association relationship between the risk identification library and the loan risk early warning model, and further includes:

[0117] Step 410, analyze the similarity between the newly added loan risk early warning model and the existing loan risk early warning models.

[0118] In this step, methods such as Euclidean distance, Manhattan distance, and cosine similarity can be used to calculate the similarity between the newly added loan risk early warning model and the existing loan risk early warning models. The specific calculation method is not limited in this article.

[0119] Step 420, use the risk identification library related to the existing loan risk early warning model with a similarity greater than a second predetermined value (for example, 90%, which can be set according to requirements) as the risk identification library associated with the newly added individual loan risk early warning model.

[0120] This embodiment can improve the association speed between the newly added loan early warning model and the loan risk early warning model.

[0121] In one embodiment of this article, in addition to the early warning task screening rules, the risk identification factors also include loan screening rules, and the loan screening rules are used to screen out the suspicious loans to be identified from the early warning tasks. The early warning task screening rules are used to verify the suspicious loans to be identified.

[0122] Specifically, as Figure 5 shown, the above - mentioned step 130 verifies the early warning tasks using the risk identification factors associated with the early warning tasks, including:

[0123] Step 510, according to the loan screening rules in the risk identification factors associated with the early warning task, screen out the suspicious loans to be identified corresponding to each risk identification factor.

[0124] Step 520, use the early warning task screening rules in the risk identification factors to verify the suspicious loans to be identified.

[0125] This step can verify the early warning tasks hierarchically, improve the verification efficiency, and reduce the verification of unnecessary suspicious loans.

[0126] In one embodiment of this article, as Figure 6 shown, in addition to the above - mentioned steps 110 to 120, the method for verifying suspicious loans further includes:

[0127] Step 610: Determine the relationship coefficient between the early warning task and the risk identification factor based on the doubtful loans that identify risk issues in the historical early warning tasks and the risk identification factors associated with the early warning tasks.

[0128] In this step, the relationship coefficient between the early warning task and the risk identification factor can reflect the judgment accuracy of the risk identification factor for the early warning task. The larger the relationship coefficient, the greater the judgment accuracy of the risk identification factor for the early warning task. Specifically, as Figure 7 shown, the process of determining the relationship coefficient between the early warning task and the risk identification factor based on the doubtful loans that identify risk issues in the historical early warning tasks and the risk identification factors associated with the early warning tasks includes:

[0129] Step 710: Determine the relationship coefficient between each doubtful loan in the early warning task and the risk identification factor through a logistic regression algorithm based on the doubtful loans that identify risk issues in the historical early warning tasks and the risk identification factors associated with the early warning tasks. Specifically, the implementation process of the logistic regression algorithm can refer to the existing technology and will not be elaborated here.

[0130] Step 720: Determine the relationship coefficient between the early warning task and the risk identification factor through a decision tree algorithm based on the relationship coefficient between each doubtful loan in the early warning task and the risk identification factor. Specifically, the implementation process of the decision tree algorithm can refer to the existing technology and will not be elaborated here.

[0131] Step 620: Sort multiple early warning tasks according to the relationship coefficients between the multiple early warning tasks and the risk identification factors.

[0132] In this step, sort the multiple early warning tasks in descending order according to the relationship coefficients between the multiple early warning tasks and the risk identification factors.

[0133] Step 130 is further step 130': According to the sorting result, verify the early warning task by using the risk identification factor associated with the early warning task. For example, there are three early warning tasks A, early warning task B, and early warning task C. Sort these three early warning tasks in descending order of the relationship coefficient according to the relationship coefficient between the early warning task and the risk identification factor. The sorting result is early warning task B, early warning task A, and early warning task C. Step 630 verifies the early warning tasks according to the sorting result of early warning task B, early warning task A, and early warning task C.

[0134] This embodiment can preferentially identify early warning tasks with high risks and improve the verification efficiency of early warning tasks.

[0135] In one embodiment of the present invention, in addition to the above steps 110 to 120, steps 610 to 620, and steps 710 to 720, the method for verifying suspicious loans further includes: sorting the suspicious loans in the early warning task according to the correlation coefficient between each suspicious loan and the risk identification factor in the early warning task.

[0136] In this step, the correlation coefficient between each suspicious loan and the risk identification factor in the early warning task can reflect the judgment accuracy of the risk identification factor for the suspicious loan. The larger the correlation coefficient, the greater the judgment accuracy of the risk identification factor for the suspicious loan.

[0137] Specifically, when implementing, sort the suspicious loans in the early warning task in descending order according to the correlation coefficient between each suspicious loan and the risk identification factor in the early warning task.

[0138] When performing the verification of the early warning task in step 130, verify the suspicious loans according to the sorting result of the suspicious loans in the early warning task.

[0139] This embodiment can preferentially identify suspicious loans with high risks and improve the verification efficiency of suspicious loans.

[0140] Based on the same inventive concept, the present invention also provides a system for verifying suspicious loans as described in the following embodiments. Since the principle of the system for verifying suspicious loans to solve problems is similar to that of the method for verifying suspicious loans, the implementation of the system for verifying suspicious loans can refer to the method for verifying suspicious loans, and the repeated parts will not be described in detail. The system for verifying suspicious loans provided in this embodiment includes multiple functional modules, all of which can be implemented by dedicated or general-purpose chips, or can also be implemented by software programs, and the present invention does not limit this.

[0141] Specifically, as Figure 8 shown, the system for verifying suspicious loans includes:

[0142] A preprocessing module 810, configured to pre-establish an association relationship between a risk identification library and a loan risk early warning model; wherein, each risk identification library includes multiple risk identification factors, and the verification directions of each risk identification factor are different;

[0143] A risk identification factor determination module 820, configured to determine the risk identification factors associated with the early warning task according to the association relationship between the risk identification library and the loan risk early warning model; wherein, the early warning task includes multiple suspicious loans output by the same loan risk early warning model;

[0144] A verification module 830, configured to verify the early warning task by using the risk identification factors associated with the early warning task.

[0145] The doubtful loan verification system provided in this paper can automatically determine the verification direction of early warning tasks and conduct verifications using relevant risk identification factors by pre - establishing the association relationship between the risk identification library and the loan risk early warning model, reducing the difficulty of manual verification, and the verification direction is targeted, which can improve the verification efficiency.

[0146] To more clearly illustrate the technical solution of this paper, a specific embodiment is used for detailed description below. Specifically, the doubtful loan verification method includes:

[0147] 1. Establish a risk identification library.

[0148] 2. According to the historical loan information and historical doubtful loan information, establish the association relationship between the risk identification library and the loan risk early warning model.

[0149] (a) For each historical loan under each loan product, conduct verifications according to the screening rules of early warning tasks for each risk identification factor, and determine the combination of risk identification factors for each historical loan based on the verification results;

[0150] (b) According to the historical doubtful loans output by the loan risk early warning model and the combination of risk identification factors for each historical loan, determine the combination of risk identification factors for historical doubtful loans;

[0151] (c) According to the combination of risk identification factors for historical doubtful loans, count the number of risk identification factors triggered by the same risk early warning model under the same risk identification library. If the number of risk identification factors triggered by a risk early warning model under a risk identification library accounts for a first predetermined value of the total number of risk identification factors under this risk identification library, then associate this risk identification library with this risk early warning model.

[0152] Determine the importance of the risk identification library to the risk early warning model according to the number of risk identification factors triggered by the same risk early warning model under the same risk identification library.

[0153] 3. Calculate the relationship coefficient between the early warning task and the risk identification factor and the relationship coefficient between the doubtful loan and the risk identification factor.

[0154] (a) According to the association relationship between the risk identification library and the loan risk early warning model and the relationship between the early warning task and the loan risk early warning model, determine the risk identification factors associated with the early warning task.

[0155] (b) According to the doubtful loans that determine risk problems in historical early warning tasks and the risk identification factors associated with the early warning task, analyze through the logistic regression algorithm to determine the relationship coefficient between each doubtful loan in the early warning task and the risk identification factor.

[0156] (c) Determine the relationship coefficient between the early warning task and the risk identification factor through the decision tree algorithm according to the relationship coefficient between each suspicious loan in the early warning task and the risk identification factor.

[0157] 4. Hierarchical management of early warning tasks and suspicious loans in early warning tasks

[0158] (a) Sort the suspicious loans in the early warning task in descending order according to the relationship coefficient between each suspicious loan in the early warning task and the risk identification factor.

[0159] (b) Sort multiple early warning tasks in descending order according to the relationship coefficient between the early warning task and the risk identification factor. According to the sorting result, divide every N (for example, three) early warning tasks into a group.

[0160] 5. Verify the suspicious information according to the hierarchical result in 4.

[0161] Retrieve the early warning tasks in the order of the early warning task groups, and verify the suspicious loans according to the sorting of the suspicious loans in the early warning tasks.

[0162] In a specific embodiment of this article, the loan risk early warning model is, for example, an early warning model for non-compliance of the purpose of express loans. This early warning model monitors the flow of express loan funds to the real estate and capital markets; by monitoring the express loan disbursement situation and the flow of funds in the borrower's account, it warns of suspicious information about the flow of express loan funds.

[0163] The risk identification class library associated with this early warning model is the abnormal fund flow class, and the risk identification factor in the abnormal fund flow class library is the risk identification of equal-amount fund reflux.

[0164] The relationship coefficient between the equal-amount fund reflux risk identification factor and this early warning model is 1 (that is, the relevance is close). By mining the historical suspicious points of this early warning model, it is possible to directly determine whether the suspicious loans of this early warning model are compliant through this equal-amount fund reflux risk identification factor.

[0165] In an embodiment of this article, a computer device is also provided. A suspicious loan verification system is installed in the computer device, such as Figure 9As shown, the computer device 902 may include one or more processors 904, such as one or more central processing units (CPUs), and each processing unit may implement one or more hardware threads. The computer device 902 may also include any memory 906 for storing any kind of information such as code, settings, data, etc. By way of non-limiting example, for instance, the memory 906 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory devices, hard disks, optical discs, etc. More generally, any memory may store information using any technology. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 902. In one case, when the processor 904 executes the associated instructions stored in any memory or combination of memories, the computer device 902 may perform any operation of the associated instructions. The computer device 902 also includes one or more drive mechanisms 908 for interacting with any memory, such as a hard disk drive mechanism, an optical disc drive mechanism, etc.

[0166] The computer device 902 may also include an input / output module 910 (I / O) for receiving various inputs (via the input device 912) and for providing various outputs (via the output device 914). A specific output mechanism may include a presentation device 916 and an associated graphical user interface 918 (GUI). In other embodiments, the input / output module 910 (I / O), the input device 912, and the output device 914 may not be included, and it may only be a computer device in a network. The computer device 902 may also include one or more network interfaces 920 for exchanging data with other devices via one or more communication links 922. One or more communication buses 924 couple the components described above together.

[0167] The communication link 922 may be implemented in any way, for example, via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 922 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0168] In one embodiment herein, a computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is run by a processor, it executes the steps of the above-mentioned doubtful loan verification method.

[0169] In one embodiment herein, a computer-readable instruction is also provided, and when the processor executes the instruction, the program therein causes the processor to execute the above-mentioned doubtful loan verification method.

[0170] It should be understood that in various embodiments herein, the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments herein.

[0171] It should also be understood that in the embodiments herein, the term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0172] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this article.

[0173] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0174] In the several embodiments provided herein, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling or direct coupling or communication connection to each other can be an indirect coupling or communication connection through some interfaces, devices, or units, and can also be in an electrical, mechanical, or other form of connection.

[0175] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments herein.

[0176] In addition, each functional unit in the various embodiments of this document may be integrated into a processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0177] If the above-mentioned integrated unit is implemented in the form of 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 essence of the technical solution in this document, or the part that contributes to the prior art, or all or part of this 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 enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this document. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0178] Specific embodiments are used in this document to elaborate on the principles and implementation manners of this document. The description of the above embodiments is only used to help understand the method and its core idea in this document; at the same time, for those of ordinary skill in the art, according to the idea in this document, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this document.

Claims

1. A method for verifying doubtful loans, characterized in that, Including: Pre - establish the association relationship between the risk identification library and the loan risk early - warning model; wherein, each risk identification library includes multiple risk identification factors, and the verification directions of each risk identification factor are different; According to the association relationship between the risk identification library and the loan risk early - warning model, determine the risk identification factors associated with the early - warning task; wherein, the early - warning task includes multiple suspected loans output by the same loan risk early - warning model; Use the risk identification factors associated with the early - warning task to verify the early - warning task; Among them, pre - establishing the association relationship between the risk identification library and the loan risk early - warning model includes: Verify each historical loan under each loan product according to each risk identification factor, and determine the risk identification factor combination of each historical loan according to the verification results; According to the historical suspected loans output by the loan risk early - warning model and the risk identification factor combination of each historical loan, determine the risk identification factor combination of the historical suspected loans; According to the risk identification factor combination of the historical suspected loans, count the number of risk identification factors triggered by the same risk early - warning model under the same risk identification library; If the number of risk identification factors triggered by a risk early - warning model under a risk identification library accounts for a first predetermined value of the total number of risk identification factors under this risk identification library, then associate this risk identification library with this risk early - warning model.

2. The doubtful loan verification method according to claim 1, wherein Pre - establishing the association relationship between the risk identification library and the loan risk early - warning model also includes: According to the number of risk identification factors triggered by the same risk early - warning model under the same risk identification library, determine the importance degree of the risk identification library to the risk early - warning model; According to the importance degree of the risk identification library to the risk early - warning model, determine the association order between the risk identification library and the risk early - warning model.

3. The doubtful loan verification method according to claim 1, characterized in that Pre - establishing the association relationship between the risk identification library and the loan risk early - warning model also includes: Analyze the similarity between the new loan risk early - warning model and the existing loan risk early - warning models; Take the risk identification libraries related to the existing loan risk early - warning models with similarity greater than a second predetermined value as the risk identification libraries associated with the new loan risk early - warning model.

4. The doubtful loan verification method according to claim 1, wherein, The risk identification factors include: loan screening rules and early - warning task screening rules; The loan screening rules are used to screen out the suspected loans to be identified from the early - warning task; The early - warning task screening rules are used to verify the suspected loans to be identified.

5. The doubtful loan verification method according to claim 1, characterized in that, Using the risk identification factors associated with the early - warning task to verify the early - warning task includes: The early - warning task screens out the suspected loans to be identified corresponding to each risk identification factor according to the loan screening rules of its associated risk identification factors; Use the early - warning task screening rules in the risk identification factors to verify the suspected loans to be identified.

6. The doubtful loan verification method according to claim 1, characterized in that, Also including: According to the suspected loans that determine risk problems in the historical early - warning tasks and the risk identification factors associated with the early - warning task, determine the relationship coefficient between the early - warning task and the risk identification factor; Sort multiple early - warning tasks according to the relationship coefficients between multiple early - warning tasks and risk identification factors; According to the sorting results, verify the early - warning tasks.

7. The doubtful loan verification method according to claim 6, wherein, Determine the relationship coefficient between the early warning task and the risk identification factors based on the doubtful loans that identify risk issues in the historical early warning tasks and the risk identification factors associated with the early warning tasks, including: Based on the doubtful loans that identify risk issues in the historical early warning tasks and the risk identification factors associated with the early warning tasks, through logistic regression algorithm analysis, determine the relationship coefficient between each doubtful loan in the early warning task and the risk identification factors; Based on the relationship coefficient between each doubtful loan in the early warning task and the risk identification factors, through decision tree algorithm, determine the relationship coefficient between the early warning task and the risk identification factors.

8. The doubtful loan verification method according to claim 7, wherein Also included: Sort the doubtful loans in the early warning task according to the relationship coefficient between each doubtful loan in the early warning task and the risk identification factors; When conducting the verification of the early warning task, verify the doubtful loans according to the sorting result of the doubtful loans in the early warning task.

9. The doubtful loan verification method according to claim 8, characterized in that Sort the doubtful loans in the early warning task according to the relationship coefficient between each doubtful loan in the early warning task and the risk identification factors, including: Sort the doubtful loans in the early warning task in descending order according to the relationship coefficient between each doubtful loan in the early warning task and the risk identification factors.

10. The doubtful loan verification method according to claim 6, characterized in that, Sort multiple early warning tasks according to the relationship coefficients between multiple early warning tasks and the risk identification factors, including: Sort multiple early warning tasks in descending order according to the relationship coefficients between multiple early warning tasks and the risk identification factors.

11. The doubtful loan verification method according to claim 1, characterized in that, The early warning task is generated in the following manner: Group the multiple doubtful loans output by the same loan risk early warning model according to the pre-set grouping rules; Each grouping result constitutes an early warning task.

12. A doubtful loan verification system, characterized in that, Including: A preprocessing module for pre-establishing the association relationship between the risk identification library and the loan risk early warning model; wherein, each risk identification library includes multiple risk identification factors, and the verification directions of each risk identification factor are different; A risk identification factor determination module for determining the risk identification factors associated with the early warning task according to the association relationship between the risk identification library and the loan risk early warning model; wherein, the early warning task includes multiple doubtful loans output by the same loan risk early warning model; A verification module for verifying the early warning task by using the risk identification factors associated with the early warning task; Among them, pre-establishing the association relationship between the risk identification library and the loan risk early warning model includes: Verify each historical loan under each loan product according to each risk identification factor, and according to the verification results, determine the risk identification factor combination of each historical loan; According to the historical doubtful loans output by the loan risk early warning model and the risk identification factor combination of each historical loan, determine the risk identification factor combination of the historical doubtful loans; According to the risk identification factor combination of the historical doubtful loans, count the number of risk identification factors triggered by the same risk early warning model under the same risk identification library; If the number of risk identification factors triggered by a risk early warning model under a risk identification library accounts for a first predetermined value of the total number of risk identification factors under the risk identification library, then associate the risk identification library with the risk early warning model.

13. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the doubtful loan verification method described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an executable computer program, and when the computer program is executed by a processor, it implements the doubtful loan verification method described in any one of claims 1 to 11.

Images