A method, device, electronic device and storage medium for determining risky loans

Abstract

The embodiment of the present invention discloses a method, device, electronic device and storage medium for determining risky loans. The method comprises: obtaining loan information, and determining a borrower group and a fund supplier group according to the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds; obtaining characteristic data of the borrower group and the fund supplier group, and inputting the characteristic data into a pre-constructed risk judgment model to obtain candidate risky loans; screening the candidate risky loans according to preset rules to determine the target risky loans. By running the technical solution provided by the embodiment of the present invention, it is possible to solve the problem that the scope and time point of fund transactions between individuals are screened according to preset rules, and the risk warning is issued after manual analysis of the suspicious loans at each time point screened, which is time-consuming and labor-intensive, and the accuracy is difficult to measure, and there is a high risk of manual intervention, which is not conducive to risk management and control, and the efficiency and accuracy of risk loan estimation are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to computer technology, and more particularly to a method, device, electronic device and storage medium for determining risky loans. Background Art

[0002] Personal housing loan business is one of the main asset businesses of commercial banks. It refers to the loans that commercial banks provide to borrowers for the purpose of purchasing first-time housing (i.e. housing developed and constructed by real estate developers or other qualified development entities and then sold to individuals). However, in recent years, there have been cases of loan fraud, loan arbitrage, or non-compliant loan use or suspicious sources of down payments.

[0003] In the prior art, the estimation of risky loans is based on the scope and time points of personal financial transactions, which are screened according to preset rules. The suspicious loans at each time point are manually analyzed and risk warnings are issued. This method is time-consuming and labor-intensive, and its accuracy is difficult to measure. It brings a lot of unnecessary labor to the staff, and there is a high risk of manual intervention, which is not conducive to risk management. Summary of the invention

[0004] The embodiments of the present invention provide a risky loan determination method, device, electronic device and storage medium to achieve the effect of improving the efficiency and accuracy of risky loan estimation.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining risky loans, the method comprising:

[0006] Obtaining loan information, and determining a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds;

[0007] Acquire characteristic data of the borrower group and the fund supplier group, and input the characteristic data into a pre-built risk judgment model to obtain candidate risk loans;

[0008] The candidate risky loans are screened according to preset rules to determine target risky loans.

[0009] In a second aspect, an embodiment of the present invention further provides a risk loan device, the device comprising:

[0010] A loan information acquisition module, used to acquire loan information and determine a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds;

[0011] A feature data acquisition module, used to acquire feature data of the borrower group and the fund supplier group, and input the feature data into a pre-built risk judgment model to acquire candidate risk loans;

[0012] The loan determination module is used to screen the candidate risk loans according to preset rules to determine the target risk loans.

[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0014] one or more processors;

[0015] a storage device for storing one or more programs,

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the risk loan determination method as described above.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the risky loan determination method as described above.

[0018] The embodiment of the present invention obtains loan information and determines a borrower group and a fund supplier group according to the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds; obtains characteristic data of the borrower group and the fund supplier group, and inputs the characteristic data into a pre-constructed risk judgment model to obtain candidate risk loans; and screens the candidate risk loans according to preset rules to determine target risk loans. The problem that the scope and time point of fund transactions between individuals are screened according to preset rules, and risk warnings are issued after manual analysis of suspicious loans at each time point screened out is time-consuming and labor-intensive, and the accuracy is difficult to measure, and there is a high risk of manual intervention, which is not conducive to risk management, is solved, and the efficiency and accuracy of risk loan estimation are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a method for determining risky loans provided in Embodiment 1 of the present invention;

[0020] Figure 2 A flow chart of a method for determining risky loans provided in Embodiment 2 of the present invention;

[0021] Figure 3 A schematic diagram of the structure of a risk loan device provided in Embodiment 3 of the present invention;

[0022] Figure 4 A schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0023] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0024] Embodiment 1

[0025] Figure 1 This is a flow chart of a risky loan determination method provided in the first embodiment of the present invention. This embodiment is applicable to the case of risk prediction of loans. The method can be executed by the risky loan determination device provided in the embodiment of the present invention. The device can be implemented in software and / or hardware. Figure 1 The risky loan determination method provided in this embodiment includes:

[0026] Step 110, obtaining loan information, and determining a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group composed of persons associated with the borrower, and the fund supplier group is a group composed of persons associated with the borrower's down payment funds.

[0027] Among them, loan information refers to all information related to the loan, which can be divided into internal information and external information. For example, when the loan is a housing loan, the loan information can be information related to the borrower's down payment. Internal information can include the borrower's personal information, credit information, credit card business information, relationship information, credit information, etc. that can be obtained within the bank. External information can include intermediary information and broker information from real estate big data related to the purpose of the loan.

[0028] A borrower gang is a group of people associated with the borrower, including, for example, the borrower's spouse, co-borrowers, mortgagors, etc.

[0029] The fund supplier group is a group of people associated with the borrower's down payment funds, that is, the objects that may provide the borrower with down payment funds. Examples include real estate developers, real estate agents, etc.

[0030] It should be noted that there may be overlap among the members of the above-mentioned gangs. For example, the spouse of the borrower may also be a real estate agent.

[0031] According to the loan information, for each loan that needs to be analyzed, the corresponding borrower group and fund supplier group are obtained.

[0032] In this embodiment, optionally, determining the borrower group and the fund supplier group according to the loan information includes:

[0033] Based on knowledge graph technology, obtain the borrower set and fund supplier set;

[0034] Score the intimacy of all members of the borrower set and the fund supplier set respectively; wherein the intimacy of the borrower set refers to the intimacy between the members of the borrower set and the borrowers, and the intimacy of the fund supplier set refers to the intimacy between the members of the fund supplier set and the entity to which the fund supplier belongs;

[0035] According to the intimacy scores, the members constituting the borrower group and the fund supplier group are respectively determined from the borrower group and the fund supplier group.

[0036] Among them, the knowledge graph is essentially a semantic network, a graph-based data structure consisting of nodes and edges. In the knowledge graph, each node represents an "entity" that exists in the real world, and each edge is a "relationship" between entities. The knowledge graph is a relationship network obtained by connecting all different types of information together, providing the ability to analyze problems from the perspective of "relationships." Through the knowledge graph technology, combined with the obtained loan information, the borrower set and the fund supplier set are obtained. The borrower set is a set of all objects associated with the borrower, and the fund supplier set is a set of all objects associated with the borrower's down payment funds.

[0037] Score the intimacy of all members in the borrower set and the fund supplier set; wherein, the intimacy of the borrower set is the intimacy between the members of the borrower set and the borrower, and the intimacy of the fund supplier set is the intimacy between the members of the fund supplier set and the entity where the fund supplier is located. The entity where the fund supplier is located is the company or enterprise where the fund supplier is located. The intimacy between each member of the borrower set and the fund supplier set and the borrower can be scored based on complex network relationships and graph database technology. According to the intimacy score, the borrower set and the fund supplier set are screened to determine the members that constitute the borrower group and the fund supplier group. Exemplarily, the top 50 objects with the intimacy score in the borrower set and the fund supplier set are selected as the members that constitute the borrower group and the fund supplier group. That is, the borrower group includes 50 members, and the fund supplier group includes 50 members. This is to avoid obtaining too many objects that are not closely related to the borrower, increase the workload of subsequent processing, and thus improve the efficiency of estimation. It should be noted that when the number of members in a borrower group and a fund supplier group is less than 50, they can be handled according to the actual number.

[0038] Step 120: Obtain characteristic data of the borrower group and the fund supplier group, and input the characteristic data into a pre-built risk judgment model to obtain candidate risk loans.

[0039] Among them, the characteristic data is the data related to the loan of the borrower group and the fund supplier group that may affect the risk estimation of the loan. Exemplarily, the characteristic data may include data related to the loan: transaction time, transaction amount, counterparty, etc.; basic data of members: occupation, education, nature of work, annual income, asset management scale value, etc.; member credit data: social security information, tax data, etc. The characteristic data is input into the risk judgment model to determine whether the current loan may be a risky loan. If the judgment result is yes, the current loan can be used as a candidate risk loan and added to the suspicious loan pool for further analysis of the candidate risk loans. Among them, the risk judgment model can be any model that can be used for risk estimation. Exemplarily, it can be a decision tree model. The model prediction effect can be evaluated by cross-using indicators such as receiver operating characteristics and Gini coefficient to optimize the construction of the decision tree model, thereby improving the accuracy and efficiency of the risk prediction of the decision tree model.

[0040] In this embodiment, optionally, obtaining characteristic data of the borrower group and the fund supplier group includes:

[0041] Obtaining the borrower group data and the fund supplier group data;

[0042] Performing data cleaning on the borrower group data and the fund supplier group data; wherein the data cleaning includes: filling missing data, processing outliers and deleting duplicate data;

[0043] Determine characteristic data from the cleaned borrower group data and fund supplier group data.

[0044] The borrower group data and fund supplier group data are obtained by collecting data from each member of the borrower group and fund supplier group, including specific data of each member, such as personal data (such as gender, name, age, etc.), account data including specific amounts, transaction data, etc.

[0045] Data cleaning can be performed on the borrower group data and the fund supplier group data. Duplicate values ​​in the collected data can be eliminated according to business rules, abnormal values ​​can be replaced with average values, and missing values ​​can be supplemented based on historical data.

[0046] Determine the characteristic data from the cleaned data. The characteristic data can be determined by assigning different weight values ​​to different characteristic data. The weight value is also called IV value, which is an indicator value to measure the importance of characteristic data. Through repeated iterations, select the data with greater impact on loan risk as characteristic data. By acquiring and cleaning the data, the accuracy and completeness of characteristic data acquisition can be improved, providing a data basis for the construction of risk judgment models and actual operations.

[0047] Step 130: Screen the candidate risky loans according to preset rules to determine target risky loans.

[0048] The preset rules may be rules obtained based on expert experience, which is not limited in this embodiment. After the candidate risky loans are screened according to the preset rules, the screened loans may be manually reviewed to determine whether they are risky loans. If they are determined to be risky loans, the risky loans may be recovered in advance to avoid possible losses caused by the risky loans.

[0049] The technical solution provided in this embodiment obtains loan information and determines the borrower group and the fund supplier group based on the loan information; wherein the borrower group is a group of people associated with the borrower, and the fund supplier group is a group of people associated with the borrower's down payment funds; obtains characteristic data of the borrower group and the fund supplier group, and inputs the characteristic data into a pre-constructed risk judgment model to obtain candidate risk loans; and screens the candidate risk loans according to preset rules to determine the target risk loans. This solves the problem that the scope and time point of fund transactions between individuals are screened according to preset rules, and the risk warning is issued after manual analysis of the suspicious loans at each time point that are screened, which is time-consuming and labor-intensive, and the accuracy is difficult to measure, and there is a high risk of manual intervention, which is not conducive to risk management and control, so as to achieve the effect of improving the efficiency and accuracy of risk loan estimation.

[0050] Embodiment 2

[0051] Figure 2 This is a flowchart of a method for determining risky loans provided in Embodiment 2 of the present invention. This technical solution is to supplement the process of screening candidate risky loans according to preset rules. Compared with the above solution, this solution is specifically optimized to screen the candidate risky loans according to preset rules to determine the target risky loans, including:

[0052] Obtaining transaction flow associated with the candidate risky loan within a preset period;

[0053] The transaction flow is screened according to preset rules to determine the target risk loans. Specifically, the flow chart of the data interaction method is as follows: Figure 2 As shown:

[0054] Step 210: Obtain loan information, and determine the borrower group and the fund supplier group based on the loan information.

[0055] Among them, the borrower group is a group composed of people associated with the borrowers, and the fund supplier group is a group composed of people associated with the borrowers' down payment funds.

[0056] Step 220: Obtain characteristic data of the borrower group and the fund supplier group, and input the characteristic data into a pre-built risk judgment model to obtain candidate risk loans.

[0057] Step 230: Obtain transaction flow associated with the candidate risky loan within a preset period.

[0058] The transaction flows associated with the candidate risk loans are the transaction flows associated with the borrowers of the candidate risk loans, and illustratively, may be all the transaction flows in which funds are transferred to the borrower group within one month before the loan application date.

[0059] Step 240: Screen the transaction flow according to preset rules to determine target risk loans.

[0060] The acquired transaction flow is screened according to preset rules, and the problematic transaction flow screened out can be submitted for manual review. If it is determined that there is a problem with the transaction flow, the candidate risk loan associated with the transaction flow is judged to be the target risk loan.

[0061] In this embodiment, optionally, the transaction flow is screened according to rules to determine target risk loans, including:

[0062] Screening out candidate transaction flows that meet the first preset rule;

[0063] Determining whether the candidate transaction flow is a problematic transaction flow;

[0064] If the candidate risk loan includes problematic transaction flow, the candidate risk loan is determined to be a target risk loan.

[0065] The first preset rule is used to filter out transaction flows that do not need to be processed later. For example, for a specific bank, it may include:

[0066] a. The transfer account names of transaction flows that are not real estate developers cooperating with the bank and their close associates will be excluded. The close associates can be obtained by scoring the intimacy.

[0067] b. The transaction records corresponding to the bank's external accounts, internal accounts, and personal accounts of members of the borrower gang shall be excluded;

[0068] c. Transaction flows containing the following keywords in the name of the transfer-out account shall be eliminated: "reserve fund" but not "real estate", personal loan system settlement, fund supervision, agency payment, agency collection, agency payment, temporarily suspended account, income account, withdrawal account, suspended account, transitional account, affordable housing, housing and urban and rural construction, housing provident fund, redemption account, pending processing, and payment shall be eliminated;

[0069] d. The total amount transferred to the borrower group within one month before the loan application date is less than 20,000 yuan;

[0070] e. The corresponding transaction flows of a single fund supplier group with less than 2 associated persons will be removed;

[0071] f. The corresponding transaction flow of a single fund supplier group transferring less than 2 funds to a borrower group will be eliminated;

[0072] g. Transaction flows corresponding to a single fund supplier group where the number of borrower groups corresponding to the group is less than 3 will be eliminated.

[0073] The transaction flow removed according to the first preset rule is used as a candidate transaction flow, and it is determined whether the candidate transaction flow is a problem transaction flow. If the transaction flow is determined to be a problem transaction flow, and the candidate risk loan includes any problem transaction flow, the candidate risk loan is determined to be a target risk loan. By screening the transaction flow according to the first preset rule, the transaction flow that cannot be processed or has a low possibility of risk is removed, so as to reduce the amount of data processing and improve the efficiency of subsequent problem transaction flow judgment.

[0074] In this embodiment, optionally, determining whether the candidate transaction flow is a problematic transaction flow includes:

[0075] If the fund supplier group and the borrower group associated with the candidate transaction flow have a preset relationship, determining whether the candidate transaction flow satisfies a second preset rule;

[0076] If the second preset rule is met, the candidate transaction flow is determined to be a problematic transaction flow.

[0077] The preset relationship is a possible relationship between the fund supplier group and the borrower group. For example, the developer corresponding to the fund supplier group is the developer of the real estate project purchased by the borrower group with the loan, that is, the purpose of the borrower's loan may be to purchase a house developed by the fund supplier's company.

[0078] If there is a preset relationship, it is determined whether the candidate transaction flow meets the second preset rule, wherein the second preset rule is used to further screen the candidate transaction flow. Exemplarily, the second preset rule may be a transaction flow in which the amount transferred from the fund supplier group to the borrower group / the amount of the borrower group's down payment is between 0.5 and 1.2, that is, assuming the down payment amount is 200,000, then the transaction flow in which the fund supplier group transfers an amount of 100,000 to 240,000 to the borrower group is a problematic transaction flow. If there is a preset relationship, it indicates that the possibility of risk has increased, and the candidate transaction flow is directly screened by the second preset rule to improve the efficiency and accuracy of obtaining problematic transaction flows.

[0079] In this embodiment, optionally, determining whether the candidate transaction flow is a problematic transaction flow includes:

[0080] If the fund supplier group and the borrower group associated with the candidate transaction flow do not have a preset relationship, screening the candidate transaction flow according to a third preset rule;

[0081] Determine whether the selected candidate transaction flow meets the second preset rule;

[0082] If so, the candidate transaction flow is determined to be a problematic transaction flow.

[0083] If there is no preset relationship between the fund supplier group and the borrower group associated with the candidate transaction flow, the candidate transaction flow is screened according to the third preset rule, wherein the third preset rule can be to screen the top 20 fund suppliers of each branch with the total transaction amount with the borrower group. If there are less than 20, the actual number is used. Then determine whether the screened candidate transaction flow meets the second preset rule. Exemplarily, obtain the transaction flow in which the amount transferred from the screened fund supplier group to the borrower group / the amount of the borrower group's down payment is between 0.5 and 1.2. If there is no preset relationship, it indicates that there may be a connection between the fund supplier group and the borrower group that is not easy to find directly. The candidate transaction flow is further analyzed through the third preset rule to obtain transaction flows that may be risky, so as to improve the accuracy of risk loan determination.

[0084] The embodiment of the present invention obtains loan information, and determines the borrower group and the fund supplier group according to the loan information; obtains the characteristic data of the borrower group and the fund supplier group, and inputs the characteristic data into a pre-constructed risk judgment model to obtain candidate risk loans; obtains the transaction flow associated with the candidate risk loan within a preset period; and screens the transaction flow according to preset rules to determine the target risk loan. The transaction flow associated with the candidate loan is analyzed according to the preset rules to predict the loan that may be at risk, thereby improving the efficiency and accuracy of determining the risk loan and avoiding the problems of lag and low accuracy in risk judgment.

[0085] Embodiment 3

[0086] Figure 3 This is a schematic diagram of the structure of a risk loan device provided in the third embodiment of the present invention. The device can be implemented by hardware and / or software, can execute a risk loan method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. Figure 3 As shown, the device comprises:

[0087] The loan information acquisition module 310 is used to acquire loan information and determine the borrower group and the fund supplier group according to the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds;

[0088] A feature data acquisition module 320 is used to acquire feature data of the borrower group and the fund supplier group, and input the feature data into a pre-built risk judgment model to obtain candidate risk loans;

[0089] The loan determination module 330 is used to screen the candidate risk loans according to preset rules to determine target risk loans.

[0090] The embodiment of the present invention obtains loan information and determines a borrower group and a fund supplier group according to the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds; obtains characteristic data of the borrower group and the fund supplier group, and inputs the characteristic data into a pre-constructed risk judgment model to obtain candidate risk loans; and screens the candidate risk loans according to preset rules to determine target risk loans. The problem that the scope and time point of fund transactions between individuals are screened according to preset rules, and risk warnings are issued after manual analysis of suspicious loans at each time point screened out is time-consuming and labor-intensive, and the accuracy is difficult to measure, and there is a high risk of manual intervention, which is not conducive to risk management, is solved, and the efficiency and accuracy of risk loan estimation are improved.

[0091] Based on the above technical solutions, optionally, the loan information acquisition module includes:

[0092] The collection acquisition submodule is used to acquire the borrower collection and the fund supplier collection based on the knowledge graph technology;

[0093] The intimacy scoring submodule is used to score the intimacy of all members in the borrower set and the fund supplier set respectively; wherein the intimacy of the borrower set is the intimacy between the members of the borrower set and the borrowers, and the intimacy of the fund supplier set is the intimacy between the members of the fund supplier set and the entity to which the fund supplier belongs;

[0094] The member determination submodule is used to determine the members constituting the borrower group and the fund supplier group from the borrower set and the fund supplier set respectively according to the intimacy score.

[0095] On the basis of the above technical solutions, optionally, the feature data acquisition module includes:

[0096] A data acquisition submodule, used to acquire the borrower group data and the fund supplier group data;

[0097] A data cleaning submodule is used to clean the borrower group data and the fund supplier group data; wherein the data cleaning includes: missing data filling, outlier processing and duplicate data deletion;

[0098] The characteristic data determination submodule is used to determine the characteristic data from the cleaned borrower group data and the fund supplier group data.

[0099] Based on the above technical solutions, optionally, the loan determination module includes:

[0100] A transaction flow acquisition submodule, used to acquire transaction flow associated with the candidate risky loan within a preset period;

[0101] The loan determination submodule is used to screen the transaction flow according to preset rules to determine the target risk loans.

[0102] Based on the above technical solutions, optionally, the loan determination submodule includes:

[0103] A transaction flow screening unit, used to screen out candidate transaction flows that meet a first preset rule;

[0104] A transaction flow determination unit, used to determine whether the candidate transaction flow is a problematic transaction flow;

[0105] The loan determination unit is configured to determine the candidate risk loan as a target risk loan if the candidate risk loan includes a problematic transaction flow.

[0106] On the basis of the above technical solutions, optionally, the transaction flow determination unit includes:

[0107] The first rule judgment subunit judges whether the candidate transaction flow satisfies a second preset rule if the fund supplier group and the borrower group associated with the candidate transaction flow have a preset relationship;

[0108] The first transaction flow determination subunit is used to determine that the candidate transaction flow is a problematic transaction flow if the second preset rule is met.

[0109] On the basis of the above technical solutions, optionally, the transaction flow determination unit includes:

[0110] a transaction flow screening subunit, configured to screen the candidate transaction flow according to a third preset rule if the fund supplier group and the borrower group associated with the candidate transaction flow do not have a preset relationship;

[0111] The second rule judgment subunit is used to judge whether the screened candidate transaction flow satisfies the second preset rule;

[0112] The second transaction flow determination subunit is used to determine the candidate transaction flow as a problematic transaction flow if the conditions are met.

[0113] Embodiment 4

[0114] Figure 4 A schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention is shown in FIG. Figure 4As shown, the electronic device includes a processor 40, a memory 41, an input device 42 and an output device 43; the number of processors 40 in the electronic device can be one or more. Figure 4 A processor 40 is taken as an example; the processor 40, the memory 41, the input device 42 and the output device 43 in the electronic device can be connected by a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0115] The memory 41 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the risky loan determination method in the embodiment of the present invention. The processor 40 executes various functional applications and data processing of the electronic device by running the software programs, instructions and modules stored in the memory 41, that is, implements the above-mentioned risky loan determination method.

[0116] The memory 41 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 41 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 41 may further include a memory remotely arranged relative to the processor 40, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0117] Embodiment 5

[0118] Embodiment 5 of the present invention further provides a storage medium including computer executable instructions, wherein the computer executable instructions are used to execute a risky loan determination method when executed by a computer processor, the method comprising:

[0119] Obtaining loan information, and determining a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds;

[0120] Acquire characteristic data of the borrower group and the fund supplier group, and input the characteristic data into a pre-built risk judgment model to obtain candidate risk loans;

[0121] The candidate risky loans are screened according to preset rules to determine target risky loans.

[0122] Of course, the computer executable instructions of a storage medium containing computer executable instructions provided by an embodiment of the present invention are not limited to the operations of the method described above, and can also execute related operations in the risk loan determination method provided by any embodiment of the present invention.

[0123] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0124] It is worth noting that in the above-mentioned embodiment of the risky loan determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0125] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining risky loans, characterized in that: include: Obtaining loan information, and determining a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds; Acquire characteristic data of the borrower group and the fund supplier group, and input the characteristic data into a pre-built risk judgment model to obtain candidate risk loans; Screening the candidate risk loans according to preset rules to determine target risk loans, and recovering the target risk loans in advance; Wherein, the step of determining the borrower group and the fund supplier group according to the loan information includes: Based on knowledge graph technology, obtain the borrower set and fund supplier set; The intimacy scores are respectively given to all members of the borrower set and the fund supplier set; wherein the intimacy of the borrower set is the intimacy between the members of the borrower set and the borrowers, and the intimacy of the fund supplier set is the intimacy between the members of the fund supplier set and the entity to which the fund supplier belongs; wherein the entity to which the fund supplier belongs is the company or enterprise to which the fund supplier belongs; wherein the intimacy scores are respectively given to all members of the borrower set and the fund supplier set based on complex network relationships and graph database technology; According to the intimacy scores, members constituting the borrower group and the fund supplier group are determined from the borrower group and the fund supplier group respectively; The step of screening the candidate risk loans according to preset rules to determine target risk loans includes: Obtaining transaction flow associated with the candidate risky loan within a preset period; Screening the transaction flow according to preset rules to determine target risk loans; The step of screening the transaction flow according to preset rules to determine target risk loans includes: Screening out candidate transaction flows that meet the first preset rule; Determining whether the candidate transaction flow is a problematic transaction flow; If the candidate risk loan includes problematic transaction flow, determining the candidate risk loan as a target risk loan; The step of determining whether the candidate transaction flow is a problematic transaction flow includes: If the fund supplier group and the borrower group associated with the candidate transaction flow have a preset relationship, determining whether the candidate transaction flow satisfies a second preset rule; If the second preset rule is met, determining the candidate transaction flow as a problematic transaction flow; The step of determining whether the candidate transaction flow is a problematic transaction flow includes: If the fund supplier group and the borrower group associated with the candidate transaction flow do not have a preset relationship, screening the candidate transaction flow according to a third preset rule; Determine whether the selected candidate transaction flow meets the second preset rule; If so, the candidate transaction flow is determined to be a problematic transaction flow.

2. The method according to claim 1, characterized in that Obtaining characteristic data of the borrower group and the fund supplier group, including: Obtaining the borrower group data and the fund supplier group data; Performing data cleaning on the borrower group data and the fund supplier group data; wherein the data cleaning includes: filling missing data, processing outliers and deleting duplicate data; Determine characteristic data from the cleaned borrower group data and fund supplier group data.

3. A risk loan device, characterized in that: include: A loan information acquisition module, used to acquire loan information and determine a borrower group and a fund supplier group based on the loan information; wherein the borrower group is a group of persons associated with the borrower, and the fund supplier group is a group of persons associated with the borrower's down payment funds; A feature data acquisition module, used to acquire feature data of the borrower group and the fund supplier group, and input the feature data into a pre-built risk judgment model to acquire candidate risk loans; A loan determination module, used to screen the candidate risk loans according to preset rules to determine target risk loans, and to recover the target risk loans in advance; Wherein, the loan information acquisition module includes: The collection acquisition submodule is used to acquire the borrower collection and the fund supplier collection based on the knowledge graph technology; The intimacy scoring submodule is used to score the intimacy of all members of the borrower set and the fund supplier set respectively; wherein the intimacy of the borrower set is the intimacy between the members of the borrower set and the borrowers, and the intimacy of the fund supplier set is the intimacy between the members of the fund supplier set and the entity where the fund supplier is located; wherein the entity where the fund supplier is located is the company or enterprise where the fund supplier is located; wherein the intimacy scoring of all members of the borrower set and the fund supplier set is realized based on complex network relationships and graph database technology; A member determination submodule, for determining members constituting a borrower group and a fund supplier group from the borrower group and the fund supplier group respectively according to the intimacy score; Wherein, the loan determination module includes: A transaction flow acquisition submodule, used to acquire transaction flow associated with the candidate risky loan within a preset period; The loan determination submodule is used to screen the transaction flow according to preset rules to determine the target risk loan; Wherein, the loan determination submodule includes: A transaction flow screening unit, used to screen out candidate transaction flows that meet a first preset rule; A transaction flow determination unit, used to determine whether the candidate transaction flow is a problematic transaction flow; A loan determination unit, configured to determine the candidate risk loan as a target risk loan if the candidate risk loan includes a problematic transaction flow; Wherein, the flow determination unit comprises: A first rule judgment subunit is used to judge whether the candidate transaction flow satisfies a second preset rule if the fund supplier group and the borrower group associated with the candidate transaction flow have a preset relationship; A first transaction flow determination subunit, configured to determine the candidate transaction flow as a problematic transaction flow if the second preset rule is satisfied; Wherein, the flow determination unit comprises: a transaction flow screening subunit, configured to screen the candidate transaction flow according to a third preset rule if the fund supplier group and the borrower group associated with the candidate transaction flow do not have a preset relationship; The second rule judgment subunit is used to judge whether the screened candidate transaction flow satisfies the second preset rule; The second transaction flow determination subunit is used to determine the candidate transaction flow as a problematic transaction flow if the conditions are met.

4. An electronic device, characterized in that: The electronic device comprises: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the risk loan determination method as described in any one of claims 1-2.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the risky loan determination method as described in any one of claims 1-2 is implemented.

Images