Loan knowledge graph creation method, loan approval method and device
By forming a loan knowledge graph and linking historical loan information, the problem of low accuracy of fraud identification in vehicle loan approval is solved, the approval speed and the accuracy of risk assessment are improved, and the risk of fraud is reduced.
Patent Information
- Application Number
- CN202410823434.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-25
- Publication Date
- 2024-11-05
AI Technical Summary
The failure to correctly identify fraud during vehicle loan approval process leads to low identification accuracy and inability to comprehensively analyze, which increases the risks and losses of financial lending institutions.
By obtaining historical loan application information, historical loan approval information and post-loan monitoring information, a loan knowledge graph is formed, and different historical loan information are collected using preset relationships to form a graph to facilitate viewing of information on various stages of the loan application life cycle, and used for new loans Approval decisions for application.
It improves the speed of loan approval, reduces the time for manual search, provides more accurate risk assessment, reduces the possibility of wrong judgments and fraudulent behavior, and enhances the ability to identify fraudulent behavior.
Smart Images

Figure CN118898291A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of vehicle financial loan technology, and in particular, relates to a loan knowledge graph creation method, a loan approval method and a device. Background Art
[0002] As a common consumer credit product, vehicle loans provide convenience to a certain extent, allowing consumers to purchase vehicles that they cannot pay for in one lump sum and repay them gradually through installments.
[0003] However, the failure to correctly identify fraud in the vehicle loan approval process is an issue that cannot be ignored. Currently, fraud identification in the vehicle loan approval process mainly relies on manual and expert rules. This identification method has certain limitations, such as low identification accuracy and inability to conduct comprehensive analysis, which brings higher risks and losses to financial lending institutions. Summary of the invention
[0004] The embodiments of the present application provide a loan knowledge graph creation method, a loan approval method and a device, which can solve the problem that the fraud identification means in the current vehicle loan approval process rely on manual and expert rules for identification, which has low recognition accuracy and cannot be comprehensively analyzed.
[0005] In a first aspect, an embodiment of the present application provides a method for creating a loan knowledge graph, including:
[0006] Obtain all historical loan information, including historical loan application information, historical loan approval information and post-loan monitoring information;
[0007] Associating the historical loan application information, the historical loan approval information and the post-loan monitoring information in the same historical loan information;
[0008] According to a preset relationship set, different historical loan application information is associated and different post-loan monitoring information is associated to form a loan knowledge graph.
[0009] In a possible implementation of the first aspect, associating different historical loan application information according to a preset relationship set includes:
[0010] If the application data in the first historical loan application information and the application data in the second historical loan application information meet the first preset relationship in the preset relationship set, the application data in the first historical loan application information is used as the first data node, and the application data in the second historical loan application information is used as the second data node; the first historical loan application information is any one of the multiple historical loan application information, and the second historical loan application information is any one of the multiple historical loan application information except the first historical loan application information;
[0011] The first data node and the second data node are connected through a first node connection line, and the first preset relationship is added to the first node connection line.
[0012] In a possible implementation of the first aspect, associating different post-loan monitoring information according to a preset relationship set includes:
[0013] If the post-loan data in the first post-loan monitoring information and the post-loan data in the second post-loan monitoring information meet the second preset relationship in the preset relationship set, the post-loan data in the first post-loan monitoring information is used as the third data node, and the post-loan data in the second post-loan monitoring information is used as the fourth data node; the first post-loan monitoring information is any one of the multiple post-loan monitoring information, and the second post-loan monitoring information is any one of the multiple post-loan monitoring information except the first post-loan monitoring information;
[0014] The third data node and the fourth data node are connected through a second node connection line, and the second preset relationship is added to the second node connection line.
[0015] In a possible implementation manner of the first aspect, the preset relationship in the preset relationship set includes at least one of an ID number relationship, a mobile phone number relationship, an address relationship, a personnel relationship, and an information input device relationship.
[0016] In a second aspect, the present application embodiment provides a loan approval method, including:
[0017] Obtain new loan application information;
[0018] Searching for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information in a loan knowledge graph; the loan knowledge graph is created by the loan knowledge graph creation method described in any one of the first aspects above;
[0019] A correlation graph between the new loan application information and the target historical loan information is displayed.
[0020] In a possible implementation manner of the second aspect, the loan approval method further includes:
[0021] Determining the risk level of the new loan application information based on the target historical loan information;
[0022] Displays the stated risk level.
[0023] In a possible implementation manner of the second aspect, the loan approval method further includes:
[0024] When the new loan application information is approved, the new loan application information, the new loan approval information and the new post-loan monitoring information in the new loan information are associated;
[0025] According to the preset relationship set, the new loan application information is associated with the historical loan application information and the new post-loan monitoring information is associated with the post-loan monitoring information to update the loan knowledge graph.
[0026] In a third aspect, the present application embodiment provides a loan knowledge graph creation device, including:
[0027] A first acquisition module is used to acquire all historical loan information, wherein the historical loan information includes historical loan application information, historical loan approval information and post-loan monitoring information;
[0028] An association module, used for associating the historical loan application information, the historical loan approval information and the post-loan monitoring information in the same historical loan information;
[0029] A formation module is used to associate different historical loan application information and different post-loan monitoring information according to a preset relationship set to form a loan knowledge graph.
[0030] In a fourth aspect, an embodiment of the present application provides a loan approval device, including:
[0031] The second acquisition module is used to acquire new loan application information;
[0032] A search module, used to search in a loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information; the loan knowledge graph is created by the loan knowledge graph creation method described in any one of the first aspects above;
[0033] A display module is used to display a correlation map between the new loan application information and the target historical loan information.
[0034] In the fifth aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the loan knowledge graph creation method as described in any one of the first aspect above is implemented, or, when the processor executes the computer program, the loan approval method as described in any one of the second aspect above is implemented.
[0035] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0036] The loan knowledge graph creation method provided in the first aspect above can integrate data from multiple data sources related to the field of vehicle loans by acquiring all historical loan information including historical loan application information, historical loan approval information and post-loan monitoring information, so that financial loan institutions can obtain comprehensive data about vehicle loans in the entire process of application, approval and post-loan monitoring, by associating historical loan application information, historical loan approval information and post-loan monitoring information in the same historical loan information; and associating different historical loan application information and different post-loan monitoring information according to a preset relationship set, thereby forming a loan knowledge graph. The loan knowledge graph creation method provided in the embodiment of the present application can obtain a loan knowledge graph containing all historical loan information, and then use the loan knowledge graph to more clearly view the information at each stage in the loan application life cycle, providing a decision-making basis for the approval of subsequent new loan applications.
[0037] The loan approval method provided in the second aspect above, when approving a new loan application, first, obtains the new loan application information, and secondly, the loan approval personnel can search for historical loan information associated with the new loan application in the loan knowledge graph created by the first aspect, that is, search the loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information, and finally, displays the association graph between the new loan application information and the target historical loan information. The loan approval method provided in the embodiment of the present application can quickly find the historical loan information related to the new loan application, reducing the time for manual search, thereby speeding up the loan approval speed; displaying and analyzing the association graph between the new loan application information and the historical loan information can more conveniently provide financial lending institutions with associated data related to the approval of new loan applications, helping financial lending institutions to more accurately assess the risk level of new loan applications, reduce the possibility of misjudgment, and avoid the occurrence of fraudulent loan behaviors.
[0038] It can be understood that the beneficial effects of the above-mentioned third aspect can be referred to the relevant description in the above-mentioned first aspect, the beneficial effects of the above-mentioned fourth aspect can be referred to the relevant description in the above-mentioned second aspect, and the beneficial effects of the above-mentioned fifth aspect can be referred to the relevant description in the above-mentioned first aspect or the above-mentioned second aspect, and they will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0040] Figure 1 It is a flowchart of a method for creating a loan knowledge graph provided in an embodiment of the present application;
[0041] Figure 2 It is a flowchart of a method for creating a loan knowledge graph provided by another embodiment of the present application;
[0042] Figure 3 It is a schematic diagram of the association between historical loan application information whose preset relationship is the certificate number relationship provided by an embodiment of the present application;
[0043] Figure 4 It is a schematic diagram of the association between historical loan application information with the same device fingerprint provided by an embodiment of the present application;
[0044] Figure 5 It is a schematic diagram of an association between historical loan application information with the same remote contracting address provided by an embodiment of the present application;
[0045] Figure 6 It is a schematic diagram of the association between historical loan application information with the same residence / work address provided by an embodiment of the present application;
[0046] Figure 7 It is a flowchart of a method for creating a loan knowledge graph provided by another embodiment of the present application;
[0047] Figure 8 It is a schematic diagram of the association between historical loan application information whose preset relationship is a change of mobile phone number provided by an embodiment of the present application;
[0048] Fig. 9 is a schematic diagram of the association between overdue loans and historical loan application information provided by an embodiment of the present application;
[0049] Fig.10is a schematic diagram of the association between fraudulent loans and historical loan application information provided by an embodiment of the present application;
[0050] Fig.11 It is a flowchart of a loan approval method provided by an embodiment of the present application;
[0051] Fig.12 It is a flowchart of another loan approval method provided by an embodiment of the present application;
[0052] Fig.13 This is a flowchart of another loan approval method provided by an embodiment of the present application;
[0053] Fig.14 It is a structural schematic diagram of a loan knowledge graph creation device provided in an embodiment of the present application;
[0054] Fig.15 It is a structural diagram of a loan approval device provided in an embodiment of the present application;
[0055] Fig.16 It is a structural diagram of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0056] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0057] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.
[0058] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0059] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.
[0060] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0061] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.
[0062] With the popularization of the Internet, the Internet of Things and big data, the financial ecology and structure are undergoing profound changes. In order to enhance competitiveness, reduce operating costs and improve customer satisfaction, financial lending institutions are gradually transforming to Internet business. However, the auto finance industry, with its characteristics of "high single credit limit and high commodity circulation value", has attracted much attention from the black industry and has become an important target for fraud gangs.
[0063] Fraud gangs take advantage of the data loopholes in the current vehicle loan approval process that do not combine data sources distributed in multiple different systems to launch large-scale, multi-platform concentrated attacks on vehicle finance loan companies. For example, fraud gangs may take advantage of the lack of real-time data sharing between different systems to repeatedly submit loan applications on different platforms. Since the systems cannot verify multiple applications from the same customer in a timely manner, multiple approvals may be made, allowing the fraud gangs to obtain multiple loans; or, in the absence of unified data verification, fraud gangs may submit false information (such as forged proof of income or false credit reports) in one of the systems, and other systems cannot cross-verify the authenticity of this information, thereby defrauding the loan amount; or, in the case of multiple systems failing to share identity authentication data, fraud gangs may use stolen identity information to apply for loans. Since each system operates independently, identity theft may not be detected in a timely manner.
[0064] Therefore, the failure to correctly identify fraud in the vehicle loan approval process is an issue that cannot be ignored. Currently, fraud identification in the vehicle loan approval process mainly relies on manual and expert rules. This identification method has certain limitations, such as low identification accuracy and inability to fully analyze fraud, which brings higher risks and losses to financial lending institutions.
[0065] Based on the above problems, an embodiment of the present application provides a method for creating a loan knowledge graph. By acquiring all historical loan information including historical loan application information, historical loan approval information and post-loan monitoring information, a variety of data sources related to the field of vehicle loans can be integrated, so that financial lending institutions can obtain comprehensive data about vehicle loans in the entire process of application, approval and post-loan monitoring. By associating historical loan application information, historical loan approval information and post-loan monitoring information in the same historical loan information; associating different historical loan application information and associating different post-loan monitoring information according to a preset relationship set to form a loan knowledge graph, the information of each stage in the loan application life cycle can be more clearly viewed through the display of the loan knowledge graph, providing a decision-making basis for subsequent new loan approvals.
[0066] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.
[0067] Figure 1 The flowchart of the loan knowledge graph creation method provided by an embodiment of the present application is shown. Figure 1 As shown, the loan knowledge graph creation method may include the following steps S101 to S103.
[0068] Step S101: Acquire all historical loan information, including historical loan application information, historical loan approval information and post-loan monitoring information.
[0069] Specifically, the field of vehicle finance loans includes a variety of data sources, which are distributed in multiple different systems, each of which is used to process specific types of data. For example, the customer management system is used to manage basic information data including the customer's name, contact information, address, ID number, etc.; the loan management system is used to process application information data including loan amount, term, interest rate, etc., as well as loan approval status information such as loan approval progress and approval results; the credit scoring system is used to process credit information data including credit scores and credit reports generated based on customer credit history; the payment system is used to process payment information data including customer repayment method, repayment amount, repayment time, etc.
[0070] In order to solve the data vulnerability defect of not combining data sources distributed in multiple different systems in the current vehicle loan approval process, the embodiment of the present application integrates and summarizes the data from multiple different systems including a customer management system, a loan management system, a credit scoring system, and a payment system, to form a data pool containing all historical loan information, wherein each piece of historical loan information includes historical loan application information, historical loan approval information, and post-loan monitoring information.
[0071] Step S102: Associating historical loan application information, historical loan approval information and post-loan monitoring information in the same historical loan information.
[0072] Specifically, the loan management system will generate a unique identifier for each loan application, such as a loan ID or a loan application order number, which will accompany the entire life cycle of the loan, including the application stage, approval stage, and post-loan monitoring stage. The unique identifier is stored in the data pool together with the customer's loan application information (such as personal information, loan amount, loan purpose, etc.). After all historical loan information is stored in the data pool, the historical loan application information, historical loan approval information, and post-loan monitoring information in the same historical loan information can be associated through the unique loan ID or the unique loan application order number.
[0073] By associating the data of different stages of vehicle loans (application stage, approval stage and post-loan monitoring stage), it can be ensured that the data of each historical loan information is complete and consistent, thus effectively avoiding the problem of data isolation or duplication and improving the accuracy of the data. At the same time, the complete historical loan information of the entire process from the application stage to the approval stage to the post-loan monitoring stage helps financial lending institutions to fully understand the risk status of each loan. By analyzing the complete historical loan information data, financial lending institutions can identify potential risk factors, such as loan repayment ability, default probability, etc., so as to take corresponding risk management measures in a timely manner, such as adjusting interest rates, early repayment requirements, etc.
[0074] Step S103: According to a preset relationship set, different historical loan application information is associated and different post-loan monitoring information is associated to form a loan knowledge graph.
[0075] Specifically, a knowledge graph is a graph structure that represents the relationships between entities. It uses a graph database as a storage engine to connect different types of information together to form a complex relational network.
[0076] In an embodiment of the present application, various preset relationships such as the same ID number, the same device fingerprint, the same mobile phone number, and similar addresses are sorted out according to the data pool, and the relationship between different historical loan information is established. This relationship information is stored in, for example, a neo4j graph database, and a loan knowledge graph is formed based on the relationship information stored in the neo4j graph database.
[0077] In some embodiments, the preset relationship in the preset relationship set includes at least one of a certificate number relationship, a mobile phone number relationship, an address relationship, a personnel relationship, and an information input device relationship.
[0078] Among them, when the preset relationship is a document number relationship with the same document number, two or more loan application information with the same document number will be associated. When the preset relationship is a mobile phone number relationship with the same mobile phone number or a changed mobile phone number, two or more loan application information with the same mobile phone number will be associated, or two or more loan application information related to the loan application with a changed mobile phone number will be associated. When the preset relationship is an address relationship with the same or similar longitude and latitude for remote signing, the same or similar home address, and the same or similar work address, two or more loan application information with the same or similar longitude and latitude for remote signing, the same or similar home address, and the same or similar work address will be associated. When the preset relationship is a personnel relationship of relatives (husband and wife, brothers and sisters, father and son or mother and son, etc.) and friends, two or more loan application information with relatives (husband and wife, brothers and sisters, father and son or mother and son, etc.) and friends will be associated. When the preset relationship is an information input device relationship with the same device fingerprint, two or more loan application information with the same device fingerprint will be associated.
[0079] In other words, different historical loan application information in the data pool can be associated with different preset relationships, that is, the information belonging to the same applicant in different historical loan application information can be integrated through the document number relationship. The loan application information using the same mobile phone number in different historical loan application information can be integrated through the mobile phone number relationship, or the change history of the applicant's contact information can be tracked so that the applicant can be contacted in time when needed. The impact of changes in the applicant's residence or business location on the loan risk can be analyzed through the address relationship. The relationship between the applicant and other personnel (such as guarantors, co-borrowers, etc.) can be revealed through the personnel relationship, so as to more comprehensively assess the loan risk. Through the information input device relationship, the source and input process of different loan history loan application information can be traced to ensure the accuracy and reliability of the data.
[0080] In some embodiments, Figure 2 A flowchart of a method for creating a loan knowledge graph provided by another embodiment of the present application. Figure 2As shown, step S103 includes step S1031 to step S1032.
[0081] Step S1031: If the application data in the first historical loan application information and the application data in the second historical loan application information meet the first preset relationship in the preset relationship set, the application data in the first historical loan application information is used as the first data node, and the application data in the second historical loan application information is used as the second data node.
[0082] The first historical loan application information is any one of the multiple historical loan application information, and the second historical loan application information is any one of the multiple historical loan application information except the first historical loan application information.
[0083] Specifically, taking the first preset relationship of the same ID number as an example, when there are two historical loan application information with the same ID number in the data pool, which meets the ID number relationship in the preset relationship set, one of the two historical loan application information with the same ID number is used as the first data node, and the other historical loan application information is used as the second data node. The historical loan application information in the data pool is traversed to find all historical loan application information in the data pool that meets the same preset relationship for the same ID number.
[0084] Step S1032: Connect the first data node and the second data node through a first node connection line, and add the first preset relationship to the first node connection line.
[0085] Specifically, every two pieces of historical loan information in the historical loan application information in the data pool that conform to the same preset relationship, such as the ID number relationship, are used as the first data node and the second data node, respectively, and the first data node and the second data node are connected through the first node connection line, and the preset relationship with the same ID number is added to the first node connection line. Figure 3 As shown, there are currently 4 historical loan application information. Under each historical loan application information, the displayed information includes the name of the main borrower and the last four digits of the loan application order number. The 4 historical loan application information are "Zhang San-8616", "Li Si-8613", "Chen Ming-8607" and "Wang Hong-8608", among which "Zhang San-8616" is connected with "Li Si-8613" through the same ID number "111" of the main borrower; "Zhang San-8616" is connected with "Chen Ming-8607" through the same ID number "222" as brothers and sisters; "Zhang San-8616" is connected with "Wang Hong-8608" through the same ID number "333" as husband and wife.
[0086] In another example, every two pieces of historical loan information in the historical loan application information in the data pool of the device relationship that match the same preset relationship, such as the same device fingerprint, are respectively used as the first data node and the second data node, and the first data node and the second data node are connected through the first node connection line, and the preset relationship with the same device fingerprint is added to the first node connection line. Figure 4 As shown, there are currently three historical loan application information, namely "Bao Meina 5694", "Li Jieli 5762" and "Che Shang 5703". Among them, "Bao Meina 5694", "Li Jieli 5762" and "Che Shang 5703" are all connected through the main device fingerprint.
[0087] In another example, every two pieces of historical loan information in the data pool of historical loan application information that meet the same preset relationship, such as the same remote contract address, are used as the first data node and the second data node, respectively, and the first data node and the second data node are connected through the first node connection line, and the preset relationship of the same remote contract address is added to the first node connection line. Figure 5 As shown, there are currently 4 historical loan application information, namely "Bao Meina 6353", "Li Jieli 5762", "Liu Wei 6291" and "Quanpin Distance Optimization 6321", among which "Bao Meina 6353", "Li Jieli 5762", "Liu Wei 6291" and "Quanpin Distance Optimization 6321" are all connected through the main applet.
[0088] In another example, every two pieces of historical loan information in the data pool of historical loan application information that meet the same preset relationship, such as the same residential / work address, are used as the first data node and the second data node, respectively, and the first data node and the second data node are connected through a first node connection line, and the preset relationship of the same residential / work address is added to the first node connection line. Figure 6 As shown, there are currently 7 historical loan application information, namely "Asset Change 1 9935", "Zhong Jing 9951", "Asset Change 3 9947", "Asset Change 1 9929", "Asset Change 2 9931" and "Criminal Case 1 0046", among which the 7 historical loan application information are all connected through the same residential / unit address.
[0089] In addition, the method of associating multiple historical loan application information with preset relationships, such as address relationships with the same household registration address and mobile phone number relationships with the same mobile phone number, is similar to the corresponding association method with the same remote signing address and the same residential / unit address mentioned above, and will not be repeated here.
[0090] Through this graphical display method, you can intuitively see the association between the ID number, device fingerprint, remote signing address, residential / work address, household registration address, etc. between each historical loan application information, which helps loan approval personnel identify potential duplicate applications or association risks. In addition, this display method can also help loan approval personnel quickly understand the relationship between each historical loan application information and make corresponding evaluations and decisions.
[0091] In some embodiments, Figure 7 A schematic diagram of a loan knowledge graph creation method provided by another embodiment of the present application is shown. Figure 7 As shown, step S103 includes step S1033 to step S1034.
[0092] Step S1033: If the post-loan data in the first post-loan monitoring information and the post-loan data in the second post-loan monitoring information meet the second preset relationship in the preset relationship set, the post-loan data in the first post-loan monitoring information is used as the third data node, and the post-loan data in the second post-loan monitoring information is used as the fourth data node.
[0093] The first post-loan monitoring information is any one of the multiple post-loan monitoring information, and the second post-loan monitoring information is any one of the multiple post-loan monitoring information except the first post-loan monitoring information.
[0094] Taking the post-loan monitoring stage, the second preset relationship is the change of mobile phone number as an example. When the mobile phone number changes during the post-loan monitoring process, one of the other two historical loan application information related to the historical loan application information with the changed mobile phone number is used as the third data node, and the other historical loan application information is used as the fourth data node. The historical loan application information in the data pool is traversed to find all historical loan application information related to the historical loan application information with the changed mobile phone number in the data pool.
[0095] Step S1034: Connect the third data node and the fourth data node through the second node connection line, and add the second preset relationship to the second node connection line.
[0096] Specifically, every two pieces of historical loan information in the data pool that meet the same preset relationship, such as the change of mobile phone number, are used as the third data node and the fourth data node respectively, and the third data node and the fourth data node are connected through the second node connection line, and the preset relationship of the change of mobile phone number is added to the second node connection line. Figure 8As shown, during the post-loan monitoring stage, it was found that the mobile phone number of the principal borrower "Feiran 0098" was changed, and the changed mobile phone number was associated with the mobile phone numbers in other historical loan application information in the data pool. In this case, a certain relationship between other historical loan application information associated with the changed mobile phone number of "Feiran-0098" will be marked on the second node connection line, such as "Brother-Sister-Guo Jun" and "Brother-Sister-Qian Xia", which means that the relationship is brother and sister.
[0097] Through this graphical display method, you can intuitively see the relationship between various post-loan monitoring information, for example, historical loan applications associated with changes in mobile phone numbers, or by associating repayment records with loan contract information, you can monitor loan repayments and promptly discover overdue or default risks. By associating changes in the applicant's credit score with loan approval information, you can assess whether the applicant's credit status has changed significantly, which helps post-loan monitoring personnel quickly identify potential risks and associations, improve monitoring efficiency and accuracy, and take corresponding risk management measures.
[0098] Specifically, the loan knowledge graph obtained according to the above loan knowledge graph creation method can be displayed on the visualization interface through filtering, clicking, dragging and other operations, and the loan knowledge graph can be jumped to the detailed information of the specified historical loan information for viewing by double-clicking. When displaying the loan knowledge graph, the visualization interface provides a corresponding icon for different situations such as address duplicate checking, ID card duplicate checking, mobile phone number duplicate checking, device fingerprint duplicate checking, current order, unresolved / revoked, approval rejected, approved or cancelled, unknown order status, overdue, fraud, etc. The icon is displayed next to the corresponding loan application, so that the status of the loan application is displayed more clearly.
[0099] It should be noted that if an overdue situation is found in the post-loan monitoring information, the overdue icon will be displayed next to the corresponding loan application, such as Fig. 9 As shown, the square icon next to the loan application "Wu Wanting 5664" has "18" written in it, which means that the repayment of the loan application "Wu Wanting 5664" has been overdue for 18 days. Therefore, "Wu Wanting 56642", "Wu Wanting 5660" and "Xing Xiaohe 9844" in the post-loan monitoring related to the loan application "Wu Wanting 5664" also need to be judged whether there is a risk of overdue. If it is judged that "Wu Wanting 56642", "Wu Wanting 5660" and "Xing Xiaohe 9844" are also overdue, the square icon and its overdue days will be displayed next to the overdue loan application. If it is judged that "Wu Wanting 56642", "Wu Wanting 5660" and "Xing Xiaohe 9844" are about to be overdue, the loan applicants who are about to be overdue will be urged to repay as soon as possible.
[0100] If fraud is found in the post-loan monitoring information, a fraud icon will be displayed next to the corresponding loan application, such as Fig.10 As shown in the figure, the triangle icon next to the loan application "Baomeina 5663" indicates that the loan application "Baomeina 5663" is fraudulent. Then other loan applications related to the loan application "Baomeina 5663" ( Fig.10 The loan application shown in the figure also needs to be judged whether there is a risk of fraud. If it is judged that fraud also exists in other loan applications, a triangle icon will be displayed next to the corresponding loan application to inform the financial institution staff that this loan application is a fraudulent loan.
[0101] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0102] In order to improve the efficiency and accuracy of loan approval, an embodiment of the present application provides a loan approval method, which can quickly find historical loan information related to the new loan application by obtaining new loan application information and searching for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information in the loan knowledge graph created by the first aspect, thereby reducing the time for manual search and speeding up the loan approval process; by displaying an association graph between the new loan application information and the target historical loan information, the new loan application information can be associated with the historical loan information for analysis, thereby more conveniently providing financial lending institutions with key data related to the approval of new loan applications, helping financial lending institutions to more accurately assess the risk level of new loan applications, reduce the possibility of misjudgment, and avoid the occurrence of fraudulent loan behaviors.
[0103] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.
[0104] Fig.11 The figure shows a flow chart of a loan approval method provided by an embodiment of the present application. Fig.11 As shown, the loan approval method may include the following steps S201 to S203.
[0105] Step S201: Obtain new loan application information.
[0106] After the loan knowledge graph is created, the loan knowledge graph is applied to the approval of new loan applications. When a new loan application needs to be approved, the new loan application information is first obtained, such as vehicle loan information (loan amount, loan term, interest rate), applicant's basic information (name, mobile phone number, ID number), remote signing location information, device information used when applying for the loan, etc. Specifically, the new loan application information can be obtained by obtaining the personal basic information, loan demand, vehicle loan information, financial status and other related information entered by the user in the application interface. Alternatively, an API connection can be established with an external system (such as a cooperative bank, credit scoring agency) through an application interface to automatically capture new loan application information from an external system.
[0107] Step S202: searching the loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information; the loan knowledge graph is created by the above-mentioned loan knowledge graph creation method.
[0108] Specifically, after obtaining the new loan application information, the target historical loan information that satisfies the preset relationship with the mobile phone number, ID number or address in the new loan application information is searched in the loan knowledge graph. For example, the loan knowledge graph is searched to see if there are other loan records using the same mobile phone number or ID number as the current applicant, or to see if there are loan applications at the same address or a similar address. These queries may involve traversing multiple nodes and relationships to find the target historical loan information that meets the preset conditions. The target historical loan information is the information related to the new loan application. These target historical loan information can be used as a decision-making basis for evaluating new loan applications, providing valuable references for loan approval personnel, helping loan approval personnel identify potential risks or fraud, and understanding the applicant's loan history and credit status.
[0109] Step S203: Displaying a correlation graph between the new loan application information and the target historical loan information.
[0110] Based on the search results from the loan knowledge graph, a correlation graph between the new loan application information and the target historical loan information is generated and displayed. Specifically, the correlation graph can be displayed through a visualization tool (such as a graphical interface), with nodes representing entities (such as applicants, historical loan records), and node connection lines representing relationships (such as family members, common addresses, etc.).
[0111] In some embodiments, Fig.12 A flowchart of another loan approval method provided by an embodiment of the present application is shown. Fig.12 As shown, the loan approval method further includes steps S204 to S205.
[0112] Step S204: Determine the risk level of the new loan application information based on the target historical loan information.
[0113] Specifically, the system will comprehensively analyze the target historical loan information obtained from the loan knowledge graph, including but not limited to the applicant's past credit record, repayment history, overdue record, number of loan defaults, etc. And the system will use a pre-defined risk assessment model or algorithm to process and calculate this data. The risk assessment model can combine multiple factors, such as the applicant's credit score, income stability, debt situation, past loan behavior patterns, etc. Based on the above analysis results, the system will assign a specific risk level to the new loan application information. Risk levels are usually divided into multiple levels (such as low, medium, high, and very high) to reflect the probability that the applicant may default.
[0114] Step S205: Display the risk level.
[0115] Specifically, once the system determines the risk level of the information for a new loan application, the risk level will be clearly displayed to the loan approver. The risk level can be presented graphically through the user interface, such as using color coding, bar charts or dashboards, so that the approver can quickly and intuitively understand the applicant's risk level.
[0116] In some embodiments, Fig.13 A flowchart of another loan approval method provided by an embodiment of the present application is shown. Fig.13 As shown, the loan approval method further includes steps S206 to S207.
[0117] Step S206: When the new loan application information is approved, the new loan application information, the new loan approval information and the new post-loan monitoring information in the new loan information are associated.
[0118] When the new loan application approval process is completed, a new loan approval information is obtained, and the new loan approval information is the result of loan approval or loan rejection. When the new loan approval information obtains the result of loan rejection, since the new loan application is rejected, there is no need to perform post-loan monitoring on the new loan approval information. At this time, the new loan application information and the new loan approval information in the new loan information are associated. When the new loan approval information obtains the result of loan approval, post-loan monitoring of the new loan approval begins, and the new loan application information, new loan approval information, and new post-loan monitoring information in the new loan information are associated.
[0119] Step S207: According to the preset relationship set, the new loan application information is associated with the historical loan application information and the new post-loan monitoring information is associated with the post-loan monitoring information to update the loan knowledge graph.
[0120] Specifically, according to the preset relationships in the preset relationship set, such as the ID number relationship, mobile phone number relationship, address relationship, etc., the current new loan application information is associated with the historical loan application information. In the post-loan monitoring stage, new post-loan monitoring data of new loan applications (such as repayment records, overdue conditions, changes in financial status, etc.) will be continuously generated. According to the preset relationships in the preset relationship set, the new post-loan monitoring data of these new loan applications are associated with the existing post-loan monitoring information of the historical loan application information to form a dynamically updated loan knowledge graph, that is, after a new loan application is approved using the loan knowledge graph, the approved new loan application is added to the loan knowledge graph.
[0121] The loan approval method provided in the embodiment of the present application can quickly identify new loan applications during approval, and assist in approval based on the associated orders of historical loan information. During approval, you can view the map centered on the specified mobile phone number, ID number, address, etc., and you can intuitively view the information of the associated loan orders (including fraud, overdue, approval status and detailed information, etc.) through the hierarchical and associated dimensions, which is used to simply and intuitively determine whether there are problems with the current new loan application, so as to improve approval efficiency.
[0122] The loan approval method provided in the embodiment of the present application can also define and identify suspicious gangs, mine and deduce the formation process of fraud gangs through loan knowledge graphs, deeply understand the formation path of fraud gangs, and thus formulate effective strategies to intercept fraudulent customers. Through the suspicious case number confirmed by the original anti-fraud investigation, query the relevant graph of this number, identify suspicious orders according to the association relationship, and add them to the original gang; according to the graph, multiple orders use the same device to report orders, the longitude and latitude addresses of multiple remote reports are close, and the ID number and mobile phone number that have been fraudulent are reported again. The scene can be regarded as a suspected fraud gang. In addition, it is also possible to investigate and summarize problematic cases such as rejected cases or fraud cases, analyze the relationship between multiple suspected fraud orders and the related orders of suspected fraud gangs, and learn from these relationships to obtain the common points of new fraud gangs to specify relevant fraud rules and strategies, and feed back pre-loan strategies.
[0123] The loan approval method provided in the embodiment of the present application can also generate derived variables based on the graph after the final review is completed in combination with anti-fraud rules, sort out and summarize the data in the variables and the relationship between order fraud, suspected fraud, and gang fraud, and mine out variables with a higher probability of fraud. These variables are used to reversely determine whether there is fraud in the case, and generate corresponding models based on multiple variables for determining fraud, and finally reach fraud identification rules based on these variables and models.
[0124] The loan approval method provided in the embodiment of the present application can also formulate node weights, calculate fraud gang scores, output fraud gang reports, and generate corresponding scores and order reports through the number of associations and degrees of association in different dimensions. According to the dimensions of the duplicate check nodes associated with each order (such as nodes in dimensions such as certificate number, mobile phone number, address, etc.), the corresponding weights are matched, and the order status, overdue days, fraud situation, whether it is gang fraud, etc. of the associated orders at different levels (first-level association and multi-level association) are combined to calculate the comprehensive fraud score of each order, and output the comprehensive fraud gang report of each dimension; at the same time, the order number, order quality and other information associated with different dimensions and different levels are used to calculate the score of the order, and a comprehensive order report is generated, which is provided to the approval personnel to assist in the approval of the current order, and provided to the anti-fraud personnel to assist in the investigation and judgment of the current single order anti-fraud, etc. Based on the loan knowledge map and derived variables, the quantity and quality of the associated orders at different levels of different dimensions are used as feature variables, and added to the anti-fraud rules, which can improve the accuracy and effectiveness of identifying fraudulent behavior.
[0125] Corresponding to the loan knowledge graph creation method described in the above embodiment, Fig.14 This is a structural block diagram of the loan knowledge graph creation device provided in the embodiment of the present application. For the sake of convenience of explanation, only the parts related to the embodiment of the present application are shown.
[0126] Reference Fig.14 , the loan knowledge graph creation device 14 includes:
[0127] The first acquisition module 141 is used to acquire all historical loan information, where the historical loan information includes historical loan application information, historical loan approval information and post-loan monitoring information.
[0128] The association module 142 is used to associate the historical loan application information, the historical loan approval information and the post-loan monitoring information in the same historical loan information.
[0129] The forming module 143 is used to associate different historical loan application information and different post-loan monitoring information according to a preset relationship set to form a loan knowledge graph.
[0130] In some embodiments, the forming module 143 is further configured to:
[0131] If the application data in the first historical loan application information and the application data in the second historical loan application information meet the first preset relationship in the preset relationship set, the application data in the first historical loan application information is used as the first data node, and the application data in the second historical loan application information is used as the second data node; the first historical loan application information is any one of the multiple historical loan application information, and the second historical loan application information is any one of the multiple historical loan application information except the first historical loan application information;
[0132] The first data node and the second data node are connected through a first node connection line, and the first preset relationship is added to the first node connection line.
[0133] In some embodiments, the forming module 143 is further configured to:
[0134] If the post-loan data in the first post-loan monitoring information and the post-loan data in the second post-loan monitoring information meet the second preset relationship in the preset relationship set, the post-loan data in the first post-loan monitoring information is used as the third data node, and the post-loan data in the second post-loan monitoring information is used as the fourth data node; the first post-loan monitoring information is any one of the multiple post-loan monitoring information, and the second post-loan monitoring information is any one of the multiple post-loan monitoring information except the first post-loan monitoring information;
[0135] The third data node and the fourth data node are connected through a second node connection line, and the second preset relationship is added to the second node connection line.
[0136] In some embodiments, the preset relationship in the preset relationship set includes at least one of a certificate number relationship, a mobile phone number relationship, an address relationship, a personnel relationship, and an information input device relationship.
[0137] Corresponding to the loan approval method described in the above embodiment, Fig.15 This is a structural block diagram of the loan approval device provided in the embodiment of the present application. For the sake of convenience of explanation, only the parts related to the embodiment of the present application are shown.
[0138] Reference Fig.15 , the loan approval device 15 comprises:
[0139] The second acquisition module 151 is used to acquire new loan application information.
[0140] The search module 152 is used to search the loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information; the loan knowledge graph is created by the loan knowledge graph creation method described in any one of claims 1-4.
[0141] The display module 153 is used to display a correlation diagram between the new loan application information and the target historical loan information.
[0142] In some embodiments, the loan approval device 15 further includes:
[0143] The determination module 154 is used to determine the risk level of the new loan application information based on the target historical loan information.
[0144] The risk level display module 155 is used to display the risk level.
[0145] In some embodiments, the loan approval device 15 further includes:
[0146] The new loan information associating module 156 is used to associate the new loan application information, the new loan approval information and the new post-loan monitoring information in the new loan information after the new loan application information is approved.
[0147] The updating module 157 is used to associate the new loan application information with the historical loan application information and associate the new post-loan monitoring information with the post-loan monitoring information according to the preset relationship set, so as to update the loan knowledge graph.
[0148] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the loan knowledge graph creation method / loan approval method implementation example of this application. Their specific functions and technical effects can be found in the loan knowledge graph creation method / loan approval method implementation example section and will not be repeated here.
[0149] in addition, Fig.14 The loan knowledge graph creation device shown Fig.15 The loan approval device shown may be a software unit, a hardware unit, or a combination of software and hardware, which is built into an existing terminal device, or may be integrated into the terminal device as an independent accessory, or may exist as an independent terminal device.
[0150] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0151] Fig.16 Schematic diagram of the structure of the terminal device provided in the embodiment of the present application. Fig.16 As shown, the terminal device 16 of this embodiment includes: at least one processor 160 ( Fig.16 Only one is shown in the figure) a processor, a memory 161, and a computer program 162 stored in the memory 161 and executable on the at least one processor 160, and when the processor 160 executes the computer program 162, the steps in any of the above-mentioned loan knowledge graph creation method / loan approval method embodiments are implemented.
[0152] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Fig.16 It is only an example of the terminal device 16 and does not constitute a limitation on the terminal device 16. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.
[0153] The processor 160 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0154] In some embodiments, the memory 161 may be an internal storage unit of the terminal device 16, such as a hard disk or memory of the terminal device 16. In other embodiments, the memory 161 may also be an external storage device of the terminal device 16, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 16. Further, the memory 161 may also include both an internal storage unit and an external storage device of the terminal device 16. The memory 161 is used to store an operating system, an application program, a boot loader, data, and other programs, such as the program code of the computer program. The memory 161 may also be used to temporarily store data that has been output or is to be output.
[0155] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0156] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0157] In the embodiments provided in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0158] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0159] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for creating a loan knowledge graph, characterized in that: include: Obtain all historical loan information, including historical loan application information, historical loan approval information and post-loan monitoring information; Associating the historical loan application information, the historical loan approval information and the post-loan monitoring information in the same historical loan information; According to a preset relationship set, different historical loan application information is associated and different post-loan monitoring information is associated to form a loan knowledge graph.
2. The loan knowledge graph creation method according to claim 1, characterized in that: According to the preset relationship set, different historical loan application information are associated, including: If the application data in the first historical loan application information and the application data in the second historical loan application information meet the first preset relationship in the preset relationship set, the application data in the first historical loan application information is used as the first data node, and the application data in the second historical loan application information is used as the second data node; the first historical loan application information is any one of the multiple historical loan application information, and the second historical loan application information is any one of the multiple historical loan application information except the first historical loan application information; The first data node and the second data node are connected through a first node connection line, and the first preset relationship is added to the first node connection line.
3. The loan knowledge graph creation method according to claim 1, characterized in that: According to the preset relationship set, different post-loan monitoring information are associated, including: If the post-loan data in the first post-loan monitoring information and the post-loan data in the second post-loan monitoring information meet the second preset relationship in the preset relationship set, the post-loan data in the first post-loan monitoring information is used as the third data node, and the post-loan data in the second post-loan monitoring information is used as the fourth data node; the first post-loan monitoring information is any one of the multiple post-loan monitoring information, and the second post-loan monitoring information is any one of the multiple post-loan monitoring information except the first post-loan monitoring information; The third data node and the fourth data node are connected through a second node connection line, and the second preset relationship is added to the second node connection line.
4. The loan knowledge graph creation method according to claim 1, characterized in that: The preset relationship in the preset relationship set includes at least one of a certificate number relationship, a mobile phone number relationship, an address relationship, a personnel relationship, and an information input device relationship.
5. A loan approval method, characterized in that: include: Obtain new loan application information; Searching the loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information; The loan knowledge graph is created by the loan knowledge graph creation method according to any one of claims 1 to 4; A correlation graph between the new loan application information and the target historical loan information is displayed.
6. The loan approval method according to claim 5, characterized in that: The loan approval method further comprises: Determining the risk level of the new loan application information based on the target historical loan information; Displays the stated risk level.
7. The loan approval method according to claim 5, characterized in that: The loan approval method further comprises: When the new loan application information is approved, the new loan application information, the new loan approval information and the new post-loan monitoring information in the new loan information are associated; According to the preset relationship set, the new loan application information is associated with the historical loan application information and the new post-loan monitoring information is associated with the post-loan monitoring information to update the loan knowledge graph.
8. A device for creating a loan knowledge graph, characterized in that: include: A first acquisition module is used to acquire all historical loan information, wherein the historical loan information includes historical loan application information, historical loan approval information and post-loan monitoring information; An association module, used for associating the historical loan application information, the historical loan approval information and the post-loan monitoring information in the same historical loan information; A formation module is used to associate different historical loan application information and different post-loan monitoring information according to a preset relationship set to form a loan knowledge graph.
9. A loan approval device, characterized in that: include: The second acquisition module is used to acquire new loan application information; A search module, used to search in a loan knowledge graph for target historical loan information that satisfies a preset relationship in a preset relationship set with the new loan application information; the loan knowledge graph is created by the loan knowledge graph creation method according to any one of claims 1 to 4; A display module is used to display a correlation map between the new loan application information and the target historical loan information.
10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the loan knowledge graph creation method as described in any one of claims 1 to 4 is implemented, or when the processor executes the computer program, the loan approval method as described in any one of claims 5 to 7 is implemented.
