Loan collection plan push method and device

Abstract

The embodiment of the present application provides a method and device for pushing a loan collection plan, which can be used in the field of financial technology. The method includes: grouping each loan user according to the geographical location information corresponding to each loan user to be collected, so as to form each loan user group; determining the unique identification of the collection personnel corresponding to each loan user group based on the preset collection personnel configuration rules; and pushing the corresponding loan collection plan to each collection personnel according to the unique identification of the collection personnel, wherein the loan collection plan contains the corresponding loan user group and the user information and geographical location information of each loan user in the loan user group. The present application can effectively improve the applicability and accuracy of the pushed loan collection plan, and thus can effectively improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, in particular to the field of financial technology, and specifically to a method and device for pushing a loan collection plan. Background Art

[0002] Due to the current complex business environment and changes in customer behavior characteristics, credit risk manifestations are diversified and unpredictable factors are increasing. Post-loan management capabilities, especially overdue loan collection capabilities, are facing severe challenges. It is urgent to formulate an intensive, efficient, and intelligent comprehensive management plan to replace the traditional manual collection task allocation and unified rule collection strategy, so as to save collection costs, improve collection efficiency, and optimize collection results.

[0003] At present, the push process of the existing loan collection scheme is usually as follows: according to the list of loan users to be collected, a number of loan users are pushed to the client devices of each collection staff in turn, so that the collection staff can collect loans from these loan users according to the received loan user information.

[0004] However, the existing loan collection plan push process has the problem of unreasonable allocation of collection tasks due to its simple allocation logic, insufficient considerations, lack of comprehensiveness, and the scattered distribution of loan users to be collected. Therefore, the loan collection plan push process does not consider the collection cost and time-consuming issues. Therefore, due to the existing loan collection plan push method, the pushed loan collection plan has poor applicability and low accuracy, which will lead to problems such as low efficiency and low success rate of loan collection. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a method and device for pushing a loan collection plan, which can effectively improve the applicability and accuracy of the pushed loan collection plan, thereby effectively improving the collection efficiency and success rate of collectors who collect debts according to the pushed plan, and effectively improving the user experience of financial institutions and collectors.

[0006] In order to solve the above technical problems, this application provides the following technical solutions:

[0007] In a first aspect, the present application provides a method for pushing a loan collection plan, comprising:

[0008] Grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups;

[0009] Based on the preset collection personnel configuration rules, determining the unique identifiers of the collection personnel corresponding to each of the loan user groups;

[0010] According to the unique identification of the collection personnel, the corresponding loan collection plan is pushed to each of the collection personnel, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographic location information of each of the loan users in the loan user group.

[0011] Furthermore, before pushing the corresponding loan collection plan to each of the collection personnel according to the unique identification of the collection personnel, the method further includes:

[0012] Obtaining user types corresponding to respective loan users in each of the loan user groups, and determining pre-stored collection strategy recommendation solutions corresponding to respective user types;

[0013] Correspondingly, the loan collection plan also includes the collection strategy recommendation plan corresponding to each of the loan users in the loan user group.

[0014] Furthermore, the determining of the unique identifiers of the debt collectors corresponding to the respective loan user groups based on the preset debt collector configuration rules includes:

[0015] Quantitative scoring is performed based on key features of the historical collection record information of each of the loan users, and the collection difficulty level of each of the loan users is determined based on a preset loan user collection difficulty grading library, wherein the loan user collection difficulty grading library is used to store the correspondence between the quantitative scoring of key features in the historical collection record information and the collection difficulty level;

[0016] Determining difficulty scores corresponding to each of the loan user groups according to an average value of the collection difficulty levels of each of the loan users in each of the loan user groups;

[0017] Based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, the corresponding collection personnel configuration rules are selected for each of the loan user groups, and based on the preset collection personnel configuration rules, the unique identification of the collection personnel corresponding to each of the loan user groups is determined.

[0018] Further, based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, respectively selecting the corresponding collection personnel configuration rule for each of the loan user groups, and determining the unique identification of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rule, including:

[0019] For the loan user group whose current difficulty score is less than or equal to the difficulty threshold, selecting a pre-stored low-difficulty collection personnel configuration rule;

[0020] According to the low-difficulty debt collection personnel configuration rule, a matching target debt collection personnel is selected from various debt collection personnel;

[0021] If there are multiple target debt collectors, the target debt collector with the closest geographical distance to the current loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the current loan user group is obtained;

[0022] Among them, the low-difficulty collection personnel configuration rules include:

[0023] Allocating a single debt collector for each of the loan users in the loan user group;

[0024] The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups;

[0025] The total collection amount corresponding to the collection personnel is within the preset total collection amount range;

[0026] Furthermore, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer.

[0027] Further, based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, respectively selecting the corresponding collection personnel configuration rule for each of the loan user groups, and determining the unique identification of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rule, including:

[0028] For the loan user group whose current difficulty score is greater than the difficulty threshold, selecting a pre-stored high-difficulty collection personnel configuration rule;

[0029] According to the high-difficulty collection personnel configuration rules, a matching target collection personnel is selected from various collection personnel;

[0030] If there are multiple target debt collectors, then the preset capability ratings corresponding to the respective target debt collectors are obtained;

[0031] The target debt collector who is geographically closest to the current loan user group and / or has the highest matching degree between the ability rating and the difficulty score of the loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the loan user group is obtained;

[0032] Among them, the high-difficulty collection personnel configuration rules include:

[0033] Allocating a single debt collector for each of the loan users in the loan user group;

[0034] The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups;

[0035] Furthermore, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer.

[0036] Furthermore, the obtaining of user types corresponding to the respective loan users in the respective loan user groups, and determining the pre-stored collection strategy recommendation solutions corresponding to the respective user types, includes:

[0037] Obtaining the indicator characteristics of the historical collection record information corresponding to each of the loan users, wherein the indicator characteristics include: user information, repayment willingness, collection performance and deterioration possibility;

[0038] According to the comprehensive scores of the indicator characteristics of the historical collection record information corresponding to each of the loan users, the user types corresponding to each of the loan users are determined respectively using a preset classification model;

[0039] Based on the user types corresponding to each of the loan users, the collection strategy recommendation schemes for each of the loan users are obtained from a preset loan user collection recommendation scheme grading library, wherein the collection recommendation scheme grading library is used to store the correspondence between the user type and the collection strategy recommendation scheme.

[0040] Furthermore, the loan users are grouped according to the geographical location information corresponding to each loan user to be collected, so as to form each loan user group, including:

[0041] Obtaining the address records corresponding to each loan user to be collected, wherein the address records contain at least one of the collateral address, the residential address, and the workplace address;

[0042] Based on the address records corresponding to the loan users, determining the geographical location information corresponding to the loan users;

[0043] According to the geographical location information corresponding to each of the loan users, a preset clustering algorithm is applied to group the loan users to form loan user groups.

[0044] In a second aspect, the present application provides a loan collection solution push device, comprising:

[0045] A user grouping module, in which the user groups the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups;

[0046] A personnel configuration module, used to determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on a preset collection personnel configuration rule;

[0047] The solution push module is used to push the corresponding loan collection solution to each of the collection personnel according to the unique identification of the collection personnel, wherein the loan collection solution includes the corresponding loan user group and the user information and geographic location information of each of the loan users in the loan user group.

[0048] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for pushing a loan collection plan when executing the computer program.

[0049] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the loan collection plan push method.

[0050] It can be seen from the above technical scheme that the present application provides a method and device for pushing a loan collection plan, the method comprising: grouping each of the loan users to be collected according to their respective geographical location information to form each loan user group; determining the unique identifier of the collection personnel corresponding to each of the loan user groups based on a preset collection personnel configuration rule; and pushing the corresponding loan collection plan to each of the collection personnel according to the unique identifier of the collection personnel, wherein the loan collection plan contains the corresponding loan user group and the user information and the geographical location information of each of the loan users in the loan user group. By grouping each of the loan users to be collected according to their respective geographical location information to form each loan user group, the loan collection plan can be grouped into groups corresponding to the loan user groups. Loan users to be collected within the same distance range are divided into a group, which can effectively improve the applicability and accuracy of the pushed loan collection plan. By selecting collection personnel for each loan user group to collect loans, the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan can be effectively improved; by pushing the loan collection plan containing the user information and geographic location information of the loan user group and each of the loan users in the loan user group to the client device of the collection personnel corresponding to the loan user group, the efficiency and convenience of the collection personnel in obtaining the required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, 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 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 work.

[0052] Figure 1 This is a first flow chart of the method for pushing a loan collection plan in an embodiment of the present application.

[0053] Figure 2 This is a second flow chart of the loan collection plan push method in the embodiment of the present application.

[0054] Figure 3 It is a specific flow chart of step 200 in the loan collection plan push method in the embodiment of the present application.

[0055] Figure 4 It is a first specific flow chart of step 230 in the loan collection plan push method in the embodiment of the present application.

[0056] Figure 5 It is a second specific flow chart of step 230 in the loan collection plan push method in the embodiment of the present application.

[0057] Figure 6 It is a specific flow chart of step 400 in the loan collection plan push method in the embodiment of the present application.

[0058] Figure 7 It is a specific flow chart of step 100 in the loan collection plan push method in the embodiment of the present application.

[0059] Figure 8 It is a structural diagram of a loan collection plan push device in an embodiment of the present application.

[0060] Fig. 9 It is a flowchart of the scheduling and allocation of debt collection personnel in the application example of this application.

[0061] Fig.10 It is a schematic diagram of the output process of the comprehensive collection strategy in the application example of this application.

[0062] Fig.11 It is an example diagram of the indicator system in the application example of this application.

[0063] Fig.12 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0065] It should be noted that the loan collection plan push method and device disclosed in this application can be used in the field of financial technology, and can also be used in any field outside the field of financial technology. The application field of the loan collection plan push method and device disclosed in this application is not limited.

[0066] Traditional collection processes and methods can no longer meet the needs of risk management. The risk management of overdue personal loans faces the following problems: First, the current collection task allocation method is relatively traditional, with insufficient considerations and lack of comprehensiveness. There are problems such as unreasonable collection task allocation and inability to meet fairness and maximize collection effects. Second, since the manual allocation of collection tasks cannot completely solve the problems of increased collection costs and reduced collection efficiency caused by geographical location, collection personnel are faced with scattered customer distribution and long door-to-door collection. How to reasonably plan collection routes and save collection time has become one of the problems faced by collection personnel in the face of many customers. The degree of collection management concentration needs to be improved. Third, in accordance with the requirements of internal risk control management, collection personnel regularly change hands for collection, resulting in the break of the original collection chain and the inability to predict the response of customers. "Who to collect first", "who to collect mainly", and "who can not collect" have all become questions for collection personnel facing the collection list. Fourth, collection personnel mainly formulate fixed collection methods for collection based on the overdue time. For customers with different collection response types, different customers have different levels of acceptance of collection methods, and there is an urgent need for more scientific, differentiated and intelligent collection strategies and processes.

[0067] Based on this, in view of the existing loan collection plan push method, because its designated logic of the plan is too simple, the pushed loan collection plan has the problems of poor applicability and low accuracy, which in turn leads to low efficiency and low success rate of loan collection. The present application proposes a loan collection plan push method, a loan collection plan push device, an electronic device and a computer-readable storage medium. By grouping each loan user to be collected according to the geographical location information corresponding to each loan user to be collected, so as to form each loan user group, the loan users to be collected within the same distance range can be divided into one group, thereby effectively improving the applicability and accuracy of the pushed loan collection plan. By selecting debt collectors for each loan user group to collect loans, the convenience, efficiency and success rate of debt collectors who collect loans according to the push solution can be effectively improved; by pushing a loan collection solution that includes the user information and geographic location information of the loan user group and each of the loan users in the loan user group to the client device of the debt collector corresponding to the loan user group, the efficiency and convenience of debt collectors in obtaining required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of debt collectors who collect loans according to the push solution, and effectively improve the user experience of financial institutions and debt collectors.

[0068] Based on the above content, the present application also provides a loan collection plan push device for implementing the loan collection plan push method provided in one or more embodiments of the present application. The loan collection plan push device can communicate with client devices and financial institution systems by itself or through a third-party server, etc., to receive a loan collection plan push request sent by a client device or a financial institution system, and after obtaining the loan collection plan push result according to the loan collection plan push request, send the loan collection plan to the client device of the collection personnel or the financial institution system for display, so that the collection personnel holding the client device or the collection personnel viewing the display content of the financial institution system can perform on-site loan collection processing on the loan users to be collected according to the loan collection plan.

[0069] It is understandable that the client device may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0070] The loan collection scheme push device can be a server or a client device, that is, all operations are completed in the client device. If all operations are completed in the client device, the client device can also include a processor for specific processing of the loan collection scheme push. In another practical application scenario, the specific selection can be made based on the processing capacity of the client device and the limitations of the user's usage scenario. This application does not limit this.

[0071] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0072] The server and the client device may communicate with each other using any suitable network protocol, including network protocols that have not yet been developed on the date of filing this application. The network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Of course, the network protocols may also include, for example, RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer) protocols used on top of the above protocols.

[0073] The details are described in detail through the following embodiments and application examples.

[0074] In order to solve the problem that the existing loan collection plan push method has poor applicability and low accuracy due to the overly simple specified logic of the plan, which leads to low efficiency and low success rate of loan collection, the present application provides an embodiment of a loan collection plan push method, see Figure 1 The loan collection plan pushing method executed by the loan collection plan pushing device specifically includes the following contents:

[0075] Step 100: Grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups.

[0076] In step 100, multiple loan users to be collected corresponding to the financial institution can be obtained in advance. Specifically, software such as statistical analysis software SAS (STATISTICAL ANALYSIS SYSTEM) can be used to connect to the database of the financial institution to extract the overdue customer information in the loan repayment record data table. According to the collection business standards, customer screening conditions are compiled to extract the customer set to be collected, such as extracting the loan customer group that is overdue for more than 15 days per month, and eliminating overdue customers who no longer need to be collected, etc.

[0077] Then, the geographical location information corresponding to each loan user can be directly obtained from the pre-stored database. It can be understood that the geographical location information refers to the longitude and latitude information. It is also possible to first obtain the address information corresponding to each loan user, and then obtain the geographical location information corresponding to each loan user based on the address information. For example, the address information of the loan user pre-stored in the financial institution can be obtained first, and then a request can be sent to the map API interface by writing Python code, with the input being the address information of the loan user, and the interface output being the fuzzy matching address information result, and the longitude and latitude corresponding to the address.

[0078] In step 100, the specific method of grouping loan users is to group multiple loan users with similar geographical location information into the same loan user group. Specifically, the geographical location information can be extracted by keywords and the keyword similarity calculation can be performed to confirm whether the geographical location information is similar. The loan users can also be grouped by clustering. It can be understood that the loan user group includes multiple loan users, and the total number of loan users in each loan user group and the total number of loan user groups can be preset, which can be set according to the actual application scenario.

[0079] Step 200: Based on the preset collection personnel configuration rules, the unique identifiers of the collection personnel corresponding to each of the loan user groups are determined.

[0080] In step 200, after completing the grouping of loan users, it is necessary to assign collection personnel to each loan user group. Specifically, the correspondence between loan user groups and collection personnel can be a one-to-one, one-to-many or many-to-one relationship. However, in order to avoid loan users being subject to collection by multiple collection personnel in one collection cycle as much as possible, in a preferred embodiment of step 200, the correspondence between loan user groups and collection personnel is only a one-to-one or many-to-one relationship. That is to say, in one collection cycle, one collection personnel can be responsible for loan collection work for a single or multiple loan user groups, but only one collection personnel is assigned to one loan user group to provide the best user experience for the loan users being collected.

[0081] It is understandable that the unique identification of the debt collector refers to the identification that can represent the unique identity of the debt collector in the financial institution, and can specifically be the debt collector's work number, ID number, or a coded combination of name and numbers in the financial institution, etc.

[0082] Step 300: According to the unique identifier of the collection personnel, the corresponding loan collection plan is pushed to each of the collection personnel, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographic location information of each of the loan users in the loan user group.

[0083] In an example of step 300, if through steps 100 and 200, it is determined that loan user groups D1, D2 and D5 all correspond to collection officer C123, then based on the unique identifier of the client device pre-recorded in the financial institution system by the collection officer C123 (such as an IP address or telephone number, etc.), a loan collection plan containing the user information and geographic location information of each of the loan users in loan user groups D1, D2 and D5 is sent to the client device.

[0084] Furthermore, if there are multiple loan user groups pushed to a collection officer, the multiple loan user groups can be sorted and displayed according to the distance between each of the multiple loan user groups and the geographical location of the collection officer, so that the collection officer can more intuitively and quickly obtain the recommended collection order of the multiple loan user groups according to the loan collection plan. For example, if the distances between the loan user groups D1, D2 and D5 and the geographical location of the collection officer are D2, D1, and D5 from near to far, then according to the distances between each of the multiple loan user groups D1, D2 and D5 and the geographical location of the collection officer, the loan collection plan sorts the loan user groups D1, D2 and D5 in the order of D2, D1, and D5, so that when the collection officer C123 sees the loan collection plan, he can immediately know that the collection order recommended to him according to the order of distance from near to far is D2, D1, and D5.

[0085] In addition, in step 300, in order to avoid leakage of loan user data, the loan collection plan may be encrypted according to a preset encryption method before pushing the corresponding loan collection plan to the client devices corresponding to each of the collection personnel. The encrypted loan collection plan is then sent to the client devices of the corresponding collection personnel, so that the collection personnel can decrypt the encrypted loan collection plan according to the decryption method uniquely corresponding to the encryption method obtained in advance from the financial institution to obtain the loan collection plan, thereby effectively improving the security of the loan user data, effectively protecting the privacy data of the loan user, and at the same time improving the user experience of the loan user.

[0086] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application, by grouping each loan user to be collected according to the geographical location information corresponding to each loan user to be collected, to form each loan user group, can divide the loan users to be collected within the same distance range into one group, thereby effectively improving the applicability and accuracy of the pushed loan collection plan, and by selecting collection personnel for each loan user group to collect loans, the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan can be effectively improved; by pushing the loan collection plan containing the loan user group and the user information and the geographical location information of each loan user in the loan user group to the client device of the collection personnel corresponding to the loan user group, the efficiency and convenience of the collection personnel in obtaining the required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel.

[0087] In order to provide a configuration method for a collection recommendation plan, an embodiment of a loan collection plan push method provided in this application is provided, see Figure 2 The method for pushing the loan collection plan further specifically includes the following contents after step 100 and before step 300:

[0088] Step 400: Obtain the user type corresponding to each loan user in each loan user group, and determine the pre-stored collection strategy recommendation corresponding to each user type.

[0089] Correspondingly, the loan collection plan also includes the collection strategy recommendation plan corresponding to each of the loan users in the loan user group.

[0090] It can be understood that, in one example, the user types can be divided into three categories, namely: the type that repays once urged, the type that repays repeatedly when urged, and the type that does not repay after repeated urging. The database of the financial institution contains collection strategy recommendation plans corresponding to the type that repays once urged, the type that repays repeatedly when urged, and the type that does not repay after repeated urging.

[0091] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application, by determining the recommended collection strategy plan corresponding to each loan user in the loan user group, and pushing the recommended collection strategy plan to the corresponding collection personnel, can further improve the applicability, pertinence and accuracy of the pushed loan collection plan, and thus can further improve the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0092] In order to provide an optimal method for allocating collection personnel, an embodiment of the loan collection plan push method provided in this application is provided, see Figure 3 Step 200 of the loan collection plan push method specifically includes the following contents:

[0093] Step 210: Perform quantitative scoring based on key features of the historical collection record information of each of the loan users, and determine the collection difficulty level of each of the loan users based on a preset loan user collection difficulty grading library, wherein the loan user collection difficulty grading library is used to store the correspondence between the quantitative scoring of key features in the historical collection record information and the collection difficulty level.

[0094] In one implementation of step 210, the quantitative scoring of the key features can be pre-entered into the database of the financial institution by the staff of the financial institution, and the loan collection plan push device can directly retrieve the key feature quantitative scoring of the historical collection record information of each of the loan users from the database.

[0095] Among them, the key feature quantitative score can be the same as the indicator feature mentioned in one or more embodiments of the present application, that is, the key feature can be an indicator, and the indicator includes: user information indicator, repayment willingness, collection performance and possibility of deterioration.

[0096] Step 220: Determine the difficulty score corresponding to each loan user group according to the average value of the collection difficulty level of each loan user in each loan user group.

[0097] Step 230: Based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, select the corresponding collection personnel configuration rules for each of the loan user groups, and determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rules.

[0098] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application can effectively increase the pertinence and effectiveness of the loan collection plan push by differentially selecting different collection personnel configuration plans according to the user's collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0099] In order to provide a preferred method for selecting different collection personnel deployment plans according to the difficulty of user collection, an embodiment of the loan collection plan push method provided in this application is provided, see Figure 4 Step 230 of the loan collection plan push method specifically includes the following contents:

[0100] Step 2311: For the loan user group whose current difficulty score is less than or equal to the difficulty threshold, select a pre-stored low-difficulty collection personnel configuration rule.

[0101] Step 2312: According to the low-difficulty debt collection personnel configuration rules, a matching target debt collection personnel is selected from among the debt collection personnel.

[0102] It can be understood that the low-difficulty collection personnel configuration rules may refer to constraints, that is, first select personnel who meet these constraints from the various collection personnel at your disposal. If there is more than one person after screening, then step 2313 may be executed to select the person who is geographically closest to the current loan user group as the collection personnel corresponding to the current loan user group.

[0103] Step 2313: If there are multiple target collection personnel, the target collection personnel with the closest geographical distance to the current loan user group is selected as the collection personnel corresponding to the current loan user group, and the unique identifier of the collection personnel corresponding to the current loan user group is obtained.

[0104] Among them, the low-difficulty collection personnel configuration rules include:

[0105] (1) Allocating a single debt collector for each of the loan users in the loan user group.

[0106] (2) The total number of loan user groups corresponding to the debt collection personnel is within a preset range of the total number of loan user groups.

[0107] (3) The total collection amount corresponding to the collection personnel is within the preset total collection amount range.

[0108] (4) The loan user group corresponding to the collection officer in the current collection cycle did not appear in the collection officer's previous collection cycle.

[0109] Specifically, an objective function representing minimizing the sum of the distances between the debt collectors and the customers can be set first. This objective function is to ensure the highest efficiency of the debt collection business and to ensure that customer clusters with closer distances are preferentially assigned to debt collectors, so that the overall debt collection distance is the shortest, thereby reducing the debt collection cost.

[0110] Then set the constraints of the objective function: each customer can only be collected by one collection officer; the total collection clusters of each collection officer are within the preset total number of loan user groups, ensuring that the collection workload of each collection officer is not much different, meeting the fairness of task allocation; the total collection amount of each collection officer is within the preset total collection amount, so that the task allocation must ensure that the collection amount allocation is not much different, meeting the fairness of task allocation; in order to prevent and control risks and meet the turnover principle, the current allocation result is different from the historical allocation result, ensuring that the collection officers of the same cluster in the previous cycle are different. Prevent the collection officers from collecting the same customer cluster for a long time and the same customer from having risk problems, so that they change the collection cluster when new tasks are allocated. Constraints can also include decision variables used to determine whether a customer is collected by a collection officer and variables used to determine whether a customer was collected by a collection officer in the previous cycle.

[0111] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application can further increase the pertinence and effectiveness of the loan collection plan push by executing low-difficulty collection personnel configuration rules for loan user groups with low collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0112] In order to provide a preferred method for selecting different collection personnel deployment plans according to the difficulty of user collection, an embodiment of the loan collection plan push method provided in this application is provided, see Figure 5 Step 230 of the loan collection plan push method further specifically includes the following contents:

[0113] Step 2321: For the loan user group whose current difficulty score is greater than the difficulty threshold, select a pre-stored high-difficulty collection personnel configuration rule.

[0114] Step 2322: According to the high-difficulty debt collection personnel configuration rules, a matching target debt collection personnel is selected from various debt collection personnel.

[0115] Step 2323: If there are multiple target debt collectors, obtain the preset capability ratings corresponding to each of the target debt collectors.

[0116] It is understandable that the ability rating of the target collection personnel can be set in advance by the staff of the financial institution based on the pre-acquired historical collection data of the target collection personnel, and stored in the database of the financial institution. The loan collection plan push device can directly extract the corresponding ability rating of each of the target collection personnel from the database of the financial institution.

[0117] It can be understood that the high-difficulty collection personnel configuration rules may refer to constraints, that is, first select personnel who meet these constraints from the various collection personnel at your disposal. If there is more than one person after screening, then through the execution of steps 2323 and 2324, one of the target collection personnel who is geographically closest to the current loan user group and / or has the highest match between his ability rating and the difficulty score of the loan user group is selected as the collection personnel corresponding to the current loan user group.

[0118] The calculation method of the matching degree between the capability rating and the difficulty score of the loan user group can be: if the capability rating is expressed in the form of a score, then the product between the capability rating and the difficulty score of the loan user group is calculated. The larger the product, the higher the matching degree between the capability rating and the difficulty score of the loan user group, and vice versa. In addition, if the capability rating is expressed in the form of a score, the ratio between the capability rating and the difficulty score of the loan user group can also be calculated. Alternatively, a comparison table between capability rating, difficulty score and matching degree can be retrieved from the database of a financial institution to directly query the corresponding matching degree.

[0119] Step 2324: select the target collection personnel who is geographically closest to the current loan user group and / or has the highest match between the ability rating and the difficulty score of the loan user group as the collection personnel corresponding to the current loan user group, and obtain the unique identifier of the collection personnel corresponding to the loan user group.

[0120] Among them, the high-difficulty collection personnel configuration rules include:

[0121] (1) assigning a single debt collector to each of the loan users in the loan user group;

[0122] (2) the total number of loan user groups corresponding to the debt collectors is within a preset total number of loan user groups;

[0123] (3) The loan user group corresponding to the collection officer in the current collection cycle did not appear in the collection officer's previous collection cycle.

[0124] Specifically, we can first set an objective function that minimizes the distance between customers and debt collectors and maximizes the matching degree between customer clusters and debt collectors. Under the premise of minimizing the debt collection distance, we can assign more difficult debt collection clusters to debt collectors with stronger abilities, so as to achieve a higher debt collection matching degree and improve the overall debt collection success rate. This allows the objective function to simultaneously meet the principles of highest debt collection efficiency and greatest effectiveness.

[0125] Then set the constraints of the objective function: each customer can only be collected by one collection officer; the total collection cluster of each collection officer is within the preset total number of loan user groups, ensuring that the collection workload of each collection officer is not much different, meeting the fairness of task allocation; in order to prevent and control risks and meet the turnover principle, the current allocation result is different from the historical allocation result, ensuring that the collection officers of the same cluster in the previous cycle are different. Prevent collection officers from collecting the same customer cluster for a long time and the same customer from having risk problems, so that they change the collection cluster when new tasks are assigned. Constraints can also include decision variables used to determine whether a customer is being collected by a collection officer.

[0126] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application can further increase the pertinence and effectiveness of the loan collection plan push by executing high-difficulty collection personnel configuration rules for loan user groups with high collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0127] In order to provide an optimal method for recommending a collection strategy, an embodiment of a method for pushing a loan collection plan provided in this application is provided. Figure 6 , step 400 of the loan collection plan push method specifically includes the following contents:

[0128] Step 410: Obtain the indicator characteristics of the historical collection record information corresponding to each of the loan users, wherein the indicator characteristics include: user information, repayment willingness, collection performance and possibility of deterioration.

[0129] Specifically, financial institutions can pay attention to customer fund changes and repayment habits from four dimensions, including user basic information, repayment willingness, possibility of deterioration, and historical collection performance, based on business experience and collection scenarios, extract multiple characteristic indicators, and build an indicator system that can truly reflect user characteristics, fund changes, and repayment performance.

[0130] Step 420: According to the comprehensive scores of the indicator characteristics of the historical collection record information corresponding to each of the loan users, the user types corresponding to each of the loan users are determined respectively using a preset classification model.

[0131] In step 420, the database of the financial institution may pre-store the correspondence between various numerical ranges of the total score of the indicator feature and the user type. For example, a comprehensive score with a total score between 0 and 30 points, and the corresponding user type is the type of user who repeatedly refuses to repay using a preset classification model.

[0132] Step 430: Based on the user types corresponding to the respective loan users, respectively obtain the collection strategy recommendation schemes for the respective loan users from a preset loan user collection recommendation scheme grading library, wherein the collection recommendation scheme grading library is used to store the correspondence between the user types and the collection strategy recommendation schemes.

[0133] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application determines the user type by grading the user's indicators, and then obtains the collection strategy recommendation plan for each of the loan users in a preset loan user collection recommendation plan grading library according to the different user types. This can effectively improve the pertinence and effectiveness of the collection strategy recommendation plan selection, and further improve the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0134] In order to provide an optimal method for grouping loan users, an embodiment of the loan collection scheme push method provided in this application is provided, see Figure 7 Step 100 of the loan collection plan push method specifically includes the following contents:

[0135] Step 110: Obtain the address records corresponding to each loan user to be collected, wherein the address records contain at least one of the collateral address, the residential address, and the workplace address.

[0136] In step 110, you can first connect to the database server, write a SQL script, extract the customer's collateral address, residential address, and work address from the database, and finally obtain a complete and real customer address.

[0137] Step 120: Based on the address records corresponding to the loan users, determine the geographical location information corresponding to the loan users.

[0138] Step 130: According to the geographical location information corresponding to each of the loan users, a preset clustering algorithm is applied to group the loan users to form loan user groups.

[0139] In step 130, the clustering algorithm may adopt a K-means clustering algorithm K-means or a constrained K-means clustering algorithm K-means_constrained and the like.

[0140] From the above description, it can be seen that the loan collection plan push method provided in the embodiment of the present application can effectively improve the accuracy, reliability and efficiency of loan user grouping by obtaining the geographic location information corresponding to each of the loan users and applying a preset clustering algorithm to group each of the loan users, and can effectively improve the accuracy and reliability of loan collection plan formulation, thereby further improving the collection convenience, efficiency and success rate of collection personnel who collect according to the push plan.

[0141] From the software level, in order to solve the problem that the existing loan collection plan push method has poor applicability and low accuracy due to the overly simple specified logic of the plan, which leads to low efficiency and low success rate of loan collection, the present application provides an embodiment of a loan collection plan push device for executing all or part of the contents of the loan collection plan push method, see Figure 8 The loan collection plan push device specifically includes the following contents:

[0142] The user grouping module 10 groups the loan users to be collected according to their corresponding geographical location information, so as to form loan user groups.

[0143] In the user grouping module 10, multiple loan users to be collected corresponding to the financial institution can be obtained in advance. Specifically, software such as statistical analysis software SAS (STATISTICAL ANALYSIS SYSTEM) can be used to connect to the database of the financial institution to extract the overdue customer information in the loan repayment record data table. According to the collection business standards, customer screening conditions are compiled to extract the customer set to be collected, such as extracting the loan customer group that is overdue for more than 15 days per month, and eliminating overdue customers who no longer need to be collected, etc.

[0144] Then, the geographical location information corresponding to each loan user can be directly obtained from the pre-stored database. It can be understood that the geographical location information refers to the longitude and latitude information. It is also possible to first obtain the address information corresponding to each loan user, and then obtain the geographical location information corresponding to each loan user based on the address information. For example, the address information of the loan user pre-stored in the financial institution can be obtained first, and then a request can be sent to the map API interface by writing Python code, with the input being the address information of the loan user, and the interface output being the fuzzy matching address information result, and the longitude and latitude corresponding to the address.

[0145] In the user grouping module 10, the specific method of grouping loan users is to group multiple loan users with similar geographical location information into the same loan user group. Specifically, the geographical location information can be extracted by keywords and the keyword similarity calculation can be performed, or the geographical location information can be confirmed to be similar by the latitude and longitude distance of the geographical location. The loan users can also be grouped by clustering. It is understandable that the loan user group contains multiple loan users, and the total number of loan users in each loan user group and the total number of loan user groups can be preset, which can be set according to the actual application scenario.

[0146] The personnel configuration module 20 is used to determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rules.

[0147] In the personnel configuration module 20, after completing the grouping of loan users, it is necessary to assign collection personnel to each loan user group. Specifically, the correspondence between the loan user group and the collection personnel can be a one-to-one, one-to-many or many-to-one relationship. However, in order to avoid loan users being collected by multiple collection personnel in one collection cycle as much as possible, in a preferred embodiment of the personnel configuration module 20, the correspondence between the loan user group and the collection personnel is only a one-to-one or many-to-one relationship. That is to say, in one collection cycle, one collection personnel can be responsible for the loan collection work of one or more loan user groups, but one loan user group is only assigned one collection personnel to maximize the user experience of the loan users being collected.

[0148] It is understandable that the unique identification of the debt collector refers to the identification that can represent the unique identity of the debt collector in the financial institution, and can specifically be the debt collector's work number, ID number, or a coded combination of name and numbers in the financial institution, etc.

[0149] The solution push module 30 is used to push the corresponding loan collection solution to each of the collection personnel according to the unique identification of the collection personnel, wherein the loan collection solution includes the corresponding loan user group and the user information and the geographic location information of each of the loan users in the loan user group.

[0150] In an example of the solution push module 30, if through steps 100 and 200, it is determined that the loan user groups D1, D2 and D5 all correspond to the collection personnel C123, then based on the unique identifier of the client device pre-recorded in the financial institution system by the collection personnel C123 (such as IP address or telephone number, etc.), a loan collection solution containing the user information and the geographic location information of each of the loan users in the loan user groups D1, D2 and D5 is sent to the client device.

[0151] Furthermore, if there are multiple loan user groups pushed to a collection officer, the multiple loan user groups can be sorted and displayed according to the distance between each of the multiple loan user groups and the geographical location of the collection officer, so that the collection officer can more intuitively and quickly obtain the recommended collection order of the multiple loan user groups according to the loan collection plan. For example, if the distances between the loan user groups D1, D2 and D5 and the geographical location of the collection officer are D2, D1, and D5 from near to far, then according to the distances between each of the multiple loan user groups D1, D2 and D5 and the geographical location of the collection officer, the loan collection plan sorts the loan user groups D1, D2 and D5 in the order of D2, D1, and D5, so that when the collection officer C123 sees the loan collection plan, he can immediately know that the collection order recommended to him according to the order of distance from near to far is D2, D1, and D5.

[0152] In addition, in the solution push module 30, in order to avoid the leakage of loan user data, before pushing the corresponding loan collection solution to the client devices corresponding to each of the collection personnel, the loan collection solution can be encrypted according to a preset encryption method, and then the encrypted loan collection solution is sent to the client device of the corresponding collection personnel, so that the collection personnel can decrypt the encrypted loan collection solution according to the decryption method uniquely corresponding to the encryption method obtained in advance from the financial institution to obtain the loan collection solution, thereby effectively improving the security of the loan user data, effectively protecting the privacy data of the loan user, and at the same time improving the user experience of the loan user.

[0153] The embodiment of the loan collection plan pushing device provided in the present application can be specifically used to execute the processing flow of the embodiment of the loan collection plan pushing method in the above embodiment. Its functions will not be repeated here, and reference can be made to the detailed description of the above method embodiment.

[0154] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can group each loan user to be collected according to the geographical location information corresponding to each loan user to be collected to form each loan user group, and can divide the loan users to be collected within the same distance range into one group, thereby effectively improving the applicability and accuracy of the pushed loan collection plan, and by selecting collection personnel for each loan user group to collect loans, the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan can be effectively improved; by pushing the loan collection plan containing the loan user group and the user information and the geographical location information of each loan user in the loan user group to the client device of the collection personnel corresponding to the loan user group, the efficiency and convenience of the collection personnel in obtaining the required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel.

[0155] In order to provide a configuration method for a collection recommendation plan, in an embodiment of a loan collection plan push device provided in the present application, the loan collection plan push device is further specifically configured to perform the following after step 100 and before step 300:

[0156] Step 400: Obtain the user type corresponding to each loan user in each loan user group, and determine the pre-stored collection strategy recommendation corresponding to each user type.

[0157] Correspondingly, the loan collection plan also includes the collection strategy recommendation plan corresponding to each of the loan users in the loan user group.

[0158] It can be understood that, in one example, the user types can be divided into three categories, namely: the type that repays once urged, the type that repays repeatedly when urged, and the type that does not repay after repeated urging. The database of the financial institution contains collection strategy recommendation plans corresponding to the type that repays once urged, the type that repays repeatedly when urged, and the type that does not repay after repeated urging.

[0159] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can further improve the applicability, pertinence and accuracy of the pushed loan collection plan by determining the recommended collection strategy plan corresponding to each loan user in the loan user group, and pushing the recommended collection strategy plan to the corresponding collection personnel, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the pushed plan.

[0160] In order to provide a preferred method for allocating collection personnel, in one embodiment of the loan collection plan push device provided in the present application, the personnel allocation module 20 of the loan collection plan push device is specifically used to perform the following contents:

[0161] Step 210: Perform quantitative scoring based on key features of the historical collection record information of each of the loan users, and determine the collection difficulty level of each of the loan users based on a preset loan user collection difficulty grading library, wherein the loan user collection difficulty grading library is used to store the correspondence between the quantitative scoring of key features in the historical collection record information and the collection difficulty level.

[0162] In one implementation of step 210, the quantitative scoring of the key features can be pre-entered into the database of the financial institution by the staff of the financial institution, and the loan collection plan push device can directly retrieve the key feature quantitative scoring of the historical collection record information of each of the loan users from the database.

[0163] Among them, the quantitative scoring of the key features can be the same as the indicator features mentioned in one or more embodiments of the present application, that is, the key features can be indicators, and the indicators include: user information indicators, repayment willingness, collection performance and possibility of deterioration.

[0164] Step 220: Determine the difficulty score corresponding to each loan user group according to the average value of the collection difficulty level of each loan user in each loan user group.

[0165] Step 230: Based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, select the corresponding collection personnel configuration rules for each of the loan user groups, and determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rules.

[0166] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can effectively increase the pertinence and effectiveness of the loan collection plan push by differentially selecting different collection personnel configuration plans according to the user's collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0167] In order to provide a preferred method for selecting different collection personnel configuration schemes according to the difficulty of user collection, in an embodiment of the loan collection scheme push device provided in the present application, the personnel configuration module 20 in the loan collection scheme push device is further specifically used to execute the following contents in step 230:

[0168] Step 2311: For the loan user group whose current difficulty score is less than or equal to the difficulty threshold, select a pre-stored low-difficulty collection personnel configuration rule.

[0169] Step 2312: According to the low-difficulty debt collection personnel configuration rules, a matching target debt collection personnel is selected from among the debt collection personnel.

[0170] It can be understood that the low-difficulty collection personnel configuration rules may refer to constraints, that is, first select personnel who meet these constraints from the various collection personnel at your disposal. If there is more than one person after screening, then step 2313 may be executed to select the person who is geographically closest to the current loan user group as the collection personnel corresponding to the current loan user group.

[0171] Step 2313: If there are multiple target collection personnel, the target collection personnel with the closest geographical distance to the current loan user group is selected as the collection personnel corresponding to the current loan user group, and the unique identifier of the collection personnel corresponding to the current loan user group is obtained.

[0172] Among them, the low-difficulty collection personnel configuration rules include:

[0173] (1) Allocating a single debt collector for each of the loan users in the loan user group.

[0174] (2) The total number of loan user groups corresponding to the debt collection personnel is within a preset range of the total number of loan user groups.

[0175] (3) The total collection amount corresponding to the collection personnel is within the preset total collection amount range.

[0176] (4) The loan user group corresponding to the collection officer in the current collection cycle did not appear in the collection officer's previous collection cycle.

[0177] Specifically, an objective function representing minimizing the sum of the distances between the debt collectors and the customers can be set first. This objective function is to ensure the highest efficiency of the debt collection business and to ensure that customer clusters with closer distances are preferentially assigned to debt collectors, so that the overall debt collection distance is the shortest, thereby reducing the debt collection cost.

[0178] Then set the constraints of the objective function: each customer can only be collected by one collection officer; the total collection clusters of each collection officer are within the preset total number of loan user groups, ensuring that the collection workload of each collection officer is not much different, meeting the fairness of task allocation; the total collection amount of each collection officer is within the preset total collection amount, so that the task allocation must ensure that the collection amount allocation is not much different, meeting the fairness of task allocation; in order to prevent and control risks and meet the turnover principle, the current allocation result is different from the historical allocation result, ensuring that the collection officers of the same cluster in the previous cycle are different. Prevent the collection officers from collecting the same customer cluster for a long time and the same customer from having risk problems, so that they change the collection cluster when new tasks are allocated. Constraints can also include decision variables used to determine whether a customer is collected by a collection officer and variables used to determine whether a customer was collected by a collection officer in the previous cycle.

[0179] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can further increase the pertinence and effectiveness of the loan collection plan push by executing low-difficulty collection personnel configuration rules for loan user groups with low collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0180] In order to provide a preferred method for selecting different collection personnel configuration schemes according to the difficulty of user collection, in an embodiment of the loan collection scheme push device provided in the present application, the personnel configuration module 20 in the loan collection scheme push device is further specifically used to execute the following contents in step 230:

[0181] Step 2321: For the loan user group whose current difficulty score is greater than the difficulty threshold, select a pre-stored high-difficulty collection personnel configuration rule.

[0182] Step 2322: According to the high-difficulty debt collection personnel configuration rules, a matching target debt collection personnel is selected from various debt collection personnel.

[0183] Step 2323: If there are multiple target debt collectors, obtain the preset capability ratings corresponding to each of the target debt collectors.

[0184] It is understandable that the ability rating of the target collection personnel can be set in advance by the staff of the financial institution based on the pre-acquired historical collection data of the target collection personnel, and stored in the database of the financial institution. The loan collection plan push device can directly extract the corresponding ability rating of each of the target collection personnel from the database of the financial institution.

[0185] It can be understood that the high-difficulty collection personnel configuration rules may refer to constraints, that is, first select personnel who meet these constraints from the various collection personnel at your disposal. If there is more than one person after screening, then through the execution of steps 2323 and 2324, one of the target collection personnel who is geographically closest to the current loan user group and / or has the highest match between his ability rating and the difficulty score of the loan user group is selected as the collection personnel corresponding to the current loan user group.

[0186] The calculation method of the matching degree between the capability rating and the difficulty score of the loan user group can be: if the capability rating is expressed in the form of a score, then the product between the capability rating and the difficulty score of the loan user group is calculated. The larger the product, the higher the matching degree between the capability rating and the difficulty score of the loan user group, and vice versa. In addition, if the capability rating is expressed in the form of a score, the ratio or difference between the capability rating and the difficulty score of the loan user group can also be calculated. Alternatively, a comparison table between capability rating, difficulty score and matching degree can be retrieved from the database of a financial institution to directly query the corresponding matching degree.

[0187] Step 2324: select the target collection personnel who is geographically closest to the current loan user group and / or has the highest match between the ability rating and the difficulty score of the loan user group as the collection personnel corresponding to the current loan user group, and obtain the unique identifier of the collection personnel corresponding to the loan user group.

[0188] Among them, the high-difficulty collection personnel configuration rules include:

[0189] (1) allocating a single debt collector to each of the loan users in the loan user group;

[0190] (2) the total number of loan user groups corresponding to the debt collectors is within a preset total number of loan user groups;

[0191] (3) The loan user group corresponding to the collection officer in the current collection cycle did not appear in the collection officer's previous collection cycle.

[0192] Specifically, we can first set an objective function that minimizes the distance between customers and debt collectors and maximizes the matching degree between customer clusters and debt collectors. Under the premise of minimizing the debt collection distance, we can assign more difficult debt collection clusters to debt collectors with stronger abilities, so as to achieve a higher debt collection matching degree and improve the overall debt collection success rate. This allows the objective function to simultaneously meet the principles of highest debt collection efficiency and greatest effectiveness.

[0193] Then set the constraints of the objective function: each customer can only be collected by one collection officer; the total collection cluster of each collection officer is within the preset total number of loan user groups, ensuring that the collection workload of each collection officer is not much different, meeting the fairness of task allocation; in order to prevent and control risks and meet the turnover principle, the current allocation result is different from the historical allocation result, ensuring that the collection officers of the same cluster in the previous cycle are different. Prevent collection officers from collecting the same customer cluster for a long time and the same customer from having risk problems, so that they change the collection cluster when new tasks are assigned. Constraints can also include decision variables used to determine whether a customer is being collected by a collection officer.

[0194] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can further increase the pertinence and effectiveness of the loan collection plan push by executing high-difficulty collection personnel configuration rules for loan user groups with high collection difficulty, thereby further improving the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0195] In order to provide a preferred method for recommending a collection strategy, in an embodiment of a loan collection plan push device provided in the present application, the loan collection plan push device is further specifically used to execute the following contents in step 400:

[0196] Step 410: Obtain the indicator characteristics of the historical collection record information corresponding to each of the loan users, wherein the indicator characteristics include: user information, repayment willingness, collection performance and possibility of deterioration.

[0197] Specifically, financial institutions can pay attention to customer fund changes and repayment habits from four dimensions, including user basic information, repayment willingness, possibility of deterioration, and historical collection performance, based on business experience and collection scenarios, extract multiple characteristic indicators, and build an indicator system that can truly reflect user characteristics, fund changes, and repayment performance.

[0198] Step 420: According to the comprehensive scores of the indicator characteristics of the historical collection record information corresponding to each of the loan users, the user types corresponding to each of the loan users are determined respectively using a preset classification model.

[0199] In step 420, the database of the financial institution may pre-store the correspondence between various numerical ranges of the total score of the indicator feature and the user type. For example, a comprehensive score with a total score between 0 and 30 points, and the corresponding user type is the type of user who repeatedly refuses to repay using a preset classification model.

[0200] Step 430: Based on the user types corresponding to the respective loan users, respectively obtain the collection strategy recommendation schemes for the respective loan users from a preset loan user collection recommendation scheme grading library, wherein the collection recommendation scheme grading library is used to store the correspondence between the user types and the collection strategy recommendation schemes.

[0201] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application determines the user type by grading the user's indicators, and then obtains the collection strategy recommendation plan for each of the loan users in a preset loan user collection recommendation plan grading library according to the different user types, which can effectively improve the pertinence and effectiveness of the collection strategy recommendation plan selection, and further improve the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan.

[0202] In order to provide a preferred method for grouping loan users, in an embodiment of the device for pushing a loan collection plan provided in the present application, the user grouping module 10 of the device for pushing a loan collection plan is specifically used to perform the following contents:

[0203] Step 110: Obtain the address records corresponding to each loan user to be collected, wherein the address records contain at least one of the collateral address, the residential address, and the workplace address.

[0204] In step 110, you can first connect to the database server, write a SQL script, extract the customer's collateral address, residential address, and work address from the database, and finally obtain a complete and real customer address.

[0205] Step 120: Based on the address records corresponding to the loan users, determine the geographical location information corresponding to the loan users.

[0206] Step 130: According to the geographical location information corresponding to each of the loan users, a preset clustering algorithm is applied to group the loan users to form loan user groups.

[0207] In step 130, the clustering algorithm may adopt a K-means clustering algorithm K-means or a constrained K-means clustering algorithm K-means_constrained and the like.

[0208] From the above description, it can be seen that the loan collection plan push device provided in the embodiment of the present application can effectively improve the accuracy, reliability and efficiency of loan user grouping by obtaining the geographical location information corresponding to each of the loan users and applying a preset clustering algorithm to group each of the loan users, and can effectively improve the accuracy and reliability of loan collection plan formulation, thereby further improving the collection convenience, efficiency and success rate of collection personnel who collect according to the push plan.

[0209] In order to further illustrate this solution, this application also provides a specific application example of a loan collection solution push method. The application example of this application focuses on the core issues in the entire collection process, and proposes two major solutions for "collection personnel scheduling and allocation" optimization management and "collection comprehensive strategy" business guidance. Four major models are designed to achieve intensive scheduling of collection personnel and efficient and scientific allocation of collection tasks, forming a targeted, strategic, and personalized collection method, optimizing collection strategies, improving work efficiency, and improving management models.

[0210] In the "Dispatch and Allocation of Collection Personnel", "Dispatch" uses the "customer clustering based on geographic location" model to divide customers with relatively concentrated distribution into unified customer clusters, and the collection personnel conduct collection in a centralized manner based on customer clusters. This realizes the transformation from decentralized collection of each customer to intensive collection of customer groups, so that collection personnel can collect customers in a centralized manner, and their dispatch is more scientific and standardized. "Allocation" is the design of "task allocation based on multi-objective optimization", which replaces traditional manual work with big data + model algorithms. The model comprehensively considers the four aspects of collection efficiency, collection effect, assessment, and risk control, and proposes four major factors: the distance between the collection personnel and the customer, the matching degree between the collection personnel's ability and the difficulty of the task, the collection amount, and the collection cycle. It not only achieves better collection efficiency and collection effect, but also meets the requirements of risk prevention and control.

[0211] The "comprehensive collection strategy" aims to solve the problems of how to judge whether a customer needs collection, how the collection response is effective, and how the collection can effectively recover the debt. Refined customer type classification and the adoption of differentiated collection strategies for different types of customers have become the research focus of comprehensive strategy management. This application example designs a "customer type identification" model, and divides customers into "repay as soon as they are urged", "repay repeatedly after repeated urging", "not repay after repeated urging" and "lost contact customers" according to their different response types to collection and repayment. After understanding the customer type, the "collection method recommendation" is to be able to collect collection from customers in a targeted manner using the collection method with the best response effect, and provide collection personnel with the most appropriate collection method for individual customers. This application example helps collection personnel determine who to collect first, whether collection is necessary, and other issues. After knowing the customer's collection and repayment response type, differentiated collection strategies can be adopted in a targeted manner, and collection time can be reasonably allocated to maximize the collection effect. This application example is based on the two major solutions of "collection personnel scheduling and allocation" optimization management and "collection comprehensive strategy" business guidance. It studies and designs four major models: "customer segmentation based on geographic location", "task allocation based on multi-objective optimization", "customer type identification" and "collection method recommendation".

[0212] Based on this, the specific application examples of the loan collection plan push method specifically include the following contents:

[0213] This application example realizes the dispatch and allocation of debt collection personnel by constructing two models: "customer grouping based on geographic location" and "task allocation based on multi-objective optimization". Fig. 9 , which is specifically implemented as the following steps 101 to 106.

[0214] Step 101: Screen and preprocess customer data to extract the set of tasks to be assigned. Use SAS to connect to the database and extract the overdue customer information in the loan repayment record data table. According to the collection business standards, write customer screening conditions to extract the customer set to be collected, such as extracting the loan customer group that is overdue for more than 15 days per month, eliminating overdue customers who no longer need to be collected, etc.

[0215] Step 102: Identification and quantification of the customer's geographic location. The application example of this application uses SAS to connect to the database server, write a SQL script, extract the customer's collateral address, residential address, and work unit address from the database, and finally obtain a complete and real customer address.

[0216] In order to quantify the customer address, this application example writes Python code to send a request to the map API interface. The input is the customer address, and the interface output is the fuzzy matching address information result and the longitude and latitude corresponding to the address. The specific example is shown in Table 1:

[0217] Table 1

[0218] Address Information longitude latitude No. 1, Floor, Building, Street, Xicheng District, Beijing 116.364* 39.883* Haidian District 116.292* 39.929* No. XX, a certain street (a certain section) in Daxing District 116.338* 39.744* A household in Xinyue Garden, No. X in Nanli, Fengtai District 116.240* 39.851* A household in an apartment in Chaoyang 116.490* 39.980* A certain merchant 116.490* 39.980* A certain unit in a certain building in the third district of a certain garden in Beiqijia Town, Dongxiaokou Town, Changping District, Beijing 116.446* 40.081* A household on a certain floor of a commercial building in a certain district in the southeast of a city in Liangxiang, Fangshan District 116.154* 39.726* A certain household in a certain building in a certain road community in Haidian District, Beijing 116.353* 40.022* A certain door of a certain building in the second district of Erli, a certain park in the Economic and Technological Development Zone 116.498* 39.804*

[0219] It should be noted that the “*” in the above Table 1 represents any positive integer from 0 to 9.

[0220] Step 103: Customer clustering. This application example uses distance as a measurement standard and uses the longitude and latitude distance formula (see formula 1) to calculate the distance between two points, where (A w ,A j ) represents the latitude and longitude of customer A, (B w ,B j ) represents the latitude and longitude of customer B. Using the improved clustering model K-means-constrained, the distance between customers is calculated, and the number of customers in each category is considered, and the final clustering result is obtained by continuous adjustment.

[0221] d AB =R E *arccos(sin(A w )sin(B w )+cos(A w )cos(B w )*cos(B j -A j )) Formula 1

[0222] R E is the radius of the Earth, A w Represents the latitude value of point A, A j Represents the longitude value of point A.

[0223] Step 104: Optimize the setting and quantification of model variables.

[0224] According to the actual scenario conditions of the collection business, under the constraints of meeting the number of collection personnel assigned to customers, the average number of collections by collection personnel, and the turnover of collections on schedule, and under the premise of ensuring relative fairness, it is possible to achieve both efficient and convenient collection of customers and maximize the collection efficiency of collection personnel.

[0225] This application example proposes the following four principles for task allocation, and uses corresponding principles in different situations according to actual business needs to achieve the effect of effectively solving business problems and optimizing task allocation:

[0226] Principle 1: Highest efficiency. Ensure that customer clusters that are closer are assigned to collection personnel first, so that the overall collection distance is the shortest, thereby reducing collection costs.

[0227] Principle 2: Maximum efficiency. Assign more difficult collection clusters to more capable collection personnel to achieve a higher collection matching degree and improve the overall collection success rate.

[0228] Principle 3: Fair assessment. Currently, the performance assessment standard for debt collectors is the amount of money successfully collected. To prevent large differences in assessment results, task allocation must ensure that the amount of money collected does not differ much.

[0229] Principle 4: Risk control. There are risks when debt collectors collect debts from the same customer cluster or the same customer for a long time. Therefore, debt collection assignments need to consider whether the debt collector has collected debts from the same customer cluster in the previous period. If so, the collection cluster will be changed when new tasks are assigned.

[0230] Based on the above task allocation principles, the application example of this application proposes four major task allocation principles, namely, the distance between the collection personnel and the customer cluster, the collection matching degree, the collection amount, and the collection cycle, to achieve the optimal allocation effect in units of collection clusters. The distance is measured by the distance between the collection personnel and the center of the customer cluster. The collection amount is the sum of the amounts to be collected in the collection cluster. The collection matching degree is measured by the ability of the collection personnel and the difficulty of collection. The historical number of collections and the historical average collection amount are used to quantify whether the collection experience is rich; the average number of collections for customers is used to quantify the degree of seriousness, thereby measuring the comprehensive ability of the collection personnel. The difficulty of collecting customer clusters is comprehensively considered based on the current overdue situation and the historical overdue situation. The collection cycle is mainly quantified by whether the customer has been collected in the past.

[0231] The relevant variables of this model are described as follows:

[0232] N: represents the number of client clusters.

[0233] M: indicates the number of debt collectors.

[0234] The size is N×M, representing the allocation result matrix, element a i j indicates whether the i-th customer cluster is assigned to the j-th debt collector.

[0235] The size is N×M, representing the distance matrix, and the element d ij Represents the distance between the center of the i-th customer cluster and the j-th debt collector.

[0236] The size is N×M, representing the historical allocation result matrix, and the element h ij Indicates whether the i-th customer cluster was assigned to the j-th debt collector in the previous period.

[0237] The size is 1×M, representing the capability matrix, and the element cj represents the capability value of the jth debt collector.

[0238] The size is 1×N, representing the difficulty matrix, element diff i Represents the collection difficulty value of the i-th collection cluster.

[0239] The size is 1×N, which represents the number of customers in the collection cluster. The element cn i Represents the number of customers included in the i-th collection cluster.

[0240] ε: represents an arbitrarily small positive number, the collection difficulty threshold;

[0241] μ: represents an arbitrarily small positive number, the threshold value of the number of debt collection customers;

[0242] θ: represents an arbitrarily small positive number, the lower limit of the collection customer cluster;

[0243] η: represents an arbitrarily small positive number, the upper limit of the collection customer cluster;

[0244] α: represents an arbitrarily small positive number, the lower limit of the collection amount;

[0245] β: represents an arbitrarily small positive number, the upper limit of the collection amount.

[0246] The goal of the model is to achieve both efficient and convenient collection of customers and maximize the collection efficiency of collectors under the premise of ensuring relative fairness. Therefore, the optimization goal of the model is to minimize the distance between collectors and collection tasks (customers) to ensure that collectors can collect debts nearby and achieve the best collection efficiency. At the same time, for customers with higher collection difficulty, the matching requirements for collection ability and task difficulty are increased. The optimization goal of the model increases the matching degree of maximizing collection ability and task difficulty to achieve the maximum collection efficiency. Finally, by limiting the constraints of the model, including the number of collection clusters, the amount of collection, the collection risk, and other business principles that meet actual management needs.

[0247] Step 105: Establish a collection task allocation model based on optimal decision theory:

[0248] First, the task allocation model for the customer cluster with collection difficulty less than or equal to ε is constructed as follows:

[0249]

[0250]

[0251]

[0252]

[0253]

[0254]

[0255] h ij a ij <1 year ij ∈A,h ij ∈H (1-7)

[0256] a ij ∈{0,1} a ij ∈A (1-8)

[0257] h ij ∈{0,1}h ij ∈H (1-9)

[0258] Among them, formula (1-1) is the objective function, which means minimizing the sum of the distances between the debt collector and the customer. In order to meet the highest efficiency of the debt collection business, this objective function ensures that customer clusters with closer distances are assigned to debt collectors first, so that the overall debt collection distance is the shortest, thereby reducing the debt collection cost.

[0259] Formula (1-2) indicates that each customer can only be subject to one debt collector.

[0260] Formula (1-3) and Formula (1-4) indicate that the total collection cluster of each collection staff is within the range of θ and η, ensuring that the collection workload of each collection staff is not much different, thus satisfying the fairness of task allocation.

[0261] Formula (1-5) and Formula (1-6) indicate that the total collection amount of each debt collector is within the range of α and β, so that the task allocation must ensure that the distribution of the collection amount is not much different, satisfying the fairness of task allocation.

[0262] Formula (1-7) is to prevent and control risks and meet the principle of turnover, so that the current allocation result is different from the historical allocation result, and to ensure that the collection personnel of the same cluster in the previous cycle are different. It prevents the collection personnel from collecting the same customer cluster for a long time and the same customer from having risks, so that they change the collection cluster when new tasks are assigned.

[0263] Formula (1-8) represents the decision variable a i j is a 0-1 variable used to determine whether the customer has been collected by the collection staff.

[0264] Formula (1-9) represents the variable h ij It is a 0-1 variable used to determine whether the customer was collected by a collection staff in the previous cycle.

[0265] Second, the task allocation model for the customer cluster with collection difficulty higher than ε is constructed as follows:

[0266]

[0267]

[0268]

[0269]

[0270] h ij a ij <1 year ij ∈Ah ij ∈H (2-5)

[0271] h ij ∈{0,1}h ij ∈H (2-6)

[0272] a ij ∈{0,1} a ij ∈A (2-7)

[0273] Formula (2-1) is the objective function, which means minimizing the distance between customers and debt collectors and maximizing the matching degree between customer clusters and debt collectors. Under the premise of minimizing the collection distance, the more difficult collection clusters are assigned to debt collectors with stronger abilities to achieve a higher collection matching degree and improve the overall collection success rate. This makes the objective function meet the principles of maximum collection efficiency and maximum effectiveness at the same time.

[0274] Formula (2-2) indicates that each customer can only be subject to one debt collector.

[0275] Formula (2-3) and Formula (2-4) indicate that the total collection cluster of each collection staff is within the range of θ and η, ensuring that the collection workload of each collection staff is not much different and satisfying a certain fairness in task allocation.

[0276] Formula (2-5) is used to control risks, meet the turnover principle, and ensure that the collection personnel in the previous cycle of the same cluster are different from those in this collection.

[0277] Formula (2-6) represents the variable h ij It is a 0-1 variable used to determine whether the customer was collected by a collection staff in the previous cycle.

[0278] Formula (2-7) represents the decision variable a i j is a 0-1 variable used to determine whether the customer has been collected by the collection staff.

[0279] Step 106: Use SAS and Python to write code to solve the model and provide the optimal solution of the model for task allocation decision-making: According to the actual business needs, adjust the model parameters, modify the objective function and constraints of the model until they meet the business requirements.

[0280] and not in Fig. 9 Step 107 shown in FIG. 1 : providing a task allocation scheme.

[0281] In order to simultaneously meet the four principles of "highest efficiency", "maximum effectiveness", "fair assessment" and "risk control", this application example proposes a classification optimization strategy. On the basis of "highest efficiency" and "risk control", for customer clusters with lower collection difficulty, the task allocation principle based on "fair assessment" is implemented; for customer clusters with higher collection difficulty, the task allocation principle based on "maximum effectiveness" is implemented.

[0282] This application example achieves the output of a comprehensive collection strategy by building two models: "customer type identification" and "collection method recommendation". Fig.10 , which is specifically implemented as the following steps 201 to 206.

[0283] Step 201: Select a customer set: This step is the same as step 101 and does not need to be repeated.

[0284] Step 202: Quantitative analysis of customer tag indicators.

[0285] In order to clearly divide customers from the data and provide label data for training classification models, this application example extracts obvious feature performance as quantitative indicators based on business experience and customer historical collection and repayment response behavior, uses clustering algorithms to distinguish customer types, and obtains clear division values ​​based on clustering results, thereby labeling customer samples.

[0286] Step 203: Constructing the customer classification model indicator system.

[0287] Based on business experience and collection scenarios, this application example focuses on customer fund changes and repayment habits from four dimensions: customer basic information, repayment willingness, possibility of deterioration, and historical collection performance, and extracts multiple (e.g., 216) characteristic indicators. Fig.11 , build a set of indicators that can truly reflect customer characteristics, capital changes and repayment performance.

[0288] Step 204: Customer classification model.

[0289] From the perspective of business meaning, we study the three types of customers: "repay as soon as you ask for it", "repay repeatedly after repeated requests", and "repay repeatedly without paying back". We find that the three types of customers have a nested relationship. Therefore, the three-classification problem of customers in the application example of this application is jointly implemented by two two-classification models. First, model one identifies "repay repeatedly without paying back" customers among all customers. Then, model two divides non-"repay repeatedly without paying back" customers into two types: "repay as soon as you ask for it" and "repay repeatedly without repeated requests". At the same time, in order to ensure the balance of data volume and prevent the large disparity in data proportion from directly entering the model and affecting the model effect, this paper adjusts the data proportion of model one by downsampling and SMOTE upsampling to achieve the effect of data balance. The model cleans the extracted data through several steps such as data deduplication and labeling, single value processing, and missing value supplementation. It performs feature engineering on the data of the customer set, and successively performs classification variable mapping, variable correlation judgment, continuous variable binning, variable screening based on IV values, collinearity diagnosis and other steps to complete variable integration, compression and screening. Finally, the two models screen out feature variables to build models respectively, and select random forest models from multiple classification models to achieve customer classification.

[0290] Step 205: Recommend a collection strategy.

[0291] Based on the customer type discrimination model, customers are divided into three categories: "repay as soon as prompted", "repay repeatedly after being prompted", and "do not repay despite being prompted". The core of customer classification - the customer's repayment habits and repayment ability - is used as the basis for strategy recommendation. Combined with actual collection experience, the collection strategy is judged according to the customer's category and historical response.

[0292] For customers who repay as soon as they are urged, their historical willingness and ability to repay are excellent, and their current repayment ability is high. Judging from the historical collection response, almost all customers can successfully collect debts through SMS. Combining the three points, the model will use SMS-based collection strategies to complete the collection of customers who repay as soon as they are urged.

[0293] For customers who are repeatedly urged to repay, their historical willingness to repay is good, but their repayment ability is weaker than that of the first type of customers. Judging from the historical collection response, the successful collection method for this type of customers is mainly SMS, supplemented by telephone collection. According to the current repayment ability of the customer, the intensity of the collection strategy is gradually increased, and the collection efforts are strengthened, and SMS, SMS and telephone combined, and telephone are used in turn for collection.

[0294] For customers who fail to repay despite repeated demands, their historical willingness to repay is weak and their repayment ability is poor. From the historical collection responses, it can also be seen that the collection effect of text messages is relatively small, and telephone is the main successful collection method. Therefore, combined with the repayment ability, telephone and above collection strategies are directly adopted.

[0295] Step 206: The debt collector collects debts based on the customer type and strategy.

[0296] This application example will identify the customer type of the customers to be collected one by one, and recommend corresponding collection strategies for reference. The collection personnel will conduct differentiated collection on customers based on the provided customer feature information, customer category information and recommended collection strategies, combined with their own collection experience.

[0297] In summary, the application example of this application can have the following effects and advantages:

[0298] 1. This application example integrates and innovates machine learning technology and operational optimization technology, and uses it in the collection business in the post-loan risk control field. It also combines management difficulties and business pain points to build an industry-leading intelligent collection comprehensive system.

[0299] 2. Use the application examples of this application to solve the core problems of the entire collection process for relevant business departments, realize an intensive, refined and intelligent comprehensive collection system, and realize the upgrade and transformation from traditional collection to intelligent collection.

[0300] 3. By using the application examples of this application, customers with relatively concentrated distribution can be divided into unified customer clusters, and collection personnel can collect debts in a centralized manner based on customer clusters, thereby achieving a transformation from a decentralized collection model of each customer to an intensive collection model based on customer groups, thereby improving the degree of collection intensiveness.

[0301] 4. This application example has been used to implement a new task allocation model in the field of personal loan collection, providing effective intelligent task allocation solutions and decision-making references for relevant business departments. By replacing traditional manual task allocation with "big data + model algorithm", technology is used to liberate manpower, reduce subjective influences, and achieve efficient, objective, and fair task allocation.

[0302] 5. Use the application examples of this application to provide references for the collection strategies of front-line collection personnel. Through "model algorithm + experience judgment", refine the classification of customer types, thereby helping collection personnel to determine who to prioritize collection, whether collection is necessary, and other issues, and provide references for differentiated collection strategies.

[0303] From the hardware level, in order to solve the problem that the existing loan collection plan push method has poor applicability and low accuracy due to the overly simple specified logic of the plan, which leads to low efficiency and low success rate of loan collection, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the loan collection plan push method, and the electronic device specifically includes the following contents:

[0304] Fig.12 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig.12 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig.12 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0305] In one embodiment, the loan collection plan push function may be integrated into the central processing unit. The central processing unit may be configured to perform the following control:

[0306] Step 100: Grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups.

[0307] In step 100, multiple loan users to be collected corresponding to the financial institution can be obtained in advance. Specifically, software such as statistical analysis software SAS (STATISTICAL ANALYSIS SYSTEM) can be used to connect to the database of the financial institution to extract the overdue customer information in the loan repayment record data table. According to the collection business standards, customer screening conditions are compiled to extract the customer set to be collected, such as extracting the loan customer group that is overdue for more than 15 days per month, and eliminating overdue customers who no longer need to be collected, etc.

[0308] Then, the geographical location information corresponding to each loan user can be directly obtained from the pre-stored database. It can be understood that the geographical location information refers to the longitude and latitude information. It is also possible to first obtain the address information corresponding to each loan user, and then obtain the geographical location information corresponding to each loan user based on the address information. For example, the address information of the loan user pre-stored in the financial institution can be obtained first, and then a request can be sent to the map API interface by writing Python code, with the input being the address information of the loan user, and the interface output being the fuzzy matching address information result, and the longitude and latitude corresponding to the address.

[0309] In step 100, the specific method of grouping loan users is to group multiple loan users with similar geographical location information into the same loan user group. Specifically, the geographical location information can be extracted by keywords and the keyword similarity calculation can be performed to confirm whether the geographical location information is similar. The loan users can also be grouped by clustering. It can be understood that the loan user group includes multiple loan users, and the total number of loan users in each loan user group and the total number of loan user groups can be preset, which can be set according to the actual application scenario.

[0310] Step 200: Based on the preset collection personnel configuration rules, the unique identifiers of the collection personnel corresponding to each of the loan user groups are determined.

[0311] In step 200, after completing the grouping of loan users, it is necessary to assign collection personnel to each loan user group. Specifically, the correspondence between loan user groups and collection personnel can be a one-to-one, one-to-many or many-to-one relationship. However, in order to avoid loan users being subject to collection by multiple collection personnel in one collection cycle as much as possible, in a preferred embodiment of step 200, the correspondence between loan user groups and collection personnel is only a one-to-one or many-to-one relationship. That is to say, in one collection cycle, one collection personnel can be responsible for loan collection work for a single or multiple loan user groups, but only one collection personnel is assigned to one loan user group to provide the best user experience for the loan users being collected.

[0312] It is understandable that the unique identification of the debt collector refers to the identification that can represent the unique identity of the debt collector in the financial institution, and can specifically be the debt collector's work number, ID number, or a coded combination of name and numbers in the financial institution, etc.

[0313] Step 300: According to the unique identifier of the collection personnel, the corresponding loan collection plan is pushed to each of the collection personnel, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographic location information of each of the loan users in the loan user group.

[0314] In an example of step 300, if through steps 100 and 200, it is determined that loan user groups D1, D2 and D5 all correspond to collection officer C123, then based on the unique identifier of the client device pre-recorded in the financial institution system by the collection officer C123 (such as an IP address or telephone number, etc.), a loan collection plan containing the user information and geographic location information of each of the loan users in loan user groups D1, D2 and D5 is sent to the client device.

[0315] Furthermore, if there are multiple loan user groups pushed to a collection officer, the multiple loan user groups can be sorted and displayed according to the distance between each of the multiple loan user groups and the geographical location of the collection officer, so that the collection officer can more intuitively and quickly obtain the recommended collection order of the multiple loan user groups according to the loan collection plan. For example, if the distances between the loan user groups D1, D2 and D5 and the geographical location of the collection officer are D2, D1, and D5 from near to far, then according to the distances between each of the multiple loan user groups D1, D2 and D5 and the geographical location of the collection officer, the loan collection plan sorts the loan user groups D1, D2 and D5 in the order of D2, D1, and D5, so that when the collection officer C123 sees the loan collection plan, he can immediately know that the collection order recommended to him according to the order of distance from near to far is D2, D1, and D5.

[0316] In addition, in step 300, in order to avoid leakage of loan user data, the loan collection plan may be encrypted according to a preset encryption method before pushing the corresponding loan collection plan to the client devices corresponding to each of the collection personnel. The encrypted loan collection plan is then sent to the client devices of the corresponding collection personnel, so that the collection personnel can decrypt the encrypted loan collection plan according to the decryption method uniquely corresponding to the encryption method obtained in advance from the financial institution to obtain the loan collection plan, thereby effectively improving the security of the loan user data, effectively protecting the privacy data of the loan user, and at the same time improving the user experience of the loan user.

[0317] From the above description, it can be seen that the electronic device provided in the embodiment of the present application can group each loan user to be collected according to the geographical location information corresponding to each loan user to form each loan user group, and can divide the loan users to be collected within the same distance range into one group, thereby effectively improving the applicability and accuracy of the pushed loan collection plan, and by selecting collection personnel for each loan user group to collect loans, the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan can be effectively improved; by pushing the loan collection plan containing the loan user group and the user information and the geographical location information of each loan user in the loan user group to the client device of the collection personnel corresponding to the loan user group, the efficiency and convenience of the collection personnel in obtaining the required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel.

[0318] In another embodiment, the loan collection plan push device can be configured separately from the central processor 9100. For example, the loan collection plan push device can be configured as a chip connected to the central processor 9100, and the loan collection plan push function is realized through the control of the central processor.

[0319] like Fig.12 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig.12 In addition, the electronic device 9600 may also include Fig.12 For components not shown, reference may be made to the prior art.

[0320] like Fig.12 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0321] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0322] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0323] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0324] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0325] The communication module 9110 is a transmitter / receiver 9110 that sends and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0326] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module, etc. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0327] The embodiment of the present application also provides a computer-readable storage medium capable of implementing all the steps in the method for pushing a loan collection plan in the above embodiment. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all the steps in the method for pushing a loan collection plan in the above embodiment are implemented by the execution subject being a server or a client. For example, when the processor executes the computer program, the following steps are implemented:

[0328] Step 100: Grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups.

[0329] In step 100, multiple loan users to be collected corresponding to the financial institution can be obtained in advance. Specifically, software such as statistical analysis software SAS (STATISTICAL ANALYSIS SYSTEM) can be used to connect to the database of the financial institution to extract the overdue customer information in the loan repayment record data table. According to the collection business standards, customer screening conditions are compiled to extract the customer set to be collected, such as extracting the loan customer group that is overdue for more than 15 days per month, and eliminating overdue customers who no longer need to be collected, etc.

[0330] Then, the geographical location information corresponding to each loan user can be directly obtained from the pre-stored database. It can be understood that the geographical location information refers to the longitude and latitude information. It is also possible to first obtain the address information corresponding to each loan user, and then obtain the geographical location information corresponding to each loan user based on the address information. For example, the address information of the loan user pre-stored in the financial institution can be obtained first, and then a request can be sent to the map API interface by writing Python code, with the input being the address information of the loan user, and the interface output being the fuzzy matching address information result, and the longitude and latitude corresponding to the address.

[0331] In step 100, the specific method of grouping loan users is to group multiple loan users with similar geographical location information into the same loan user group. Specifically, the geographical location information can be extracted by keywords and the keyword similarity calculation can be performed to confirm whether the geographical location information is similar. The loan users can also be grouped by clustering. It can be understood that the loan user group includes multiple loan users, and the total number of loan users in each loan user group and the total number of loan user groups can be preset, which can be set according to the actual application scenario.

[0332] Step 200: Based on the preset collection personnel configuration rules, the unique identifiers of the collection personnel corresponding to each of the loan user groups are determined.

[0333] In step 200, after completing the grouping of loan users, it is necessary to assign collection personnel to each loan user group. Specifically, the correspondence between loan user groups and collection personnel can be a one-to-one, one-to-many or many-to-one relationship. However, in order to avoid loan users being subject to collection by multiple collection personnel in one collection cycle as much as possible, in a preferred embodiment of step 200, the correspondence between loan user groups and collection personnel is only a one-to-one or many-to-one relationship. That is to say, in one collection cycle, one collection personnel can be responsible for loan collection work for a single or multiple loan user groups, but only one collection personnel is assigned to one loan user group to provide the best user experience for the loan users being collected.

[0334] It is understandable that the unique identification of the debt collector refers to the identification that can represent the unique identity of the debt collector in the financial institution, and can specifically be the debt collector's work number, ID number, or a coded combination of name and numbers in the financial institution, etc.

[0335] Step 300: According to the unique identifier of the collection personnel, the corresponding loan collection plan is pushed to each of the collection personnel, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographic location information of each of the loan users in the loan user group.

[0336] In an example of step 300, if through steps 100 and 200, it is determined that loan user groups D1, D2 and D5 all correspond to collection officer C123, then based on the unique identifier of the client device pre-recorded in the financial institution system by the collection officer C123 (such as an IP address or telephone number, etc.), a loan collection plan containing the user information and geographic location information of each of the loan users in loan user groups D1, D2 and D5 is sent to the client device.

[0337] Furthermore, if there are multiple loan user groups pushed to a collection officer, the multiple loan user groups can be sorted and displayed according to the distance between each of the multiple loan user groups and the geographical location of the collection officer, so that the collection officer can more intuitively and quickly obtain the recommended collection order of the multiple loan user groups according to the loan collection plan. For example, if the distances between the loan user groups D1, D2 and D5 and the geographical location of the collection officer are D2, D1, and D5 from near to far, then according to the distances between each of the multiple loan user groups D1, D2 and D5 and the geographical location of the collection officer, the loan collection plan sorts the loan user groups D1, D2 and D5 in the order of D2, D1, and D5, so that when the collection officer C123 sees the loan collection plan, he can immediately know that the collection order recommended to him according to the order of distance from near to far is D2, D1, and D5.

[0338] In addition, in step 300, in order to avoid leakage of loan user data, the loan collection plan may be encrypted according to a preset encryption method before pushing the corresponding loan collection plan to the client devices corresponding to each of the collection personnel. The encrypted loan collection plan is then sent to the client devices of the corresponding collection personnel, so that the collection personnel can decrypt the encrypted loan collection plan according to the decryption method uniquely corresponding to the encryption method obtained in advance from the financial institution to obtain the loan collection plan, thereby effectively improving the security of the loan user data, effectively protecting the privacy data of the loan user, and at the same time improving the user experience of the loan user.

[0339] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application can group each loan user to be collected according to the geographical location information corresponding to each loan user to form each loan user group, and can divide the loan users to be collected within the same distance range into one group, thereby effectively improving the applicability and accuracy of the pushed loan collection plan, and by selecting collection personnel for each loan user group to collect loans, the collection convenience, efficiency and success rate of the collection personnel who collect according to the push plan can be effectively improved; by pushing the loan collection plan containing the loan user group and the user information and the geographical location information of each loan user in the loan user group to the client device of the collection personnel corresponding to the loan user group, the efficiency and convenience of the collection personnel in obtaining the required information can be effectively improved, and the push process is efficient and targeted, which can further improve the collection efficiency and success rate of the collection personnel who collect according to the push plan, and effectively improve the user experience of financial institutions and collection personnel.

[0340] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (devices), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the functions in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0341] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0342] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should be understood as a limitation of the present invention.

Claims

1. A method for pushing a loan collection plan, characterized in that: include: Grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups; Based on the preset collection personnel configuration rules, determining the unique identifiers of the collection personnel corresponding to each of the loan user groups; According to the unique identifier of the debt collector, a corresponding loan collection plan is pushed to each debt collector, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographical location information of each loan user in the loan user group; The method of determining the unique identifiers of the debt collectors corresponding to the loan user groups based on the preset debt collector configuration rules includes: Quantitative scoring is performed based on the key features of the historical collection record information of each of the loan users, and the collection difficulty level of each of the loan users is determined based on a preset loan user collection difficulty grading library, wherein the loan user collection difficulty grading library is used to store the corresponding relationship between the quantitative scoring of the key features in the historical collection record information and the collection difficulty level; the key features include: user information, repayment willingness, collection performance and possibility of deterioration; Determining difficulty scores corresponding to each of the loan user groups according to an average value of the collection difficulty levels of each of the loan users in each of the loan user groups; Based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, respectively select the collection personnel configuration rule corresponding to each of the loan user groups, and determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rule; The selecting of the collection personnel configuration rules corresponding to each loan user group based on the comparison result between the difficulty score corresponding to each loan user group and the difficulty threshold, and determining the unique identifier of the collection personnel corresponding to each loan user group based on the preset collection personnel configuration rules, includes: For the loan user group whose current difficulty score is less than or equal to the difficulty threshold, selecting a pre-stored low-difficulty collection personnel configuration rule; According to the low-difficulty debt collection personnel configuration rule, a matching target debt collection personnel is selected from various debt collection personnel; If there are multiple target debt collectors, the target debt collector with the closest geographical distance to the current loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the current loan user group is obtained; Among them, the low-difficulty collection personnel configuration rules include: Allocating a single debt collector for each of the loan users in the loan user group; The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups; The total collection amount corresponding to the collection personnel is within the preset total collection amount range; And, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer; The selecting of the collection personnel configuration rules corresponding to each loan user group based on the comparison result between the difficulty score corresponding to each loan user group and the difficulty threshold, and determining the unique identifier of the collection personnel corresponding to each loan user group based on the preset collection personnel configuration rules, includes: For the loan user group whose current difficulty score is greater than the difficulty threshold, selecting a pre-stored high-difficulty collection personnel configuration rule; According to the high-difficulty collection personnel configuration rules, a matching target collection personnel is selected from various collection personnel; If there are multiple target debt collectors, then the preset capability ratings corresponding to the respective target debt collectors are obtained; The target debt collector who is geographically closest to the current loan user group and / or has the highest matching degree between the ability rating and the difficulty score of the loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the loan user group is obtained; Among them, the high-difficulty collection personnel configuration rules include: Allocating a single debt collector for each of the loan users in the loan user group; The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups; Furthermore, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer.

2. The method for pushing a loan collection plan according to claim 1, characterized in that: Before pushing the corresponding loan collection plan to each of the collection personnel according to the unique identifier of the collection personnel, the method further includes: Obtaining user types corresponding to respective loan users in each of the loan user groups, and determining pre-stored collection strategy recommendation solutions corresponding to respective user types; Correspondingly, the loan collection plan also includes the collection strategy recommendation plan corresponding to each of the loan users in the loan user group.

3. The method for pushing a loan collection plan according to claim 2, characterized in that: The obtaining of the user type corresponding to each loan user in each loan user group, and determining the pre-stored collection strategy recommendation corresponding to each user type, includes: Obtaining the indicator characteristics of the historical collection record information corresponding to each of the loan users, wherein the indicator characteristics include: user information, repayment willingness, collection performance and deterioration possibility; According to the comprehensive scores of the indicator characteristics of the historical collection record information corresponding to each of the loan users, the user types corresponding to each of the loan users are determined respectively using a preset classification model; Based on the user types corresponding to each of the loan users, the collection strategy recommendation schemes for each of the loan users are obtained from a preset loan user collection recommendation scheme grading library, wherein the collection recommendation scheme grading library is used to store the correspondence between the user type and the collection strategy recommendation scheme.

4. The method for pushing a loan collection plan according to any one of claims 1 to 3, characterized in that: The step of grouping the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups, includes: Obtaining the address records corresponding to each loan user to be collected, wherein the address records contain at least one of the collateral address, the residential address, and the workplace address; Based on the address records corresponding to the loan users, determining the geographical location information corresponding to the loan users; According to the geographical location information corresponding to each of the loan users, a preset clustering algorithm is applied to group the loan users to form loan user groups.

5. A device for pushing loan collection plans, characterized in that: include: A user grouping module, in which the user groups the loan users according to the geographical location information corresponding to the loan users to be collected, so as to form loan user groups; A personnel configuration module, used to determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on a preset collection personnel configuration rule; A plan push module, used to push a corresponding loan collection plan to each of the collection personnel according to the unique identifier of the collection personnel, wherein the loan collection plan includes the corresponding loan user group and the user information and the geographical location information of each of the loan users in the loan user group; The staffing module is specifically used for: Quantitative scoring is performed based on the key features of the historical collection record information of each of the loan users, and the collection difficulty level of each of the loan users is determined based on a preset loan user collection difficulty grading library, wherein the loan user collection difficulty grading library is used to store the corresponding relationship between the quantitative scoring of the key features in the historical collection record information and the collection difficulty level; the key features include: user information, repayment willingness, collection performance and possibility of deterioration; Determining difficulty scores corresponding to each of the loan user groups according to an average value of the collection difficulty levels of each of the loan users in each of the loan user groups; Based on the comparison result between the difficulty score corresponding to each of the loan user groups and the difficulty threshold, respectively select the collection personnel configuration rule corresponding to each of the loan user groups, and determine the unique identifier of the collection personnel corresponding to each of the loan user groups based on the preset collection personnel configuration rule; The selecting of the collection personnel configuration rules corresponding to each loan user group based on the comparison result between the difficulty score corresponding to each loan user group and the difficulty threshold, and determining the unique identifier of the collection personnel corresponding to each loan user group based on the preset collection personnel configuration rules, includes: For the loan user group whose current difficulty score is less than or equal to the difficulty threshold, selecting a pre-stored low-difficulty collection personnel configuration rule; According to the low-difficulty debt collection personnel configuration rule, a matching target debt collection personnel is selected from various debt collection personnel; If there are multiple target debt collectors, the target debt collector with the closest geographical distance to the current loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the current loan user group is obtained; Among them, the low-difficulty collection personnel configuration rules include: Allocating a single debt collector for each of the loan users in the loan user group; The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups; The total collection amount corresponding to the collection personnel is within the preset total collection amount range; And, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer; The selecting of the collection personnel configuration rules corresponding to each loan user group based on the comparison result between the difficulty score corresponding to each loan user group and the difficulty threshold, and determining the unique identifier of the collection personnel corresponding to each loan user group based on the preset collection personnel configuration rules, includes: For the loan user group whose current difficulty score is greater than the difficulty threshold, selecting a pre-stored high-difficulty collection personnel configuration rule; According to the high-difficulty collection personnel configuration rules, a matching target collection personnel is selected from various collection personnel; If there are multiple target debt collectors, then the preset capability ratings corresponding to the respective target debt collectors are obtained; The target debt collector who is geographically closest to the current loan user group and / or has the highest matching degree between the ability rating and the difficulty score of the loan user group is selected as the debt collector corresponding to the current loan user group, and the unique identifier of the debt collector corresponding to the loan user group is obtained; Among them, the high-difficulty collection personnel configuration rules include: Allocating a single debt collector for each of the loan users in the loan user group; The total number of loan user groups corresponding to the debt collectors is within a preset total number range of loan user groups; Furthermore, the loan user group corresponding to the collection officer in the current collection cycle did not appear in the previous collection cycle of the collection officer.

6. An electronic 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 collection plan pushing method described in any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the loan collection plan push method described in any one of claims 1 to 4 is implemented.

Images