Overdue fund collection method and device, storage medium and computer equipment

Abstract

The invention relates to the technical field of financial information processing, and discloses an overdue fund collection method and device, a storage medium and computer equipment, and the device comprises an information acquisition module, an overdue account grading module, a collection strategy making module, a relative account coordination analysis module, a multi-account synchronous collection module and a mobile equipment behavior data calling and analysis module. According to the overdue fund collection method and device, the storage medium and the computer equipment, more comprehensive account information can be obtained by combining the overdue account and the relative account for analysis, so that the accuracy of judging the reason of the overdue behavior is improved, and the user experience is improved. And synchronous collection reminding can be carried out through multiple account channels through the third-party data platform and the account query interface, so that the probability that a debtor receives the collection information is increased, and meanwhile, a list of applications installed in the mobile equipment of the overdue user is obtained, so that the behavior characteristics of the overdue user can be known more comprehensively, and more bases are provided for collection decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of financial information processing, and in particular to a method, device, storage medium and computer device for collecting overdue payments. Background Art

[0002] With the development of big data analysis technology, message push technology has emerged. Through message push technology and combined with accurate user portrait analysis, a computer can actively push appropriate content to appropriate users in appropriate scenarios, so as to greatly improve the click-through rate of messages. In the scenario of collecting credit card debts, applying message push technology can determine the collection methods corresponding to different credit card users and push collection messages to credit card users through the corresponding collection methods, so as to improve the collection efficiency of credit cards.

[0003] A common method for collecting overdue payments is to first obtain data such as the overdue time and amount of users, and use the obtained data to score and rate overdue users, then use credit scoring technology to automatically determine the credit risks of overdue users and divide corresponding risk levels, and finally formulate targeted collection strategies for overdue users with different risk levels; However, since this method only obtains the overdue time and amount of overdue users, when it is determined to be medium-term and long-term overdue, manual collection means will be used to handle overdue loans in the financial industry. This manual collection method not only requires a large amount of cost investment in the training of collection personnel, but also has the problem of low collection work efficiency. In addition, only judging based on the overdue time and amount of overdue users will cause one-sidedness of overdue data, and it is impossible to formulate effective measures according to specific situations, reducing the overall collection efficiency. At the same time, only by sending text messages during the initial collection, the collection channels are single, which is likely to cause debtors to ignore the text message reminders, reducing the repayment rate of early-stage overdue users and not meeting the work requirements of financial information processing. Therefore, a method, device, storage medium and computer device for collecting overdue payments are proposed. Summary of the Invention (I) Technical Problems to be Solved Aiming at the deficiencies of the prior art, the present invention provides a method, device, storage medium and computer device for collecting overdue payments to solve the technical problems that only obtaining the overdue time and amount of overdue users causes one-sidedness of overdue data, and only by sending text messages, the collection channels are single, which is likely to cause debtors to ignore the text message reminders.

[0004] (II) Technical Solutions To achieve the above object, the present invention provides the following technical solutions: A method for collecting overdue payments, including the following steps: S1 obtains the overdue information of overdue accounts through the collection system. From the database storing loan-related data, according to the preset overdue judgment rules, it extracts the key information of overdue accounts, including the borrowing amount, borrowing date, due date, and actual repayment amount. It analyzes the historical repayment data through data mining algorithms to identify repayment behavior patterns. There are data encryption blocks and decryption modules in the database of the account. Before sending a query request, it will encrypt the sensitive information in the query conditions to ensure the security of data during transmission. When receiving the query result from the database, it will decrypt the encrypted data in the result for subsequent processing in the collection system; S2 classifies the overdue accounts according to different overdue information; S21 extracts the basic information of overdue accounts from the core database of the credit institution Extract the basic information of overdue accounts from the core database of the credit institution, including name, ID number, contact information, borrowing amount, and overdue duration, and synchronously collect the transaction records related to the overdue accounts, including the loan disbursement time, repayment plan, and actual repayment records, to analyze their repayment behavior patterns; S22 associates the relative accounts of the overdue accounts According to the identity information of the overdue accounts, search for related relative accounts through the relational database. The association basis can be registration at the same address, emergency contact setting, and common financial association. For the found relative accounts related to the overdue accounts, synchronously collect their basic information and transaction records, including data on income sources and consumption habits, to comprehensively understand the financial status of the relative accounts. According to the data obtained above, the clustering analysis method is used to classify the accounts with similar overdue behavior patterns, and at the same time, association rule mining is used to discover the potential association rules between the overdue accounts and the relative accounts, and timely discover the relationship between the high-consumption behavior of certain relative accounts and the overdue risk of the overdue accounts. Through the above steps, the overdue users and relative users without association, that is, those who are simply overdue within 1 - 15 days and over-limit, are classified as first-level overdue users, while those with an association between the overdue users and relative users are classified as second-level overdue users. The historical data is used for model training between the relative accounts found by the relational database and the overdue accounts. The model is optimized according to the prediction accuracy and recall rate indicators of the model. Then, the integrated joint account dataset is input into the optimized analysis model to obtain the analysis results of the overdue behavior of the overdue accounts and their relative accounts, so as to output the overdue risk score of the overdue accounts and the potential influencing factors of the relative accounts on the overdue behavior, including the fund transfer situation of the relative accounts. At the same time, the data is compared with the actual overdue collection results to verify the accuracy of the analysis model. When a deviation is found, the deviation information is fed back into the model for adjustment to continuously improve the accuracy of the model in obtaining overdue behavior data; S23 Linkage Accounts Query for Overdue Accounts For overdue users with an overdue duration exceeding 15 days obtained in step S21, they are classified as third-level overdue users. Through a third-party data platform and an account query interface, query the information of social software, game accounts, and application software accounts registered under the mobile phone number of the overdue user, and verify the obtained account information to exclude invalid or incorrect accounts. Then, organize and classify the valid account information according to the account type for subsequent operations. Also, send a request for authorization to obtain mobile device behavior data to the third-level overdue users. The third-party data platform and the account query interface utilize the uniqueness of the mobile phone number when registering on different platforms to establish an association between the overdue account and various Internet accounts. When querying the third-party data platform, use the mobile phone number as an index to find other account information associated with this mobile phone number, and simultaneously establish a relationship mapping model between accounts. When it is analyzed that there is an association between a certain game account and a specific social account in terms of registration information and usage habits, determine whether they belong to the same user. This relationship mapping helps in more comprehensive information transmission during the collection process. The collection system internally adopts a microservices architecture model, splitting the collection system into multiple independent microservices, including an account query microservice, a collection strategy microservice, and a message sending microservice, and uses containerization technology for deployment to improve the portability and scalability of the system. At the same time, establish a version control system within the collection system to manage the code of the collection system, conduct regular code reviews and vulnerability scans, and promptly repair the discovered problems. There is also a data analysis and marking module within the collection system. In the case of the Android system, the collection system interacts with the collection system device through the developer interface provided by the Android system to obtain data such as application information. For the Apple system, according to Apple's developer agreement, use iOS development tools and related APIs to interact with the collection system device. The data analysis and marking module can start the corresponding data collection program based on the mobile device type of the overdue user, and then transfer the collected data to the analysis module. The analysis module analyzes the data according to preset rules and focuses on marking the behaviors of overdue users who have downloaded or logged in to VPNs or external network high-risk software; S24 Query Behavioral Data of Mobile Devices of Overdue Accounts For overdue users with an overdue duration exceeding 30 days obtained in step S23, they are classified as fourth-level overdue users, and an authorization request is sent to the overdue user via text message, APP notification, and sending a document, clearly informing the purpose, scope, and confidentiality measures for obtaining the data. After the user agrees, for Android devices, use device management tools and cooperate with mobile phone manufacturers to obtain the list of installed applications inside the mobile device of the overdue user, and check whether there are VPN, external network-related applications, or high-risk software marked by the internal security system of the mobile phone. Mark the overdue users who have downloaded and logged in to VPNs, external networks, and those with high-risk software as key targets and classify them as fourth-level high-risk users; S3 formulates different collection methods for overdue accounts at different levels For first-level users, adopt a gentle collection method, mainly sending text message reminders to the overdue account holder, and recording the basic information of the overdue account; For second-level users, a joint collection strategy can be formulated to contact both the overdue account and the relative's account simultaneously for repayment negotiation; For third-level users, start sending reminder messages to the social accounts associated with the user's mobile phone number. On the basis of the initial reminder, the message content is increased with an explanation of the credit impact. For the game accounts and application software accounts associated with the user's mobile phone number, in the premise of not affecting the normal use of the user, send in-site notification reminders; For fourth-level users and fourth-level high-risk users, a more aggressive collection method of door-to-door collection is required, and the collection frequency and legal means are increased.

[0005] A device for collecting overdue payments, comprising: An information acquisition module, an overdue account grading module, a collection strategy formulation module, a relative account coordination analysis module, a multi-account synchronous collection module, and a mobile device behavior data retrieval and analysis module. The information acquisition module is used to collect the basic information of the overdue account from relevant data sources, including the borrowing amount, borrowing date, due repayment date, and actual repayment amount. The overdue account grading module is used to classify the overdue account according to the set grading criteria based on the overdue account data provided by the overdue information acquisition module. The collection strategy formulation module is used to formulate personalized collection methods for different levels of overdue accounts according to the grading results provided by the overdue account grading module. The relative account coordination analysis module is used to analyze the relationship between the overdue account and its relative's account, and assist in judging the repayment ability and repayment intention of the overdue account by obtaining information such as the credit status and repayment history of the relative's account. The multi-account synchronous collection module is used to associate the mobile phone number of the overdue account with its social software, game accounts, and application software accounts at different overdue stages, and conduct collection reminders simultaneously on multiple platforms to increase the probability of the customer being exposed to the collection information. The mobile device behavior data retrieval and analysis module is used to retrieve the behavior data of the mobile device of the overdue user at different overdue stages.

[0006] A computer device for collecting overdue payments, comprising: A processor, a memory, a network interface, an input / output interface, and an operating system and related software modules. The processor is responsible for executing various data processing tasks and adopts a high-frequency and high-cache processor. The network interface is used for the connection of the computer device to an external network to achieve communication with other systems. The input / output interface is used to receive data instructions input by external devices and to output the results processed by the computer device to external devices.

[0007] A storage medium readable by a computer device for overdue payment collection, comprising: Random access memory, read-only memory and storage device, the random access memory is used to provide temporary data storage space for computer equipment, used to store running collection-related programs and data, the read-only memory is used to store fixed programs of the basic input and output system of the computer equipment, these programs are loaded when the computer is started, and provide support for hardware initialization and basic operations of the computer equipment, and the storage device is used for long-term storage of data and programs related to the collection of overdue payments.

[0008] (III) Beneficial effects Compared with the prior art, the present invention provides a method, device, storage medium and computer equipment for collecting overdue payments, which have the following beneficial effects: 1. The overdue payment collection method, device, storage medium and computer equipment can obtain more comprehensive account information by analyzing overdue accounts and related accounts, thereby improving the accuracy of judging the cause of overdue behavior, and the data integration and analysis of multiple related accounts can reduce the one-sidedness of the data, making the prediction and evaluation of overdue behavior more accurate. Based on the accurate overdue behavior data, a more personalized and effective collection strategy can be formulated, and based on the accurate data analysis results, the collection process can be optimized. Instead of using the same collection process for all overdue accounts, differentiated processing is performed according to the risk level and overdue cause of the account, reducing unnecessary collection links and improving overall collection efficiency; 2. The collection method, device, storage medium and computer equipment for overdue payments can be used to simultaneously collect reminders from multiple account channels through a third-party data platform and an account query interface, thereby increasing the probability that the debtor will receive collection information. In addition, in the long-term overdue stage, in-depth collection measures for different accounts will have an impact on the debtor's life or entertainment experience, prompting them to repay as soon as possible to restore normal status. In addition, an authorization request is sent to the overdue user through text messages, APP notifications and file delivery to obtain a list of installed applications in the overdue user's mobile device, so as to more comprehensively understand the behavioral characteristics of the overdue user and provide more basis for collection decisions. In addition, based on the comprehensive risk score of mobile device behavior data, a personalized collection strategy is formulated to improve the targeted collection. At the same time, in this process, high-risk behaviors can be discovered in a timely manner and collection intervention can be carried out in advance to avoid further deterioration of the overdue situation. When it is discovered that a user has downloaded high-risk software and there may be financial risks, timely communication with the user is carried out to prevent the user from falling into more serious financial difficulties and being unable to repay. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1Flowchart of the method for collecting overdue payments of the present invention; Figure 2 Flowchart of the classification method for the method for collecting overdue payments of the present invention; Figure 3 Block diagram of the system structure of the computer device of the present invention. Detailed implementation manners

[0010] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0011] Please refer to Figure 1 , the method for collecting overdue payments includes the following steps: S1 Obtain the overdue information of the overdue account through the collection system. From the database storing loan-related data, according to the preset overdue determination rules, extract the key information of the overdue account, including the borrowing amount, borrowing date, due date, and actual repayment amount. Analyze the historical repayment data through data mining algorithms to identify the repayment behavior pattern. There is a data encryption block and a decryption module added to the database of the account. Before sending a query request, sensitive information in the query conditions will be encrypted to ensure the security of data during transmission. When receiving the query result from the database, the encrypted data in the result will be decrypted for subsequent processing in the collection system; Please refer to Figure 2 , S2 Classify the overdue accounts according to different overdue information; S21 Extract the basic information of the overdue account from the core database of the credit institution Extract the basic information of the overdue account from the core database of the credit institution, including name, ID number, contact information, borrowing amount, and overdue duration, and synchronously collect the transaction records related to the overdue account, including the borrowing release time, repayment plan, and actual repayment record, to analyze its repayment behavior pattern; S22 Associate the relative accounts of the overdue account According to the identity information of overdue accounts, search for related relative accounts in the relational database. The association basis can be registration at the same address, emergency contact setting, and common financial association. For the relative accounts found to be related to the overdue accounts, synchronously collect their basic information and transaction records, including data on income sources and consumption habits, to comprehensively understand the financial status of the relative accounts. According to the data obtained above, clustering analysis method is used to classify accounts with similar overdue behavior patterns, and at the same time, association rule mining is used to discover potential association rules between overdue accounts and relative accounts, and timely discover the relationship between the high-consumption behavior of certain relative accounts and the overdue risk of overdue accounts. Through the above steps, those overdue users and relative users who have no association and are only simply overdue within 1-15 days and over-limit users are classified as first-level overdue users, while those with an association between overdue users and relative users are classified as second-level overdue users. The relative accounts found to be associated with the overdue accounts in the relational database are used for model training with historical data between the two. The model is optimized according to the prediction accuracy and recall rate indicators of the model. Then, the integrated joint account dataset is input into the optimized analysis model to obtain the analysis results of the overdue behavior of the overdue accounts and their relative accounts, so as to output the overdue risk score of the overdue accounts and the potential influencing factors of the relative accounts on the overdue behavior, including the fund transfer situation of the relative accounts. At the same time, the data is compared with the actual overdue collection results to verify the accuracy of the analysis model. When a deviation is found, the deviation information is fed back into the model for adjustment to continuously improve the accuracy of the model in obtaining overdue behavior data; S23 Query the linked accounts of overdue accounts For overdue users with an overdue duration exceeding 15 days obtained in step S21, they are classified as level-three overdue users. Through a third-party data platform and an account query interface, query the information of social software, game accounts, and application software accounts registered under the mobile phone number of the overdue user, and verify the obtained account information to exclude invalid or incorrect accounts. Then, organize and classify the valid account information according to the account type for subsequent operations. Send a request for authorization to obtain mobile device behavior data to the level-three overdue users. The third-party data platform and the account query interface use the uniqueness of the mobile phone number when registering on different platforms to establish an association between the overdue account and various Internet accounts. When querying the third-party data platform, use the mobile phone number as an index to find other account information associated with this mobile phone number, and at the same time establish a relationship mapping model between accounts. When it is analyzed that there is an association between a certain game account and a specific social account in terms of registration information and usage habits, they are classified as belonging to the same user. This relationship mapping helps in more comprehensive information transmission during the collection process. The collection system internally adopts a microservice architecture model, splitting the collection system into multiple independent microservices, including an account query microservice, a collection strategy microservice, and a message sending microservice, and uses containerization technology for deployment to improve the portability and scalability of the system. At the same time, establish a version control system within the collection system to manage the code of the collection system, conduct regular code reviews and vulnerability scans, and promptly repair the discovered problems. There is a data analysis and marking module added inside the collection system. In the case of the Android system, the collection system interacts with the collection system device through the developer interface provided by the Android system to obtain data such as application information. For the Apple system, according to the Apple developer agreement, use iOS development tools and relevant APIs to interact with the collection system device. The data analysis and marking module can start corresponding data collection programs according to the mobile device type of the overdue user, and then transfer the collected data to the analysis module. The analysis module analyzes the data according to preset rules and focuses on marking the behaviors of overdue users who have downloaded or logged in to VPNs or external network high-risk software; S24 Query the behavior data of the overdue account's mobile device For overdue users with an overdue duration exceeding 30 days obtained in step S23, they are classified as level-four overdue users. Send an authorization request to the overdue user by means of text messages, APP notifications, and sending documents, clearly informing the purpose, scope, and confidentiality measures for obtaining data. After the user agrees, for Android devices, use device management tools and cooperate with mobile phone manufacturers to obtain the list of installed applications inside the mobile device of the overdue user, and check whether there are VPN, external network-related applications, or high-risk software marked by the mobile phone's internal security system. Mark the overdue users who have downloaded and logged in to VPNs, external networks, and those with high-risk software as key targets and classify them as level-four high-risk users; S3 Develop different collection methods for overdue accounts at different levels For first-level users, a mild debt collection method is adopted, mainly sending text message reminders to the overdue account holders themselves and recording the basic information of the overdue accounts. For second-level users, a joint debt collection strategy can be formulated to contact both the overdue account and the relative's account simultaneously for repayment negotiation. For third-level users, reminder messages are sent to the social accounts associated with the user's mobile phone number. Based on the initial reminder content, an explanation of the credit impact is added. For the game accounts and application software accounts associated with the user's mobile phone number, in-app notifications can be sent without affecting the normal use of the user. For fourth-level users and fourth-level high-risk users, a more aggressive debt collection method of door-to-door collection is required, and the debt collection frequency is increased and legal means are involved.

[0012] Please refer to Figure 3 This solution also provides an overdue debt collection device. The device includes an information acquisition module, an overdue account grading module, a debt collection strategy formulation module, a relative account coordination and analysis module, a multi-account synchronous debt collection module, and a mobile device behavior data retrieval and analysis module. The information acquisition module is used to collect the basic information of the overdue account from relevant data sources, including the borrowing amount, borrowing date, due date, and actual repayment amount. The overdue account grading module is used to classify the overdue accounts according to the set grading criteria based on the overdue account data provided by the overdue information acquisition module. The debt collection strategy formulation module is used to formulate personalized debt collection methods for different levels of overdue accounts according to the grading results provided by the overdue account grading module. The relative account coordination and analysis module is used to analyze the relationship between the overdue account and its relative's account, and assist in judging the repayment ability and repayment willingness of the overdue account by obtaining information such as the credit status and repayment history of the relative's account. The multi-account synchronous debt collection module is used to associate the mobile phone number of the overdue account with its social software, game accounts, and application software accounts at different overdue stages, and conduct debt collection reminders on multiple platforms simultaneously to increase the probability of the customer being exposed to the debt collection information. The mobile device behavior data retrieval and analysis module is used to retrieve the behavior data of the overdue user's mobile device at different overdue stages.

[0013] Based on the overdue debt collection device, this solution provides a computer device. The device includes a processor, a memory, a network interface, an input / output interface, and an operating system and related software modules. The processor is responsible for executing various data processing tasks and uses a high-frequency and high-cache processor. The network interface is used for the connection of the computer device to the external network to achieve communication with other systems. The input / output interface is used to receive data instructions input by external devices and to output the results processed by the computer device to external devices.

[0014] In addition, the computer device of the present solution is internally provided with a readable storage medium, which includes a random access memory, a read-only memory and a storage device. The random access memory is used to provide temporary data storage space for the computer device, and is used to store the collection-related programs and data being run. The read-only memory is used to store fixed programs of the basic input and output system of the computer device. These programs are loaded when the computer is started to provide support for the hardware initialization and basic operations of the computer device. The storage device is used for long-term storage of data and programs related to the collection of overdue payments.

[0015] In summary, this solution has successfully achieved the ability to obtain more comprehensive account information through a series of innovative designs and intelligent management strategies such as joint analysis of overdue accounts and related accounts and third-party data platforms and account query interfaces for simultaneous collection reminders on multiple account channels, thereby improving the accuracy of judging the causes of overdue behavior. The data integration and analysis of multiple related accounts can reduce the one-sidedness of the data, making the prediction and evaluation of overdue behavior more accurate. Based on accurate overdue behavior data, more personalized and effective collection strategies can be formulated. Based on accurate data analysis results, the collection process can be optimized. Instead of using the same collection process for all overdue accounts, differentiated processing is performed based on the account's risk level and overdue cause, reducing unnecessary collection links, improving overall collection efficiency, and increasing debt. The probability of a person receiving a collection message is reduced, and in the long-term overdue stage, in-depth collection measures for different accounts will have an impact on the debtor's life or entertainment experience, prompting them to repay as soon as possible to restore normal status, and send authorization requests to the overdue user through SMS, APP notifications and sending documents to obtain the list of installed applications in the overdue user's mobile device, so as to have a more comprehensive understanding of the behavioral characteristics of the overdue user and provide more basis for collection decisions. In addition, based on the comprehensive risk score of mobile device behavior data, personalized collection strategies are formulated to improve the targeted nature of collection. At the same time, in this process, high-risk behaviors can be discovered in time and collection intervention can be carried out in advance to avoid further deterioration of the overdue situation. When it is discovered that a user has downloaded high-risk software and there may be financial risks, timely communication with the user is carried out to prevent the user from falling into more serious financial difficulties and being unable to repay.

[0016] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0017] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for collecting overdue payments, characterized in that, It includes the following steps: S1 Obtain the overdue information of overdue accounts through the collection system From the database storing loan-related data, according to the preset overdue judgment rules, extract the key information of overdue accounts, including the borrowing amount, borrowing date, due date, and actual repayment amount. Analyze the historical repayment data through data mining algorithms to identify the repayment behavior patterns; S2 Classify the overdue accounts according to different overdue information; S21 Extract the basic information of overdue accounts from the core database of the credit institution Including name, ID number, contact information, borrowing amount, and overdue duration, and synchronously collect the transaction records related to the overdue accounts, including the loan disbursement time, repayment plan, and actual repayment records, to analyze their repayment behavior patterns; S22 Associate the related accounts of the overdue accounts According to the identity information of the overdue accounts, search for related related accounts through the relational database. The association basis can be registered at the same address, emergency contact setting, and common financial association. For the related accounts found related to the overdue accounts, synchronously collect their basic information and transaction records, including data on income sources and consumption habits, to comprehensively understand the financial status of the related accounts. According to the data clustering analysis method obtained above, it is used to classify the accounts with similar overdue behavior patterns, and at the same time, association rule mining is used to discover the potential association rules between the overdue accounts and the related accounts, and timely discover the relationship between the high consumption behavior of some related accounts and the overdue risk of the overdue accounts. Through the above steps, the overdue users and the related users are not associated, and only the users who are simply overdue within 1-15 days and over the limit are classified as first-level overdue users, while the overdue users and the related users who are associated are classified as second-level overdue users; S23 Query the linked accounts of the overdue accounts For the overdue users whose overdue duration obtained in step S21 exceeds 15 days, they are classified as third-level overdue users. Through the third-party data platform and the account query interface, query the information of the social software, game accounts, and application software accounts registered under the mobile phone number of the overdue user, and verify the obtained account information, exclude invalid or incorrect accounts, and sort and classify the valid account information according to the account type for subsequent operations, and send a request for authorization to obtain the mobile device behavior data to the third-level overdue users; S24 Query the behavior data of the mobile devices of the overdue accounts For the overdue users whose overdue duration obtained in step S23 exceeds 30 days, they are classified as fourth-level overdue users, and send an authorization request to the overdue user by means of text message, APP notification, and sending documents, clearly informing the purpose, scope, and confidentiality measures of obtaining the data. After the user agrees, for Android devices, use the device management tool and cooperate with the mobile phone manufacturer to obtain the list of installed applications inside the mobile device of the overdue user, check whether there are VPN, external network-related applications, or high-risk software marked by the internal security system of the mobile phone, and focus on marking the overdue users who have downloaded and logged in to VPN, external network, and have high-risk software, and classify them as fourth-level high-risk users; S3 Develop different collection methods for overdue accounts at different levels For first-level users, a mild collection method is adopted, mainly sending SMS reminders to the overdue account holder himself / herself, and recording the basic information of the overdue account; For second-level users, a joint collection strategy can be formulated to contact both the overdue account and the relative's account simultaneously for repayment negotiation; For third-level users, reminder messages are sent to the social accounts associated with the user's mobile phone number. On the basis of the initial reminder, the message content is increased with an explanation of the credit impact. For the game accounts and application software accounts associated with the user's mobile phone number, in the premise of not affecting the normal use of the user, in-site notification reminders can be sent; For fourth-level users and fourth-level high-risk users, a more aggressive collection method of door-to-door collection is required, and the collection frequency and legal means of intervention are increased.

2. The method for collecting overdue payments according to claim 1, wherein: A data encryption block and a decryption module are added to the account database. Before sending a query request, sensitive information in the query conditions will be encrypted to ensure the security of data during transmission. When the query result is received from the database, the encrypted data in the result will be decrypted for subsequent processing in the collection system.

3. The overdue payment collection method according to claim 1, characterized in that: For the relative's account found by the relational database that is associated with the overdue account, historical data is used for model training between the two. The model is optimized according to the prediction accuracy and recall rate indicators of the model. Then the integrated joint account dataset is input into the optimized analysis model to obtain the overdue behavior analysis results of the overdue account and its relative's account, so that the overdue risk score of the overdue account can be output, as well as the potential influencing factors of the relative's account on the overdue behavior, including the fund transfer situation of the relative's account. At the same time, the data is compared with the actual overdue collection results to verify the accuracy of the analysis model. When a deviation is found, the deviation information is fed back to the model for adjustment to continuously improve the accuracy of the model in obtaining overdue behavior data.

4. The method for collecting overdue payments according to claim 1, characterized in that: The third-party data platform and the account query interface use the uniqueness of the mobile phone number when registering on different platforms to establish an association between the overdue account and various Internet accounts. When querying the third-party data platform, the mobile phone number is used as an index to find other account information associated with the mobile phone number, and a relationship mapping model between accounts is established. When it is analyzed whether a certain game account is associated with a specific social account in terms of registration information and usage habits and belongs to the same user, this relationship mapping helps to carry out more comprehensive information transmission during the collection process.

5. The collection method for overdue payments according to claim 1, characterized in that: The collection system internally adopts a microservices architecture mode, splitting the collection system into multiple independent microservices, including an account query microservice, a collection strategy microservice, and a message sending microservice, and using containerization technology for deployment to improve the portability and scalability of the system. At the same time, a version control system is established within the collection system to manage the code of the collection system, and code reviews and vulnerability scans are carried out regularly to repair the problems found in a timely manner.

6. The collection method of overdue payments according to claim 1, characterized in that: The debt collection system is internally equipped with a data analysis and marking module. In the case of the Android system, the debt collection system interacts with the debt collection system device through the developer interface provided by the Android system to obtain application information data. In the case of the Apple system, according to the Apple developer agreement, iOS development tools and related APIs are used to interact with the debt collection system device.

7. The collection method for overdue payments according to claim 1, characterized in that: The data analysis and marking module can start corresponding data collection programs according to the mobile device types of overdue users, and then transfer the collected data to the analysis module. The analysis module analyzes the data according to preset rules and focuses on marking the behaviors of overdue users who are found to download or log in to VPNs or high-risk external software.

8. A collection device for overdue payments, comprising the method for collecting overdue payments as claimed in claims 1-7, characterized in that: The debt collection system is internally equipped with an information acquisition module, an overdue account grading module, a debt collection strategy formulation module, a relative account coordination and analysis module, a multi-account synchronous debt collection module, and a mobile device behavior data retrieval and analysis module. The information acquisition module is used to collect basic information of overdue accounts from relevant data sources, including the borrowing amount, borrowing date, due repayment date, and actual repayment amount. The overdue account grading module is used to classify overdue accounts according to the grading criteria set based on the overdue account data provided by the overdue information acquisition module. The debt collection strategy formulation module is used to formulate personalized debt collection methods for different levels of overdue accounts according to the grading results provided by the overdue account grading module. The relative account coordination and analysis module is used to analyze the relationship between the overdue account and its relative accounts, and assist in judging the repayment ability and repayment willingness of the overdue account by obtaining information on the credit status and repayment history of the relative accounts. The multi-account synchronous debt collection module is used to associate the mobile phone number of the overdue account with its social software, game accounts, and application software accounts at different overdue stages, and conduct debt collection reminders on multiple platforms simultaneously to increase the probability of customers being exposed to debt collection information. The mobile device behavior data retrieval and analysis module is used to retrieve the behavior data of the mobile devices of overdue users at different overdue stages.

9. A computer device for collecting overdue payments, including the method for collecting overdue payments as claimed in claims 1 - 7, characterized in that: It includes a processor, a memory, a network interface, an input / output interface, an operating system, and software modules. The processor is responsible for executing data processing tasks and uses a high-frequency and high-cache processor. The network interface is used for the connection of the computer device to the external network to achieve communication with other systems. The input / output interface is used to receive data instructions input by external devices and output the results processed by the computer device to external devices.

10. A storage medium, including a storage medium readable in the overdue payment collection computer device as described in claim 9, characterized in that: It includes a random access memory, a read-only memory, and a storage device. The random access memory is used to provide data storage space for the computer device, and is used to store the running debt collection-related programs and data. The read-only memory is used to store the fixed programs of the basic input / output system of the computer device, and these programs are loaded when the computer starts to provide support for the hardware initialization and basic operations of the computer device. The storage device is used to store data and programs related to the collection of overdue payments.

Images