Defective asset complete dispatching method, device and equipment and medium

Abstract

The invention discloses a method, a device, equipment and a medium for the exhaustion of unhealthy assets. The method comprises the following steps: acquiring to-be-processed asset data of a debtor, and performing structured processing on the to-be-processed asset data; carrying out preliminary exhaustion on the structured asset data; performing re-inspection processing on the asset data after the preliminary complete scheduling to obtain processed data; inputting the processed data into a standardized exhaustion dispatch model to obtain an exhaustion dispatch guidance scheme; the standardized exhaustion scheduling model comprises model data, the model data is stored in a database after a rule matching algorithm is converted into a general strategy which can be identified by a computer, and the rule matching algorithm is constructed by analyzing historical unhealthy asset data; and sorting out all-in-one content obtained after the all-in-one guiding scheme is passed, and generating an all-in-one report of the to-be-processed asset data. According to the scheme, the exhaustion call element structuring can be realized, various information of the debtor is more comprehensively considered, the generated exhaustion call report is more accurate, and the working efficiency is further improved.

Description

Technical Field

[0001] The present invention generally relates to the field of asset management technology, and specifically relates to a due diligence method, device, equipment and medium for non-performing assets. Background Art

[0002] With the continuous development and innovation of the financial market, the problem of non-performing assets of banks has become increasingly prominent. Non-performing assets mainly refer to financial assets that cannot generate normal interest income or are even difficult to recover costs in the process of financial institutions carrying out credit business. In the process of receiving and managing non-performing assets, due diligence is an important link. It is particularly important to evaluate and manage non-performing assets, which can comprehensively evaluate the quality, value and possibility of recovery of non-performing assets, and provide a basis for subsequent disposal decisions. In order to prevent and control risks and ensure the healthy development of banking business, it is particularly important to conduct due diligence on non-performing assets.

[0003] Currently, the relevant technologies implement due diligence on non-performing assets through data integration, automated document review, and generation of due diligence reports. However, this solution has some data missing, which makes the generated due diligence conclusions relatively simple and confusing, resulting in low accuracy of the due diligence results on non-performing assets. Summary of the invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a due diligence method, device, equipment and medium for non-performing assets.

[0005] In a first aspect, the present invention provides a due diligence method for non-performing assets, the method comprising:

[0006] Acquiring the debtor's asset data to be processed and performing structured processing on the asset data to be processed to obtain structured asset data;

[0007] Conducting preliminary due diligence on the structured asset data;

[0008] Performing a back-check on the asset data of the preliminary due diligence to obtain processed data;

[0009] Input the processed data into a standardized due diligence model to obtain a due diligence guidance plan; the standardized due diligence model includes model data, and the model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy, and the rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data;

[0010] The due diligence content obtained after the due diligence guidance plan is sorted out to generate a due diligence report on the asset data to be processed.

[0011] In one embodiment, the asset data of the preliminary due diligence is back-checked to obtain processed data, including:

[0012] Using due diligence review to determine whether there are any data element updates in the asset data of the preliminary due diligence, where the data element updates include data element changes or data element additions;

[0013] When there is a change in data elements, the changed data elements are obtained and added to the asset data of the preliminary due diligence to obtain the changed asset data;

[0014] When there is a requirement to add new data, obtain the new data elements and add the new data elements to the asset data of the preliminary due diligence to obtain the new asset data;

[0015] Based on the changed asset data and the newly added asset data, the processed data is obtained.

[0016] In one embodiment, the structured processing of the asset data to be processed includes:

[0017] Performing data cleaning and verification processing on the asset data to be processed to obtain verified asset data;

[0018] Determining unstructured data from the verified asset data, and converting the unstructured data into structured data using a natural language processing method;

[0019] The verified asset data and the structured data are sorted according to a standard format to obtain the structured asset data.

[0020] In one embodiment, the processed data is input into a standardized due diligence model to obtain a due diligence guidance plan, including:

[0021] Obtaining creditor's rights information, principal debtor information, guarantor information, mortgage information, and property clue information from the processed data;

[0022] The credit information, principal debtor information, guarantor information, mortgage information, and property clue information are analyzed through the standardized due diligence model to analyze the credit situation, principal debtor, guarantor, mortgage, and property clues, and generate a due diligence guidance plan; the due diligence guidance plan includes any one of the following: due diligence methods and due diligence content.

[0023] In one embodiment, the standardized due diligence model is constructed in the following manner:

[0024] Acquire historical non-performing asset data; the historical non-performing asset data includes multiple information categories;

[0025] Verifying, correcting and structuring the historical non-performing asset data to obtain processed historical data;

[0026] Converting the processed data into a preset format;

[0027] Under the preset format, performing feature extraction processing on the processed data to determine key features, and performing standardization and normalization processing on the key features to obtain processed key features;

[0028] Dividing the processed key features into a training set, a validation set, and a test set; the training set is annotated with historical due diligence results;

[0029] The initial model is trained by using the training set and the validation set to obtain a due diligence model to be verified, and the performance of the due diligence model to be verified is tested by using the test set, and the due diligence model to be verified with the best performance test is used as the standardized due diligence model.

[0030] In one embodiment, the historical non-performing asset data is verified, corrected and structured to obtain processed historical data, including:

[0031] Verify the historical non-performing asset data to determine whether the historical non-performing asset data is complete, consistent and accurate;

[0032] When the integrity, consistency and accuracy are met, the historical non-performing asset data is checked and processed, abnormal data is determined and corrected; the abnormal data includes any of the following: wrong data, invalid data, missing data;

[0033] The unstructured data in the non-performing asset data is converted into structured data using a natural language processing method to obtain the processed historical data.

[0034] In one embodiment, the initial model is trained by using the training set and the verification set to obtain a due diligence model to be verified, including:

[0035] Inputting the training set into the initial model to obtain an output result;

[0036] Constructing a loss function according to the output result and the historical due diligence results corresponding to the training set;

[0037] According to minimizing the loss function, iteratively training the initial model to obtain the due diligence model to be trained;

[0038] The verification set is input into the due diligence model to be trained to obtain the due diligence model to be verified.

[0039] In a second aspect, an embodiment of the present application provides a due diligence device for non-performing assets, the device comprising:

[0040] An acquisition module, used for acquiring the debtor's asset data to be processed and performing structured processing on the asset data to be processed to obtain structured asset data;

[0041] A preliminary due diligence module, used to conduct a preliminary due diligence on the structured asset data;

[0042] A backcheck processing module is used to perform backcheck processing on the asset data of the preliminary due diligence to obtain processed data;

[0043] A due diligence guidance module, used for inputting the processed data into a standardized due diligence model to obtain a due diligence guidance scheme; the standardized due diligence model includes model data, and the model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy, and the rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data;

[0044] The report generation module is used to organize the due diligence content obtained after the due diligence guidance plan is passed, and generate a due diligence report on the asset data to be processed.

[0045] In a third aspect, an embodiment of the present application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the due diligence method for non-performing assets provided in the above embodiment.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the due diligence method for non-performing assets provided in the above-mentioned embodiment.

[0047] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0048] The embodiments of the present application provide a due diligence method, device, equipment and medium for non-performing assets. The method includes: obtaining the debtor's pending asset data and performing structured processing on the pending asset data to obtain structured asset data, performing preliminary due diligence on the structured asset data, performing back-checking processing on the preliminary due diligence asset data to obtain processed data, and then inputting the processed data into a standardized due diligence model to obtain a due diligence guidance scheme. The standardized due diligence model includes model data. The model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy. The rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data, and then the due diligence content obtained after the due diligence guidance scheme is sorted out to generate a due diligence report on the pending asset data. Compared with the existing technology, this technical solution can achieve structuring of due diligence elements by structuring the asset data to be processed, converting the originally scattered and disordered data into structured data, and rechecking the asset data of the preliminary due diligence. It can conduct a comprehensive review of the asset data that has undergone preliminary due diligence, so as to quickly find the changed or newly added data, making the processed data obtained more comprehensive. In addition, since a standardized due diligence model is constructed through the analysis of historical non-performing asset data, the processed data can be processed through the standardized due diligence model, which can consider various information of the debtor more comprehensively, thereby extracting the true situation of the debtor's assets more accurately and comprehensively, making the determined due diligence guidance plan more accurate, thereby improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 This is an application environment diagram of the due diligence method for non-performing assets provided in one embodiment of the present application;

[0051] Figure 2 A flowchart of a due diligence method for non-performing assets provided in one embodiment of the present application;

[0052] Figure 3 A schematic diagram of a process for forming a standardized due diligence model provided in one embodiment of the present application;

[0053] Figure 4 A schematic diagram of a structure for forming a standardized due diligence model provided in one embodiment of the present application;

[0054] Figure 5A schematic diagram of functional modules of a due diligence device for non-performing assets provided by another embodiment of the present application;

[0055] Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only 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.

[0057] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0058] As mentioned in the background technology, in the related technology, due diligence on non-performing assets is achieved through data integration, automated document review, and generation of due diligence reports. However, this solution has some data missing, which makes the generated due diligence conclusions relatively simple and confusing, resulting in low accuracy of the due diligence results on non-performing assets.

[0059] Based on the above-mentioned defects, the present application provides a due diligence method for non-performing assets. Compared with the prior art, the technical solution can achieve structuring of due diligence elements by structuring the asset data to be processed, converting the originally scattered and disordered data into structured data, and rechecking the asset data of the preliminary due diligence. It can conduct a comprehensive review of the asset data that has undergone preliminary due diligence, so as to quickly find the changed or newly added data, making the acquired processed data more comprehensive, and because a standardized due diligence model is constructed through the analysis of historical non-performing asset data, the processed data can be processed through the standardized due diligence model, which can consider various information of the debtor more comprehensively, thereby more accurately and comprehensively extracting the true situation of the debtor's assets, making the determined due diligence guidance plan more accurate, thereby improving work efficiency.

[0060] The due diligence method for non-performing assets provided in the embodiments of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the debtor's pending asset data to the server 104. After receiving the debtor's pending asset data, the server 104 structures, conducts preliminary due diligence and back-check processing on the pending asset data to obtain the processed data, and inputs the processed data into the standardized due diligence model to obtain the due diligence guidance plan, and organizes the due diligence content that passes the due diligence guidance plan to generate a corresponding due diligence report, thereby providing feedback to the terminal 102. In addition, in some embodiments, the due diligence method for non-performing assets can also be implemented by the server 104 or the terminal 102 alone, such as the terminal 102 can directly conduct due diligence processing on the pending asset data, or the server 104 can obtain the pending asset data from the data storage system and conduct due diligence processing on the non-performing assets on the pending asset data.

[0061] Optionally, the standardized due diligence model can be constructed by the server 104 or the terminal 102 based on analysis of historical non-performing asset data.

[0062] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0063] In an exemplary embodiment, Figure 2 As shown, a method for due diligence of non-performing assets is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the method, which includes the following steps S201 to S205. Among them:

[0064] Step S201, obtaining the debtor's asset data to be processed and performing structured processing on the asset data to be processed to obtain structured asset data.

[0065] It should be noted that the pending asset data of the above-mentioned debtors are data that require due diligence, and the pending asset data may include various types of information, such as: basic information of the debtor, debt situation, analysis of the principal debtor, information of the guarantor, pledge situation, repayment ability, and other relevant information. Among them, the basic information of the debtor includes: name / title, contact information, ID number / unified social credit code, etc.; the creditor's rights situation includes: total creditor's rights, due date, amount repaid, risk situation (overdue days, amount owed), analysis of the principal debtor includes the principal debtor's financial situation (balance sheet, profit and loss statement), business operation status, industry background and market position; guarantor information includes: qualifications of legal person guarantors, personal financial situation of natural person guarantors, and details of guarantee contracts; pledge situation includes: assessed value of mortgage / pledge, property status (property clarity, legal disputes), pledge ratio; collection litigation situation includes: litigation progress, relevant legal documents; repayment ability includes: repayment history, income stability; other relevant information includes: related party relationships, illegal or unethical behavior records.

[0066] Optionally, the data on the pending assets of the above-mentioned debtors can be obtained through an existing database, such as obtaining the customer's past transaction records, credit data, etc. through an internal system; it can also be obtained by using crawler tools to obtain the latest corporate developments, legal litigation information, etc. from public websites and announcements; it can also be obtained by verification and entry by professionals.

[0067] In this embodiment, since the debtor's pending asset data is obtained, good data guidance information can be provided for the subsequent data review and preprocessing.

[0068] Furthermore, after obtaining the asset data to be processed, structured data and unstructured data can be determined from the asset data to be processed. Structured data refers to asset data that has been processed in a structured manner, and unstructured data refers to asset data that has not been processed in a structured manner. Then, the structured asset data can be cleaned and verified. For example, data can be extracted, transformed and loaded using ETL tools, and data quality tools can be used to check and correct missing values, abnormal values, duplicates, etc. For unstructured data, natural language processing can be used to process it, extract and parse key information, and convert it into structured data, thereby realizing automated entity recognition, sentiment analysis and other functions, assisting in understanding the essence of the content, and organizing the verified asset data and structured data in a standard format, and obtaining structured asset data by adjusting all data items. Among them, unstructured data can be, for example, contract texts, reports, etc.

[0069] Optionally, in the process of structuring the asset data to be processed and obtaining structured asset data, the asset data to be processed may be first cleaned and verified to obtain the verified asset data, and unstructured data may be determined from the verified asset data, and the unstructured data may be converted into structured data using natural language processing, and then the verified asset data and structured data may be organized in a standard format to obtain structured asset data.

[0070] It is understandable that in the process of arranging in accordance with the standard format, standardized specifications and data dictionaries can be used to define the naming rules, units, formats and data types of data fields, and different data can be standardized according to standardized specifications and data dictionaries, so as to ensure the consistency and comparability of data in the entire due diligence process. Among them, in the structured due diligence stage, the creditor's rights, principal debtor, guarantor (legal person, natural person), and mortgage and pledge are analyzed, and specifically subdivided into the creditor's rights risk situation, collection litigation situation, repayment ability, legal attributes of guarantees, mortgage and pledge property status, etc., and key fields are conditionally constrained.

[0071] In this embodiment, by performing structured processing on multiple asset data to be processed, the data can be standardized so that the structured asset data conforms to the standard format, further improving the efficiency of due diligence.

[0072] Step S202: Conduct preliminary due diligence on the structured asset data.

[0073] Step S203, recheck the asset data of the preliminary due diligence to obtain processed data.

[0074] It should be noted that the asset data of the preliminary due diligence mentioned above is structured asset data after the preliminary due diligence processing, the processed data is the asset data of the preliminary due diligence after the back-check processing, and the processed data is the asset data to be processed after the structured processing and the back-check processing in sequence. The processed data are all structured data, and the asset data to be processed may include structured data and unstructured data. Due diligence back-check is an important guarantee mechanism in the operation of the standardized due diligence model. As an important guarantee mechanism in the operation of the non-performing asset data due diligence model, it is designed to cope with the changes or additions of data elements that may occur over time. The core of this mechanism is that the computer system can regularly and automatically conduct a comprehensive review of the collected due diligence data to verify the integrity and timeliness of the data.

[0075] Specifically, after obtaining the structured asset data, a preliminary due diligence process can be performed on the structured asset data. For example, a preset due diligence method can be used to perform preliminary due diligence process to obtain preliminary due diligence asset data, and a backcheck algorithm can be used to backcheck the preliminary due diligence asset data to determine the objectives and scope of the due diligence backcheck, and determine the key risk points and control measures that need to be evaluated. The backcheck algorithm can be an algorithm that can be recognized by a pre-built computer language.

[0076] Optionally, in the process of rechecking the asset data of the preliminary due diligence to obtain the rechecked data, a due diligence recheck method can be used to determine whether there are any data element updates in the asset data of the preliminary due diligence, and the data element updates include data element changes or data element additions; when there are data element changes, the changed data elements are obtained and the changed data elements are added to the asset data of the preliminary due diligence to obtain the changed asset data; when there are data requirements for addition, the newly added data elements are obtained and the newly added data elements are added to the asset data of the preliminary due diligence to obtain the newly added asset data; based on the changed asset data and the newly added asset data, the processed data is obtained.

[0077] In this step, by rechecking the asset data of the preliminary due diligence, in-depth data mining and analysis are achieved, and the asset data of the preliminary due diligence is inspected regularly and automatically. In the due diligence process, new data is continuously collected and compared with the previous asset data of the preliminary due diligence. When it is identified that there is a change in the data element, the changed data element is obtained and the changed data element is added to the asset data of the preliminary due diligence in a timely manner to obtain the changed asset data, which may include information status changes or information content changes; when it is identified that a new data element is added, the new data element is obtained and added to the asset data of the preliminary due diligence to obtain the newly added asset data, and the changed asset data and the newly added asset data are sorted to obtain the processed data.

[0078] During the review process, when there is data that has not been updated for more than a preset time, an update prompt instruction can be generated and sent to the terminal of the business personnel, so that the business personnel pay attention to the unupdated data and update it in time. The update prompt instruction is used to prompt the business personnel to update the data in time.

[0079] For example, taking the debtor A as an example, assuming that during the initial due diligence structuring process, due to the time limit for information acquisition and other reasons, some litigation information could not be obtained. This part of the litigation information may include the hearing time, so that some information is missing in the generated due diligence report. In this embodiment, after obtaining the asset data of the preliminary due diligence, the system regularly and automatically conducts a comprehensive review of the included asset data to be processed, and focuses on reviewing such data as litigation information, so that in the subsequent data capture or update process of the system, it is detected that there is litigation data that can be supplemented and updated. For example, if it is detected that the hearing time of the debtor A's litigation is March 7, 2020, the data supplement process will be automatically triggered to supplement the litigation information to the original data structure, thereby preventing information loss and ensuring the comprehensiveness and accuracy of the data.

[0080] Among them, during the due diligence review process, if no change record of the due diligence information related to debtor A is detected, the system will automatically trigger the update reminder mechanism according to the pre-set configuration rules and logic, and send an update reminder instruction to the corresponding business personnel to prompt the business personnel to update the process. The business personnel can be, for example, a designated account manager or business handling personnel. In this embodiment, by sending an update reminder instruction to the business personnel, it is convenient to remind the relevant business personnel to follow up and supplement the debt information of debtor A in a timely manner to ensure the timeliness and accuracy of subsequent business operations. By setting up this update reminder mechanism, not only the direction and execution of the work instructions are enhanced, but also the real-time update and integrity of the business data in the system are effectively guaranteed, thereby improving the efficiency and quality of overall business management.

[0081] In this embodiment, by conducting a due diligence review on the asset data of the preliminary due diligence, the collected asset data can be comprehensively reviewed regularly and automatically, thereby verifying the effectiveness and stability of the non-performing asset due diligence model. The review operation also facilitates the identification of potential problems in the asset data, thereby better understanding potential risk points and formulating corresponding risk management strategies, further improving the quality of decision-making.

[0082] Step S204, input the processed data into the standardized due diligence model to obtain a due diligence guidance plan; the standardized due diligence model includes model data, and the model data is stored in the database after converting the rule-based wildcard algorithm into a computer-recognizable general strategy. The rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data.

[0083] It should be noted that the above-mentioned standardized due diligence model is a pre-trained model for due diligence on non-performing asset data. The standardized due diligence model can be a machine learning model or other models for due diligence on non-performing asset data. The due diligence model can be understood as the data elements in the due diligence report, and standardization can be understood as the unification of data elements. The due diligence guidance plan can be the due diligence method and due diligence content. The due diligence method refers to interviews, review of information reports, online inquiries, on-site visits, etc. The due diligence content is the debtor's credit information, the default limit, the seizure of collateral, property clues, etc. Specifically, the due diligence guidance method can be decomposed into guidance actions, such as reviewing post-loan reports, interviewing account managers, post-loan management personnel, debtors, etc.

[0084] In the process of inputting the processed data into the standardized due diligence model to obtain the due diligence guidance plan, the creditor's rights information, principal debtor's information, guarantor's information, mortgage and pledge information, and property clue information can be obtained from the processed data, and the creditor's rights information, principal debtor's information, guarantor's information, mortgage and pledge information, and property clue information are analyzed through the standardized due diligence model to conduct creditor's rights analysis, principal debtor analysis, guarantor analysis, mortgage and pledge analysis, and property clue analysis to generate a due diligence guidance plan; the due diligence guidance plan includes any of the following: due diligence methods and due diligence content.

[0085] In the due diligence guidance process, it can be based on computer language programs and scheduling task executors, according to the unified and standardized data element system of the model, through preset logical rules and algorithms, to analyze the processed data, and use historical cases and industry experience to conduct in-depth evaluation, so as to implement precise and systematic recommendation and guidance strategies for corporate users. This process aims to maximize the improvement of data elements in the due diligence model and ensure the comprehensiveness and depth of due diligence work.

[0086] For example, taking debtor A as an example, the corresponding non-performing asset data is the asset data to be processed of debtor A. When it is first included in the scope of the standardized due diligence model, the computer equipment will call the preset standardized due diligence model and start the automated processing process. According to the established rule set and configuration information in the standardized due diligence model, the preliminary setting of the due diligence plan and the initialization processing of data elements are completed. On this basis, the system will further analyze the processed asset data, combine historical cases and industry experience, and recommend a customized due diligence guidance plan for debtor A through multiple means such as analysis of processed data and comparison of industry case studies. The due diligence guidance plan can be used for reference and implementation by corporate account managers or agents.

[0087] In this embodiment, by passing the processed data through a standardized due diligence model, it is possible to achieve a comprehensive review of the debtor's archival materials, on-site investigation and evaluation of collateral, confirmation of valuation pricing, and collection and processing of special information and other key aspects, thereby providing a detailed and practical due diligence work guide for account managers or handling personnel. Through this due diligence guidance program, it is convenient for business personnel to be more clear about the focus and direction of due diligence, avoid blind investigation and invalid work of data collection, and thus greatly improve work efficiency. At the same time, business personnel can also refer to industry cases and risk assessment information in the program to help them more accurately judge the credit status and business potential of the debtor, providing strong support for subsequent business decisions. Therefore, the recommendation process of the due diligence guidance program is not only efficient and fast, but also can significantly improve the pertinence and effectiveness of due diligence work, and further ensure the authenticity and reliability of due diligence results.

[0088] Step S205, sorting out the due diligence content obtained through the due diligence guidance plan, and generating a due diligence report for the asset data to be processed.

[0089] It should be noted that the above-mentioned due diligence report is a report generated according to the due diligence guidance plan. The due diligence report can be presented in one or more ways. The inflow can be presented in the form of text alone, or in the form of pictures alone, or in the form of tables alone, or in the form of a combination of any two of text, pictures, and tables, or in the form of a combination of text, tables, and pictures.

[0090] Optionally, after generating the due diligence guidance plan, the due diligence methods and due diligence contents in the due diligence guidance plan may be sorted out, a mapping relationship between the due diligence methods, due diligence contents and the debtor's pending asset information may be established, and then a due diligence report may be generated after analysis and sorting. The due diligence report may include multiple data elements, which may include: credit information, guarantee information, collection records, debt information, warning information, litigation information, etc.

[0091] For example, taking the debtor A as an example, after obtaining the asset data to be processed, it can be first structured, including the initial inspection operation, and the field structured check of the asset data to be processed, and automatically fill in the existing debtor information in the bank, so as to improve the efficiency of information collection, avoid manual input errors, and reduce the input workload; then call the standardized due diligence model, read the model strategy in the database through the query interface and parse it, and realize the structured due diligence of key elements, so as to generate a standardized and normalized due diligence report. When the query request sent by the user terminal is received, the obtained due diligence report can be rendered to the user terminal through the query interface. In this embodiment, by first performing structured processing on the asset data to be processed, customer tags can be automatically generated for easy management, and each household can be accurately classified according to the structured due diligence results to achieve the goal of smart household management and strengthen the strategic orientation.

[0092] Compared with the initial due diligence on the asset data to be processed, due to changes in debtor A's repayment ability, creditor's rights risk situation, and the nature and status of the collateral after a few months, and the release of relevant new policies in the macro market environment, it is necessary to collect new customer information at this time, and formulate corresponding asset disposal measures based on the current situation of debtor A. The computer equipment calls the standardized due diligence model, adjusts the data segmentation, key fields and constraints for data types such as creditor's rights, principal debtors, guarantors, and collateral, and increases the analysis of property clues. According to the changes in the due diligence content, the label classification and disposal strategy classification results are displayed in real time, and the asset value is adjusted dynamically, so that the due diligence results can continuously feed back the strategy precision recommendation and improve the accuracy of decision-making.

[0093] In this embodiment, by organizing the due diligence guidance plan, the efficiency and speed of due diligence can be improved, and the data can be found and updated in time through the data backcheck mechanism, making the standardized due diligence processing more targeted and effective, ensuring that business personnel can obtain true and reliable due diligence results, and providing guarantees for the smooth advancement of business and risk control.

[0094] The embodiment of the present application provides a due diligence method for non-performing assets, which includes: obtaining the debtor's pending asset data and performing structured processing on the pending asset data to obtain structured asset data, performing preliminary due diligence on the structured asset data, performing back-checking processing on the preliminary due diligence asset data to obtain processed data, and then inputting the processed data into a standardized due diligence model to obtain a due diligence guidance scheme. The standardized due diligence model includes model data, and the model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy. The rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data, and then the due diligence content obtained after the due diligence guidance scheme is sorted out to generate a due diligence report on the pending asset data. Compared with the existing technology, this technical solution can achieve structuring of due diligence elements by structuring the asset data to be processed, converting the originally scattered and disordered data into structured data, and rechecking the asset data of the preliminary due diligence. It can conduct a comprehensive review of the asset data that has undergone preliminary due diligence, so as to quickly find the changed or newly added data, making the processed data obtained more comprehensive. In addition, since a standardized due diligence model is constructed through the analysis of historical non-performing asset data, the processed data can be processed through the standardized due diligence model, which can consider various information of the debtor more comprehensively, thereby extracting the true situation of the debtor's assets more accurately and comprehensively, making the determined due diligence guidance plan more accurate, thereby improving work efficiency.

[0095] In another exemplary embodiment of the present application, in order to accurately perform due diligence on the asset data to be processed and generate a due diligence report, it is necessary to build a standardized due diligence model. The embodiment of the present application provides a specific implementation method for building a standardized due diligence model, such as Figure 3 As shown, the method may include the following steps:

[0096] Step S301, obtaining historical non-performing asset data; the historical non-performing asset data includes multiple information categories.

[0097] It should be noted that the above historical non-performing asset data can be the non-performing asset data with known due diligence results, and can include multiple information categories, namely, bond status, principal debtor status, guarantor information, mortgage information, etc. And there are corresponding sub-items for each information category, for example, it can include bond risk status, collection litigation status, repayment ability, legal attributes of guarantee, and mortgage status.

[0098] Optionally, the above historical non-performing asset data can be obtained through banks, courts, credit reporting agencies, auction platforms, etc., or can be obtained through a third-party agency through a data sharing agreement.

[0099] Step S302, verifying, correcting and structuring the historical non-performing asset data to obtain processed historical data.

[0100] See also Figure 4 As shown in the figure, after obtaining historical non-performing asset data, the data elements of due diligence can be structured, the data element standards can be unified, and the preset algorithms supported by computer languages ​​can be used to verify, correct and structure the historical non-performing asset data, such as deep cleaning, induction and expansion processing. Through structured processing, the originally scattered and disordered unstructured data can be converted into structured data with clear structure, easy to query and analyze, and safely transferred to an efficient database, laying a solid foundation for subsequent data mining and model training.

[0101] Optionally, in the process of obtaining processed historical data, the historical non-performing asset data may be verified to determine whether the historical non-performing asset data meets the requirements of completeness, consistency and accuracy. When the data meets the requirements of completeness, consistency and accuracy, the historical non-performing asset data may be checked to determine abnormal data and correct the abnormal data. The abnormal data may include any of the following: erroneous data, invalid data and missing data. The unstructured data in the non-performing asset data may be converted into structured data using natural language processing to obtain the processed historical data.

[0102] Specifically, after obtaining the historical non-performing asset data, the computer device can determine the corresponding data collection plan according to different data sources, and conduct preliminary verification of the obtained historical non-performing asset data, compare it with the preset verification rules, check the integrity, consistency and accuracy of the data, and when the integrity, consistency and accuracy verification pass, perform abnormal inspection and processing to determine the abnormal data that is erroneous, invalid or missing, such as removing or correcting the noise, duplicates and erroneous items in the data, deleting invalid records, processing missing values, correcting incorrect data types or formats, etc. Processing missing values ​​can be filling missing values, and interpolation, mean replacement or mode replacement methods can be used to fill missing values.

[0103] For example, in the process of structured processing, the historical non-performing asset data obtained is mainly the collection record, and it is necessary to structure the due diligence elements. Assume that the collection record is: Zhang San, a collector of collection company A, initiated a collection action on customer Li Si by phone on a specific date (March 7, 2020), and the result of the collection was that the customer was willing but unable to repay. In this scenario, key information can be extracted first by computer language, which can be Python, SQL, etc., and sorted according to the preset structured template to obtain structured data. The structured data can be in the following clear format: the customer name is Li Si, the collector is Zhang San, the collection company is A, the collection date is clearly marked as March 7, 2020, the collection method is specifically telephone collection, and the collection result is concisely summarized as willing but unable to repay. In this step, by structuring the unstructured data, it can not only facilitate subsequent statistical analysis, but also significantly improve the efficiency and accuracy of data utilization in the due diligence process.

[0104] Step S303: convert the processed data into a preset format.

[0105] It should be noted that the above preset format is a format that can be recognized by computer language. After obtaining the processed data, the bag-of-words model, TF-IDF and other methods can be used to perform text vectorization processing, convert the text data in the processed data into numerical data, convert the date and time data in the processed data into a unified date format, and convert the processed content data into dummy variables.

[0106] Step S304, in a preset format, feature extraction is performed on the processed data to determine key features, and the key features are standardized and normalized to obtain processed key features.

[0107] The above key features are feature information that needs to be input into the initial model, which can be determined according to actual business needs, for example, they can include asset type, overdue days, collateral value, litigation progress, etc. After the key features are extracted, they can be standardized and normalized to obtain the processed key features.

[0108] In this embodiment, by extracting key features and performing standardization and normalization processing, the processed key features can be made more regular, which is convenient for improving the model training effect, helps to reduce the complexity of the model, and improves the interpretability and generalization ability of the model.

[0109] Step S305, divide the processed key features into a training set, a validation set and a test set; the training set is annotated with historical due diligence results.

[0110] After obtaining the processed key features, they can be divided into training set, validation set and test set. The training set is used for model training, and the model learns the mapping relationship between input and output; the validation set is used to adjust model parameters and prevent overfitting, and monitor the performance of the model; the test set is used to evaluate the final performance of the model and the authenticity and independence of the evaluation results.

[0111] In the process of data division, the processed key features can be divided into training set, validation set and test set according to a certain ratio.

[0112] Step S306, training the initial model through the training set and the validation set to obtain the due diligence model to be verified, and performing a performance test on the due diligence model to be verified through the test set, and taking the due diligence model to be verified with the best performance test as the standardized due diligence model.

[0113] Specifically, in the process of constructing a standardized due diligence model, the training set can be input into the initial model to obtain the output result, and a loss function can be constructed based on the output result and the historical due diligence results corresponding to the training set. The initial model can be iteratively trained according to the minimization of the loss function to obtain the due diligence model to be trained, and the verification set can be input into the due diligence model to be trained to obtain the due diligence model to be verified.

[0114] After obtaining the output results, a loss function can be constructed based on the output results and the historical due diligence results annotated on the training set, and the initial model can be optimized according to the minimization of the loss function to obtain the due diligence model to be trained. According to the difference between the output results and the historical due diligence results, the parameters in the initial model are updated to achieve the purpose of training the initial model and obtain the due diligence model to be trained. Among them, updating the parameters in the initial model can be updating the matrix parameters such as the weight matrix and the bias matrix in the initial model. Among them, the above-mentioned weight matrix and bias matrix include but are not limited to the matrix parameters in the convolution layer, feedforward network layer, and fully connected layer in the initial model.

[0115] After obtaining the due diligence model to be trained, the validation set can be input into the due diligence model to be trained, and the hyperparameters in the model can be continuously adjusted through the iterative training algorithm according to the historical due diligence results corresponding to the validation set, so that the model to be trained corresponding to the optimal hyperparameters is used as the due diligence model to be verified. The performance of the due diligence model to be verified is tested on the test set, and the due diligence model to be verified with the best performance test is used as the standardized due diligence model.

[0116] Among them, in the process of building a standardized model, it is possible to first analyze the historical non-performing asset data to build a rule-based wild-matching algorithm that can be recognized by computer language, and then convert the rule-based wild-matching algorithm into a general strategy that can be recognized by the computer device through preset rules, and persist the data of the standardized due diligence model in the database through the general strategy of the computer language. In the process of using the standardized model, the general model strategy in the database can be read through the query interface by the computer language and the asset data to be processed can be parsed to generate a due diligence guidance plan, and then generate a due diligence report. The computer device can store the due diligence report in the database again, and when receiving the due diligence query request sent by the user terminal, the due diligence report is sent to the user terminal through the query interface.

[0117] The due diligence process in this embodiment can flexibly respond to changes in duration and updates to asset status through review operations, and dynamically adjust the due diligence guidance plan according to changes in the due diligence content. At the same time, it can update the input information of the standardized model in real time according to the review content, quickly adapt to special situations or provide more detailed information, ensure the effectiveness of the model under different market conditions, thereby promoting the optimization of the decision-making process, thereby improving the efficiency of non-performing asset management, and also supporting more accurate risk assessment and decision-making.

[0118] Based on the same inventive concept, the embodiment of the present application also provides a non-performing asset due diligence device for implementing the non-performing asset due diligence method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more non-performing asset due diligence devices provided below can refer to the limitations of the non-performing asset due diligence method above, and will not be repeated here.

[0119] In an exemplary embodiment, Figure 5 As shown, a due diligence device for non-performing assets is provided, including:

[0120] The acquisition module 710 is used to acquire the debtor's asset data to be processed and perform structured processing on the asset data to be processed to obtain structured asset data;

[0121] A preliminary due diligence module 720 is used to perform preliminary due diligence on structured asset data;

[0122] The backcheck processing module 730 is used to perform backcheck processing on the asset data of the preliminary due diligence to obtain processed data;

[0123] Due diligence guidance module 740, used to input the processed data into a standardized due diligence model to obtain a due diligence guidance plan; the standardized due diligence model includes model data, which is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy, and the rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data;

[0124] The report generation module 750 is used to organize the due diligence content obtained through the due diligence guidance plan and generate a due diligence report on the asset data to be processed.

[0125] As an optional implementation, the checkback processing module 730 is specifically configured to:

[0126] Use due diligence backchecking to determine whether there are any data element updates in the asset data of the initial due diligence, which includes data element changes or new data elements;

[0127] When there is a change in data elements, obtain the changed data elements and add the changed data elements to the asset data of the initial due diligence to obtain the changed asset data;

[0128] When there is a requirement to add new data, obtain the new data elements and add them to the asset data of the initial due diligence to obtain the new asset data;

[0129] Based on the changed asset data and the newly added asset data, processed data is obtained.

[0130] As an optional implementation, the acquisition module 710 is further configured to:

[0131] Perform data cleaning and verification on the asset data to be processed to obtain the verified asset data;

[0132] Determine unstructured data from the verified asset data and convert the unstructured data into structured data using natural language processing;

[0133] The verified asset data and structured data are sorted according to a standard format to obtain structured asset data.

[0134] As an optional implementation, the due diligence guidance module 730 is specifically used to:

[0135] Obtain creditor's rights information, principal debtor information, guarantor information, mortgage information, and property clue information from the processed data;

[0136] The creditor's rights information, principal debtor information, guarantor information, mortgage and pledge information, and property clue information are analyzed through a standardized due diligence model to generate a due diligence guidance plan; the due diligence guidance plan includes any one of the following: due diligence methods and due diligence content.

[0137] As an optional implementation, the standardized due diligence model is constructed in the following way:

[0138] Obtain historical non-performing asset data; historical non-performing asset data includes multiple information categories;

[0139] Verify, correct and structure the historical non-performing asset data to obtain processed historical data;

[0140] Convert the processed data into a preset format;

[0141] In a preset format, feature extraction is performed on the processed data to determine key features, and the key features are standardized and normalized to obtain processed key features;

[0142] The processed key features are divided into training set, validation set and test set; the training set is annotated with historical due diligence results;

[0143] The initial model is trained through the training set and the validation set to obtain the due diligence model to be verified, and the performance of the due diligence model to be verified is tested through the test set. The due diligence model to be verified with the best performance test is used as the standardized due diligence model.

[0144] As an optional implementation, the above device is also used for:

[0145] Verify the historical non-performing asset data to determine whether the historical non-performing asset data meets the requirements of completeness, consistency and accuracy;

[0146] When completeness, consistency and accuracy are met, historical non-performing asset data shall be checked and processed, abnormal data shall be identified and corrected; abnormal data includes any of the following: wrong data, invalid data, missing data;

[0147] Natural language processing is used to convert unstructured data in non-performing asset data into structured data to obtain processed historical data.

[0148] As an optional implementation, the above device is also used for:

[0149] Input the training set into the initial model and get the output result;

[0150] Construct a loss function based on the output results and the historical due diligence results corresponding to the training set;

[0151] According to the minimization of the loss function, the initial model is iteratively trained to obtain the due diligence model to be trained;

[0152] Input the verification set into the due diligence model to be trained to obtain the due diligence model to be verified.

[0153] Among them, by implementing this implementation method, the technical solution can achieve structuring of due diligence elements through structured processing of the asset data to be processed, convert the originally scattered and disordered data into structured data, and conduct a back check on the asset data of the preliminary due diligence. It can conduct a comprehensive review of the asset data that has undergone preliminary due diligence, so as to quickly find the changed or newly added data, making the acquired processed data more comprehensive, and because a standardized due diligence model is constructed through the analysis of historical non-performing asset data, the processed data can be processed through the standardized due diligence model, which can consider various information of the debtor more comprehensively, thereby more accurately and comprehensively extracting the true situation of the debtor's assets, making the determined due diligence guidance plan more accurate, thereby improving work efficiency.

[0154] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video tag processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a due diligence method for non-performing assets is implemented.

[0155] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0156] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0157] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0158] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0159] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0160] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0161] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0162] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0163] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A due diligence method for non-performing assets, characterized in that: The due diligence methods for the non-performing assets include: Acquiring the debtor's asset data to be processed and performing structured processing on the asset data to be processed to obtain structured asset data; Conducting preliminary due diligence on the structured asset data; Performing a back-check on the asset data of the preliminary due diligence to obtain processed data; Input the processed data into a standardized due diligence model to obtain a due diligence guidance plan; the standardized due diligence model includes model data, and the model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy, and the rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data; The due diligence content obtained after the due diligence guidance plan is sorted out to generate a due diligence report on the asset data to be processed.

2. The due diligence method for non-performing assets according to claim 1, characterized in that: The asset data of the preliminary due diligence is reviewed and processed to obtain the processed data, including: Using due diligence review to determine whether there are any data element updates in the asset data of the preliminary due diligence, where the data element updates include data element changes or data element additions; When there is a change in data elements, the changed data elements are obtained and added to the asset data of the preliminary due diligence to obtain the changed asset data; When there is a requirement to add new data, obtain the new data elements and add the new data elements to the asset data of the preliminary due diligence to obtain the new asset data; Based on the changed asset data and the newly added asset data, the processed data is obtained.

3. The due diligence method for non-performing assets according to claim 1, characterized in that: The asset data to be processed is structured, including: Performing data cleaning and verification processing on the asset data to be processed to obtain verified asset data; Determining unstructured data from the verified asset data, and converting the unstructured data into structured data using a natural language processing method; The verified asset data and the structured data are sorted according to a standard format to obtain the structured asset data.

4. The due diligence method for non-performing assets according to claim 1, characterized in that: Input the processed data into the standardized due diligence model to obtain a due diligence guidance plan, including: Obtaining creditor's rights information, principal debtor information, guarantor information, mortgage information, and property clue information from the processed data; The credit information, principal debtor information, guarantor information, mortgage information, and property clue information are analyzed through the standardized due diligence model to analyze the credit situation, principal debtor, guarantor, mortgage, and property clues, and generate a due diligence guidance plan; the due diligence guidance plan includes any one of the following: due diligence methods and due diligence content.

5. The due diligence method for non-performing assets according to claim 1, characterized in that: The standardized due diligence model is constructed in the following way: Acquire historical non-performing asset data; the historical non-performing asset data includes multiple information categories; Verifying, correcting and structuring the historical non-performing asset data to obtain processed historical data; Converting the processed data into a preset format; Under the preset format, performing feature extraction processing on the processed data to determine key features, and performing standardization and normalization processing on the key features to obtain processed key features; Dividing the processed key features into a training set, a validation set, and a test set; the training set is annotated with historical due diligence results; The initial model is trained by using the training set and the validation set to obtain a due diligence model to be verified, and the performance of the due diligence model to be verified is tested by using the test set, and the due diligence model to be verified with the best performance test is used as the standardized due diligence model.

6. The due diligence method for non-performing assets according to claim 5, characterized in that: Verify, amend and structure the historical non-performing asset data to obtain processed historical data, including: Verify the historical non-performing asset data to determine whether the historical non-performing asset data is complete, consistent and accurate; When the integrity, consistency and accuracy are met, the historical non-performing asset data is checked and processed, abnormal data is determined and corrected; the abnormal data includes any of the following: wrong data, invalid data, missing data; The unstructured data in the non-performing asset data is converted into structured data using a natural language processing method to obtain the processed historical data.

7. The due diligence method for non-performing assets according to claim 5, characterized in that: The initial model is trained by using the training set and the verification set to obtain a due diligence model to be verified, including: Inputting the training set into the initial model to obtain an output result; Constructing a loss function according to the output result and the historical due diligence results corresponding to the training set; According to minimizing the loss function, iteratively training the initial model to obtain the due diligence model to be trained; The verification set is input into the due diligence model to be trained to obtain the due diligence model to be verified.

8. A due diligence device for non-performing assets, characterized in that: The due diligence device for the non-performing assets includes: An acquisition module, used for acquiring the debtor's asset data to be processed and performing structured processing on the asset data to be processed to obtain structured asset data; A preliminary due diligence module, used to conduct a preliminary due diligence on the structured asset data; A backcheck processing module is used to perform backcheck processing on the asset data of the preliminary due diligence to obtain processed data; A due diligence guidance module, used for inputting the processed data into a standardized due diligence model to obtain a due diligence guidance scheme; the standardized due diligence model includes model data, and the model data is stored in a database after converting a rule-based wildcard algorithm into a computer-recognizable general strategy, and the rule-based wildcard algorithm is constructed by analyzing historical non-performing asset data; The report generation module is used to organize the due diligence content obtained after the due diligence guidance plan is passed, and generate a due diligence report on the asset data to be processed.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the due diligence method for non-performing assets described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the due diligence method for non-performing assets described in any one of claims 1 to 7 is implemented.

Images