Method, platform, equipment and storage medium for identifying execution of case of turning to production breaking

Abstract

The invention provides a method, a platform, equipment and a storage medium for identifying a case for executing to-break, and the method comprises the steps: querying and obtaining retrieval data of a plurality of identification indexes corresponding to the information of a permitted pedestrian according to the information of the permitted pedestrian of a to-be-identified case, and obtaining feature data; calling an expert system to obtain a first execution-to-break prediction result; on the basis of the machine learning model, obtaining a second execution to break prediction result, and determining an execution to break feasibility mark; and in response to the case query request, displaying a case information display page, and adding a feasibility mark for executing turning and breaking in the case information display page. The method realizes information connection between the execution program and the production breaking program, automatically identifies feasibility of the execution of the case to the production breaking and notifies the corresponding user, can improve identification efficiency of the execution of the case to the production breaking, reduces manual identification time, and improves user experience through multi-source data fusion and multi-factor comprehensive analysis. And the identification accuracy of the execution of the production-to-break case is improved.

Description

Background Art

[0002] The implementation of the transfer to bankruptcy refers to a mandatory system in the execution process where the people's court, upon discovering that the person subject to execution meets the bankruptcy conditions and with the consent of the applicant for execution or the person subject to execution, transfers the execution case to the court with jurisdiction for bankruptcy review. Therefore, the execution-to-bankruptcy procedure is the connection between the execution procedure and the bankruptcy procedure. By replacing execution with bankruptcy, it increases the enterprise's solvency. When an enterprise is clearly lacking in solvency, the execution-to-bankruptcy can, to the greatest extent possible, ensure the fair repayment of creditors and is conducive to diverting execution cases to meet the actual need to clear up accumulated execution cases.

[0003] However, there are some technical obstacles in the existing execution-to-bankruptcy procedure. The information connection between the execution procedure and the bankruptcy procedure is not smooth. Currently, the integrated case-handling platforms for execution across the country often only integrate the information of the execution procedure and do not build an information connection mechanism with the bankruptcy management platform. Therefore, the two-way transfer of information such as property and bankruptcy cause review information cannot be achieved. In the process of information transfer of the execution-to-bankruptcy within the court, it is limited to the "pseudo-informatization" working method of filling in pages one by one and the "reverse informatization" working method that takes paper materials as the working entity. When judging whether the execution procedure can be converted into the bankruptcy procedure, it still relies on the manual judgment of judges, which is very time-consuming and labor-intensive. When the number of execution cases is huge, the efficiency of manual review by judges is limited, and the backlog of cases will still occur. In addition, there are many considerations in the conversion of the execution procedure into the bankruptcy procedure. When the judge initiates this procedure actively, the person subject to execution and the creditor may not cooperate, resulting in the failure to smoothly implement this procedure, leading to the reversal of the process and further affecting the case handling efficiency.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of this application. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] Aiming at the problems in the prior art, the purpose of this application is to provide a method, platform, device, and storage medium for identifying execution-to-bankruptcy cases, realizing the information connection between the execution procedure and the bankruptcy procedure, automatically identifying the feasibility of execution-to-bankruptcy cases and notifying the corresponding users, improving the identification efficiency of execution-to-bankruptcy cases, reducing the manual identification time, and improving the identification accuracy of execution-to-bankruptcy cases through the comprehensive analysis of multiple factors by multi-source data fusion.

[0006] An embodiment of this application provides a method for identifying execution-to-bankruptcy cases, including the following steps:

[0007] According to the information of the person subject to enforcement in the final case to be recognized, call the multi-source data query interface to query and obtain the retrieval data of multiple recognition indicators corresponding to the information of the person subject to enforcement, and obtain the characteristic data corresponding to the information of the person subject to enforcement based on the retrieval data;

[0008] Based on the characteristic data corresponding to the information of the person subject to enforcement, call the expert system to obtain the first prediction result of execution conversion to bankruptcy. The expert system is configured to predict the feasibility of execution conversion to bankruptcy of the person subject to enforcement based on the preset execution conversion to bankruptcy prediction rules and characteristic data;

[0009] Input the characteristic data corresponding to the information of the person subject to enforcement into the trained machine learning model to obtain the second prediction result of execution conversion to bankruptcy output by the machine learning model, and determine the execution conversion to bankruptcy feasibility annotation of the final case to be recognized according to the first prediction result of execution conversion to bankruptcy and / or the second prediction result of execution conversion to bankruptcy;

[0010] In response to a case query request, retrieve the case information corresponding to the case query request, display the case information display page, and add an execution conversion to bankruptcy feasibility annotation on the case information display page.

[0011] In some embodiments, calling the multi-source data query interface to query and obtain the retrieval data of multiple recognition indicators corresponding to the information of the person subject to enforcement includes the following steps:

[0012] Determine the query interface and retrieval keywords for each recognition indicator;

[0013] Based on the query keywords of each recognition indicator, call the query interface to retrieve the data of the recognition indicator on the information platform to obtain the retrieval data of multiple recognition indicators.

[0014] In some embodiments, obtaining the characteristic data corresponding to the information of the person subject to enforcement based on the retrieval data includes the following steps:

[0015] Determine the recognition indicator and characteristic quantity generation rule corresponding to each characteristic quantity;

[0016] Based on the retrieval data of the recognition indicator corresponding to each characteristic quantity and the characteristic quantity generation rule, generate each characteristic quantity, and combine the characteristic quantities to obtain the characteristic data corresponding to the information of the person subject to enforcement.

[0017] In some embodiments, before inputting the characteristic data corresponding to the information of the person subject to enforcement into the trained machine learning model, the following steps are further included:

[0018] Judge whether the first prediction result of execution conversion to bankruptcy corresponds to high feasibility or low feasibility;

[0019] When the first prediction result of execution conversion to bankruptcy corresponds to high feasibility, determine the execution conversion to bankruptcy feasibility annotation of the final case to be recognized according to the first prediction result of execution conversion to bankruptcy;

[0020] When the first execution-to-bankruptcy prediction result corresponds to low feasibility, the characteristic data corresponding to the information of the person subject to enforcement is input into the trained machine learning model.

[0021] In some embodiments, the following steps are adopted to determine the execution-to-bankruptcy feasibility annotation of the final execution case to be identified:

[0022] When the first execution-to-bankruptcy prediction result corresponds to high feasibility, it is determined that the execution-to-bankruptcy feasibility annotation of the final execution case to be identified is extremely likely;

[0023] When the first execution-to-bankruptcy prediction result corresponds to low feasibility, and it is determined to be high feasibility according to the first execution-to-bankruptcy prediction result and the second execution-to-bankruptcy prediction result, it is determined that the execution-to-bankruptcy feasibility annotation of the final execution case to be identified is highly likely;

[0024] When the first execution-to-bankruptcy prediction result corresponds to low feasibility, and it is determined to be low feasibility according to the first execution-to-bankruptcy prediction result and the second execution-to-bankruptcy prediction result, it is determined that the execution-to-bankruptcy feasibility annotation of the final execution case to be identified is low likely.

[0025] In some embodiments, before the information of the person subject to enforcement of the final execution case to be identified, the following steps are further included:

[0026] Screen the final execution case to be identified from the execution case database based on the program preselection rules.

[0027] In some embodiments, an expert system is called to obtain the first execution-to-bankruptcy prediction result, including the following steps:

[0028] Based on each preset execution-to-bankruptcy prediction rule, determine the characteristic quantity corresponding to each execution-to-bankruptcy prediction rule, and obtain the determination result corresponding to each execution-to-bankruptcy prediction rule;

[0029] According to the determination results and prediction result generation rules corresponding to multiple execution-to-bankruptcy prediction rules, obtain the first execution-to-bankruptcy prediction result.

[0030] The embodiment of the present application further provides an execution-to-bankruptcy case fusion platform for implementing the above-mentioned execution-to-bankruptcy case identification method. The platform includes:

[0031] An information collection module, which is used to call a multi-source data query interface according to the information of the person subject to enforcement of the final execution case to be identified, query and obtain the retrieval data of multiple identification indicators corresponding to the information of the person subject to enforcement, and obtain the characteristic data corresponding to the information of the person subject to enforcement based on the retrieval data;

[0032] An expert system module, configured to call an expert system based on the feature data corresponding to the information of the person subject to enforcement, so as to obtain a first prediction result of the conversion from enforcement to bankruptcy. The expert system is configured to predict the feasibility of the conversion from enforcement to bankruptcy of the person subject to enforcement based on the preset prediction rules for the conversion from enforcement to bankruptcy and the feature data;

[0033] A result prediction module, configured to input the feature data corresponding to the information of the person subject to enforcement into a trained machine learning model, obtain a second prediction result of the conversion from enforcement to bankruptcy output by the machine learning model, and determine the annotation of the feasibility of the conversion from enforcement to bankruptcy for the case to be identified for termination of enforcement based on the first prediction result of the conversion from enforcement to bankruptcy and / or the second prediction result of the conversion from enforcement to bankruptcy;

[0034] A user interface module, configured to, in response to a case query request, retrieve the case information corresponding to the case query request, display a case information display page, and add an annotation of the feasibility of the conversion from enforcement to bankruptcy on the case information display page.

[0035] An embodiment of the present application further provides a device for identifying a case of conversion from enforcement to bankruptcy, including:

[0036] A processor;

[0037] A memory, in which executable instructions of the processor are stored;

[0038] Wherein, the processor is configured to execute the steps of the method for identifying a case of conversion from enforcement to bankruptcy by executing the executable instructions.

[0039] An embodiment of the present application further provides a computer-readable storage medium, configured to store a program, and when the program is executed by a processor, the steps of the method for identifying a case of conversion from enforcement to bankruptcy are implemented.

[0040] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure.

[0041] The method, platform, device and storage medium for identifying a case of conversion from enforcement to bankruptcy of the present application have the following

[0042] Beneficial effects:

[0043] By adopting the present application, the information connection between the enforcement procedure and the bankruptcy procedure is effectively realized, and the feasibility of the case of conversion from enforcement to bankruptcy is automatically identified and notified to the corresponding user in the form of an annotation of the feasibility of the conversion from enforcement to bankruptcy, which is convenient for the user to quickly make a decision on whether to transfer from the enforcement procedure to the bankruptcy procedure, can improve the identification efficiency of the case of conversion from enforcement to bankruptcy, reduce the manual identification time, and through the multi-source data fusion and comprehensive analysis of multiple factors, and automatic prediction based on the expert system and the machine learning model, the identification accuracy of the case of conversion from enforcement to bankruptcy is improved. Description of the Drawings

[0044] Other features, objects, and advantages of the present application will become more apparent by reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0045] Figure 1 is a flowchart of a method for identifying execution-to-bankruptcy cases according to an embodiment of the present application;

[0046] Figure 2 is a schematic diagram of the process of a method for identifying execution-to-bankruptcy cases according to an embodiment of the present application;

[0047] Figure 3 is a schematic diagram of a case information display page according to an embodiment of the present application;

[0048] Figure 4 is a schematic diagram of the structure of an execution-to-bankruptcy case integration platform according to an embodiment of the present application;

[0049] Figure 5 is a schematic diagram of the structure of an execution-to-bankruptcy case identification device according to an embodiment of the present application;

[0050] Figure 6 is a schematic diagram of the structure of a computer-readable storage medium according to an embodiment of the present application. Detailed Embodiments

[0051] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art. The features, structures, or characteristics described can be combined in any suitable manner in one or more embodiments.

[0052] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. Although the terms "first" or "second" etc. are used in this specification to denote certain features, they are only for the purpose of indication and not for limiting the quantity and importance of the specific features.

[0053] The flowcharts shown in the accompanying drawings are only illustrative and not necessarily include all the steps. For example, some steps can be decomposed, and some steps can be combined or partially combined. Therefore, the actual execution order may be changed according to the actual situation.

[0054] As Figure 1 shown, an embodiment of the present application provides a method for identifying cases transferred to bankruptcy execution, including the following steps:

[0055] S100: According to the information of the person subject to enforcement of the case to be identified as an end-of-execution case, call a multi-source data query interface to query and obtain retrieval data of multiple identification indicators corresponding to the information of the person subject to enforcement, and obtain feature data corresponding to the information of the person subject to enforcement based on the retrieval data;

[0056] The retrieval data can be structured data obtained by querying through specific query interfaces and specific keywords from various case database platforms and enterprise information platforms, and is used to extract the feature data of the person subject to enforcement for subsequent identification of the feasibility of transferring execution to bankruptcy;

[0057] S200: Based on the feature data corresponding to the information of the person subject to enforcement, call an expert system to obtain a first prediction result of transferring execution to bankruptcy. The expert system is configured to predict the feasibility of transferring execution to bankruptcy of the person subject to enforcement based on preset prediction rules for transferring execution to bankruptcy and feature data;

[0058] S300: Input the feature data corresponding to the information of the person subject to enforcement into a trained machine learning model, obtain a second prediction result of transferring execution to bankruptcy output by the machine learning model, and determine the annotation of the feasibility of transferring execution to bankruptcy of the case to be identified as an end-of-execution case according to the first prediction result of transferring execution to bankruptcy and / or the second prediction result of transferring execution to bankruptcy;

[0059] S400: In response to a case query request, retrieve the case information corresponding to the case query request, display a case information display page, and add an annotation of the feasibility of transferring execution to bankruptcy on the case information display page.

[0060] Among them, in step S400, if the case being queried is a case for which the annotation of the feasibility of transferring execution to bankruptcy has been determined before, then add an annotation of the feasibility of transferring execution to bankruptcy on the case information display page. If the case being queried does not have an annotation of the feasibility of transferring execution to bankruptcy, then "not applicable to transfer execution to bankruptcy", "not predicted" or other annotations can be displayed.

[0061] Therefore, the present invention is based on technology replacing manual judgment to quickly identify a person subject to enforcement who is likely to be insolvent and has no special circumstances where transfer is not possible. By adopting the present application, the information connection between the execution procedure and the bankruptcy procedure is effectively realized, and the feasibility of transferring execution to bankruptcy of the case is automatically identified and notified to the corresponding user in the form of an annotation of the feasibility of transferring execution to bankruptcy, which is convenient for the user to quickly decide whether to transfer from the execution procedure to the bankruptcy procedure, can improve the identification efficiency of cases transferred to bankruptcy execution, reduce the manual identification time, and through multi-source data fusion and comprehensive analysis of multiple factors, and automatic prediction based on an expert system and a machine learning model, the identification accuracy of cases transferred to bankruptcy execution is improved.

[0062] In practice, there are conflicts of interest among the parties involved in the execution-to-bankruptcy process, and no one has applied for it. The applicants for execution can be divided into first comers and later comers. The first comers with more information can obtain more benefits in the execution procedure, so they are unwilling to apply for execution-to-bankruptcy. The later comers with less information know little about the situation of the person being executed, and will not apply for execution-to-bankruptcy from the beginning. In addition, the bankruptcy of the person being executed will have a negative impact on his personal reputation. Instead of applying for execution-to-bankruptcy, he is more inclined to deal with creditors one by one to preserve his personal reputation. After determining the feasibility of execution-to-bankruptcy for the final case to be identified, the subsequent execution-to-bankruptcy initiation procedure can also be connected, such as sending the information that execution-to-bankruptcy can be applied to the person being executed or the applicant for execution. For example, taking small and micro creditors as a breakthrough, by constructing a creditor-oriented execution-to-bankruptcy synchronous early warning system, after the creditor applies for execution, the feasibility of execution-to-bankruptcy is determined through the execution-to-bankruptcy case identification method. When the feasibility of execution-to-bankruptcy is high, the creditor is reminded in time that the person being executed may be involved in the execution-to-bankruptcy procedure, realizing the automatic flow of identification and early warning, and breaking the deadlock of the existing interest constraints of all parties.

[0063] In this embodiment, according to the information of the person subject to execution of the final case to be identified, before calling the multi-source data query interface to query and obtain the search data of multiple identification indicators corresponding to the information of the person subject to execution, the following steps are also included:

[0064] Based on procedural pre-selection rules, select cases to be identified and finalized from the execution case database.

[0065] The procedure pre-selection rules include, for example, whether the execution case is a final case. First, the final cases in the execution case data are screened out from the execution case database as the final cases to be identified. The financial situation of the executed persons in the final cases is generally not optimistic, and there is a high possibility that the execution will be converted to bankruptcy procedures. In addition, the identification of the execution-to-bankruptcy procedure needs to be set in order, so as to achieve efficient execution of the identification process from shallow to deep, from easy to difficult in the identification process, and to facilitate the priority identification of final cases with a greater possibility. The procedure pre-selection rules also include classifying the final cases to be identified according to the executed persons, sorting them according to the number of final cases of the executed persons, and sorting the executed persons from high to low according to the number of final cases. Subsequently, the feasibility of execution-to-bankruptcy is predicted according to the sorting order of the executed persons. In steps S100 to S300, the feasibility of each executed person to convert to bankruptcy procedures is analyzed based on the characteristic data of each executed person, and then the feasibility mark of execution-to-bankruptcy is determined for each final case corresponding to the executor.

[0066] When predicting the feasibility of converting enforcement to bankruptcy, multiple factors need to be considered. For example, whether the person subject to enforcement has cases with final enforcement results, whether the person subject to enforcement has major negative public opinions, whether the local employment situation will be affected if the person subject to enforcement applies for bankruptcy, whether there are historical legacy issues, whether the person subject to enforcement is involved in the supply of people's livelihood services, whether the person subject to enforcement has insufficient profitability, etc. If all rely on judges to make manual judgments and analyses, judges need to spend a lot of time collecting a large amount of information for comprehensive analysis and judgment, with very low efficiency, resulting in a large backlog of cases. Moreover, due to the information barriers between various platforms and the limitations of manual retrieval and analysis capabilities, there is a high possibility of large errors in manual judgments. Once a misjudgment occurs, it will have a significant adverse impact on the applicant for enforcement, the person subject to enforcement, and even local people's livelihood. Based on this, in this application, multiple characteristic quantities of the person subject to enforcement are determined according to multiple factors to be considered, and for each characteristic quantity, the generation rule of the characteristic quantity and the information to be collected when generating the characteristic quantity are determined, and the identification indicators to be collected are determined according to the information to be collected. The retrieval data of each identification indicator is the retrieval data that can be directly retrieved from some platforms through the query interface.

[0067] In this embodiment, in step S100, a multi-source data query interface is called to query and obtain the retrieval data of multiple identification indicators corresponding to the information of the person subject to enforcement, including the following steps:

[0068] Determine the query interface and retrieval keywords of each identification indicator;

[0069] Based on the query keywords of each identification indicator, call the query interface to retrieve the data of the identification indicator on the information platform to obtain the retrieval data of multiple identification indicators. This information platform can be an enforcement case database platform or an enterprise information query platform (such as Tianyancha platform, etc.).

[0070] In step S100, based on the retrieval data, obtain the characteristic data corresponding to the information of the person subject to enforcement, including the following steps:

[0071] Determine the identification indicators and characteristic quantity generation rules corresponding to each characteristic quantity;

[0072] Based on the retrieval data of the identification indicators corresponding to each characteristic quantity and the characteristic quantity generation rules, generate each characteristic quantity, and combine the characteristic quantities to obtain the characteristic data corresponding to the information of the person subject to enforcement.

[0073] For example, through business research, without considering the applicant for enforcement, the identification conditions for converting enforcement of a case to bankruptcy can be summarized into the following business rules:

[0074] 1. There is at least one execution case that has been terminated: The financial situation of the executed parties in the terminated cases is generally not optimistic, but the execution-to-bankruptcy process needs to be set in order, so that the debt can be efficiently paid off from the shallow to the deep, from the easy to the difficult. Therefore, in the face of a large number of executed parties whose executions have been terminated, the focus of this identification model is the number of executed parties' terminated cases. First, the executed parties are arranged according to the number of terminated cases, and the executed parties with a number of terminated cases higher than 1 are screened out. Then, according to the number of terminated cases, they are included in the case database to be transferred from execution to bankruptcy in descending order.

[0075] 2. No major negative public opinion: For cases where the execution is finalized, the determination of the cause of bankruptcy is not a difficult problem. The fact of the finalization ruling itself means that the debtor is insolvent, and therefore can meet the conditions of Article 1, Paragraph 1 of the "Provisions of the Supreme People's Court on Several Issues Concerning the Application of the Enterprise Bankruptcy Law of the People's Republic of China (I)" and Article 1, Paragraph 2 of the "Guiding Opinions on Several Issues Concerning the Transfer of Execution Cases to Bankruptcy Review". However, the legality of the norm does not mean the desirability of the consequences. For the application for execution to bankruptcy that can be transferred or rejected, the judge's consideration is not limited to the rigid conditions of the judicial interpretation and the guiding opinions, but includes the consequences of bankruptcy. If the debtor has major negative public opinion, so that the debt repayment benefits brought by bankruptcy are not enough to make up for the cost of dealing with public opinion and derivative consequences, the judge should use initiative to assist the lag of the norm. Therefore, if the debtor has major negative public opinion, the automatic identification of execution to bankruptcy should be suspended and handed over to the judge for careful consideration.

[0076] 3. Will not affect the local employment situation: In addition to major negative public opinion, another possible consequence of the bankruptcy of the person subject to execution on local people's livelihood is the disturbance of the employment market. If the person subject to execution is too large and has too many employees, so that it can have a significant impact on the local employment rate, the automatic identification of execution-to-bankruptcy conversion should also be suspended and handed over to the judge for consideration.

[0077] 4. No historical problems: Some of the parties subject to enforcement have a mixed ownership reform process, so there are historical problems with the treatment of their employees. If a mixed-ownership enterprise goes bankrupt, even if the debtor is fairly repaid, its employees will face a dilemma in treatment, which needs to be coordinated by the local government. Therefore, enterprises with historical problems cannot resort to automatic identification of execution-to-bankruptcy, but should be considered by the judge.

[0078] 5. Will not affect local finances: This rule is similar to the consideration of local employment situation and also involves the impact on local people's livelihood. If the tax revenue provided by the person subject to execution accounts for too large a proportion of the local government, the execution-to-bankruptcy process should not be rashly entered, but should be handed over to the judge for careful consideration.

[0079] 6. Not involving the provision of livelihood services: Also involving the livelihood of the local people are enterprises that undertake livelihood service functions. Although some of the persons subject to execution have suffered losses for years, are insolvent, and have been executed, they still provide important livelihood services, such as tap water supply. Persons subject to execution involving the provision of livelihood services cannot be automatically identified, but should be submitted to the judge for consideration.

[0080] 7. Insufficient profitability: Profitability is related to the repayment expectations of the person subject to execution. If the person subject to execution still has profitability, the execution-to-bankruptcy procedure will not only fail to achieve efficient repayment, but will also reduce repayment expectations. Therefore, only when the person subject to execution is truly incapable of profitability can it be transferred to bankruptcy procedure.

[0081] In this embodiment, in step S200, calling the expert system to obtain the first execution-to-bankruptcy prediction result includes the following steps:

[0082] Based on the preset execution-to-bankruptcy prediction rules, the characteristic quantity corresponding to each execution-to-bankruptcy prediction rule is determined, and the determination result corresponding to each execution-to-bankruptcy prediction rule is obtained;

[0083] According to the judgment results and prediction result generation rules corresponding to the multiple execution-to-bankruptcy prediction rules, the first execution-to-bankruptcy prediction result is obtained. The prediction result generation rule is to generate the final first execution-to-bankruptcy prediction result by integrating the judgment results corresponding to the multiple execution-to-bankruptcy prediction rules. The expert system can be executed by a rule engine. When it is necessary to change, add or delete the execution-to-bankruptcy prediction rules, the rules set in the rule engine can be updated, and the rules for execution-to-bankruptcy identification can be flexibly adjusted.

[0084] Figure 2 The implementation process of the execution-to-bankruptcy case identification method is exemplarily shown in the figure. The symbol "∧" represents the "AND" operation and "∨" represents the "OR" operation. This application determines the feature quantity, the feature quantity generation rule, the identification index required to be queried when the feature quantity is generated and the corresponding query interface, the execution-to-bankruptcy prediction rule and the final prediction result generation rule according to the above business rules. Figure 2 The relationship between the identification index, the feature amount, and the identification rule in this embodiment will be specifically described.

[0085] For example, for the first business rule, the corresponding identification metric is the list of cases where enforcement against the person subject to enforcement has been terminated. By extracting information about the person subject to enforcement (such as the organization code), the list of enforcement cases of the person subject to enforcement is queried based on this information. The characteristic quantity corresponding to this query metric is the number of cases where enforcement has been terminated, and the rule for generating the corresponding characteristic quantity is to count the number of cases where enforcement has been terminated. The corresponding query interface is the case query interface of the enforcement case database. The query keyword for this identification metric is cases where enforcement has been terminated. Whether the enforcement-to-bankruptcy prediction rule corresponding to this characteristic quantity is satisfied is whether the number of cases where enforcement has been terminated is greater than 0.

[0086] For the second business rule, the corresponding identification metric is the list of news and public opinions of the person subject to enforcement. The corresponding characteristic quantities include sensitive public opinion situations (such as labor disputes, property delivery, etc.), historical legacy issues, risks in the supply of people's livelihood services, the number of negative public opinions, the proportion of negative public opinions, etc. Its query interface is the news and public opinion query interface of a specific enterprise information platform (such as the news and public opinion API provided by Tianyancha). The rule for generating the characteristic quantity of the number of negative public opinions is to count the number of negative public opinion news. The rule for generating the characteristic quantity of the proportion of negative public opinions is to count the ratio of the number of negative public opinion news to the number of all public opinion news. The query keywords corresponding to the sensitive public opinion situation are labor disputes, property delivery, etc. The rules for generating the characteristic quantities corresponding to the sensitive public opinion situation, historical legacy issues, and risks in the supply of people's livelihood services are: if such information exists, the characteristic quantity is set to 1; if not, the characteristic value of the characteristic quantity is set to 0. The enforcement-to-bankruptcy prediction rule corresponding to these characteristic quantities is whether the following conditions are satisfied: the characteristic values of the characteristic quantities corresponding to the sensitive public opinion situation, historical legacy issues, and risks in the supply of people's livelihood services are all 0, and the values of the number of negative public opinions and the proportion of negative public opinions both meet the corresponding threshold requirements.

[0087] Corresponding to the third business rule, which involves the employment resources provided by the person subject to enforcement. If the person subject to enforcement is closely related to local employment, the bankruptcy application cannot be rashly approved. The corresponding identification metric for this business rule is the enterprise scale of the person subject to enforcement and / or the latest annual report of the person subject to enforcement. The corresponding characteristic quantities are the number of employees of the person subject to enforcement and / or the enterprise scale level. The query interfaces are the enterprise scale query interface and / or the annual report query interface of the enterprise information platform, and the query keywords are scale and / or annual report. The rule for generating the characteristic quantity is to extract the number of employees / or the enterprise scale level from the enterprise scale and / or the latest annual report. The enforcement-to-bankruptcy prediction rule is whether the following conditions are satisfied: the number of employees is less than the preset number threshold, and / or, whether the enterprise scale level is medium scale or below medium scale.

[0088] Corresponding to the business scale of the fourth item, it involves historical problems of the person being executed. Some of these problems originate from the reform of state-owned enterprises, while others originate from the transfer and supply of public services. They are heterogeneous in form, have multiple causes, are complex and cumbersome, and are difficult to implement as a unified form of judgment. The identification indicator is historical problems, and the characteristic quantity is the number of historical problems. The query interface is the news and public opinion query interface of the enterprise information platform, and the query keywords are state-owned enterprises, mixed reform, etc. The characteristic quantity generation rule is that if there is at least one historical problem, the characteristic value is set to 1, otherwise it is set to 0. The execution-to-bankruptcy prediction rule is whether it is satisfied: the number of historical problems is 0.

[0089] Corresponding to the fifth business rule, it is necessary to determine the degree of impact of the executed enterprise on the local finances. This rule requires calculating the amount of revenue of the executed person, which can be directly extracted from its annual reports for the past three years. The identification indicator corresponding to this business rule is the latest annual report of the executed person, and its corresponding feature quantity is the proportion of the executed person's revenue tax to the local finances. The feature quantity generation rule is to calculate the ratio of the executed person's revenue tax to the local finances. The query interface is the annual report query interface of the enterprise information platform, and the query keyword is annual report. The execution-to-bankruptcy prediction rule is whether it meets the following conditions: the proportion of the executed person's revenue tax to the local finances is less than the preset ratio threshold.

[0090] Corresponding to the sixth business rule, it involves the people's livelihood service supply capacity of the executed enterprise. If the business of the executed enterprise directly involves people's livelihood services, such as water supply and heating, even if it suffers a large loss, it cannot enter bankruptcy proceedings. The corresponding identification indicator is the people's livelihood service supply information, and the corresponding characteristic quantity is the people's livelihood service supply risk. The characteristic quantity generation rule is to set the characteristic value to 1 when there is people's livelihood service supply information, otherwise it is set to 0. The query interface is the news and public opinion query interface of the enterprise information platform, and the query keywords are real estate, property, electricity, heating, tap water, etc. The execution-to-bankruptcy prediction rule is whether it is satisfied: the people's livelihood service supply risk is 0.

[0091] The corresponding business rule No. 7 involves the profitability of the executed enterprise. The judgment of profitability cannot be judged by clear rules and requires comprehensive consideration of various factors. Combined with the business practice of execution-to-bankruptcy, the available factors of this rule in the execution stage are: the number of patents of the executed person, the business industry, other litigation situations, and social responsibility assessment. The four available factors are set as four feature quantities. As far as patents are concerned, whether they are applied to the production practice of the enterprise or not, patents ultimately reflect the scientific research capabilities of an enterprise. Even as a defensive intellectual property weapon, patents reflect the industry status of an enterprise. The identification indicator corresponding to the number of patents is the patent list, the query interface is the patent information interface of the enterprise information platform, and the query keyword is patent. The feature quantity generation rule is to count the number of patents. As far as litigation is concerned, the more lawsuits the executed person is the defendant, the more operating problems the enterprise faces, and the more unfavorable its profit prospects. The identification indicator corresponding to the number of lawsuits is the litigation list of the executed person, the query interface is the legal litigation interface of the enterprise information platform, and the feature quantity generation rule is to count the number of lawsuits of the executed person. In terms of industry change trends, if the number of enterprises in the industry where the person subject to execution is located is approaching saturation and has shown a downward trend in recent years, the probability that the person subject to execution will continue to survive profitably in the industry will be minimal. The identification indicator corresponding to the industry change trend is the change information of the number of enterprises in the industry where the person subject to execution is located. The query interface is the advanced query front-end interface of the enterprise information platform. The characteristic quantity generation rule is that when the number of enterprises decreases or the change in the number of enterprises is less than the preset threshold, the characteristic value is set to 1, otherwise it is set to 0. In addition to the above three factors that are clearly related to profitability, the social responsibility index of the person subject to execution reflects the expectation of receiving future investment and also reflects the economic strength of the enterprise. Social responsibility disclosure requires enterprises to show care for stakeholders, and therefore involves the compliance of corporate governance and the sustainability of business operations. These directly involve whether the enterprise can meet the industry's value consensus and whether it can be accepted by the supply chain, and are therefore also considerations for corporate profitability. The identification indicator corresponding to the social responsibility assessment is the ESG (environmental, social, and corporate governance) report disclosure information of the person subject to execution in the past three years, and its query interface is the ESG report query interface of the enterprise information platform. The feature quantity generation rule is to evaluate the level of social responsibility of the person subject to execution based on the information disclosed in the ESG report. The execution-to-bankruptcy prediction rule of the multiple feature quantities is whether at least two of the following conditions are met: the number of patents of the person subject to execution is less than the patent threshold, the number of other lawsuits of the person subject to execution is less than the lawsuit threshold, the characteristic value of the trend of the number of enterprises in the industry in which the person subject to execution is 0, and the social responsibility assessment level of the person subject to execution is low.

[0092] Based on the above analysis, the relationship between the retrieval data, feature quantity, and feature quantity generation rules can be obtained as shown in Table 1 below.

[0093] Table 1 Retrieval data items and feature quantities of identification indicators

[0094]

[0095]

[0096]

[0097] Thus, the retrieved data of the recognition indicators that can be obtained include [List of Finalized Cases of the Executed Person], [List of News Public Opinions of the Executed Person], [Latest Annual Report of the Executed Person], [Revenue of the Executed Person in the Past Three Years], [List of Patents of the Executed Person], [Litigation List of the Executed Person], [Change in the New Quantity in the Industry Where the Executed Person is Located], [ESG Disclosures of the Executed Person in the Past Three Years]. From these eight data sources, 13 items such as [Number of Finalized Cases] and [Number of Negative Public Opinions] can be extracted. Based on the characteristic quantities of the executed person, a characteristic vector of the executed person can be constructed. Based on the characteristic vector, the first prediction result of execution conversion to bankruptcy can be obtained through an expert system. The generation rule of the final prediction result is, for example Figure 2 As shown, according to whether the prediction results of each execution conversion to bankruptcy prediction rule meet the execution conversion to bankruptcy conditions, if the prediction results of each execution conversion to bankruptcy prediction result all meet the execution conversion to bankruptcy conditions, it is determined that the first execution conversion to bankruptcy prediction result corresponds to a high possibility. If there is at least one prediction result of the execution conversion to bankruptcy prediction result that does not meet the execution conversion to bankruptcy conditions, it is determined that the first execution conversion to bankruptcy prediction result corresponds to a low possibility.

[0098] The identification strategy of execution-to-bankruptcy adopted in this application is not to comprehensively evaluate the bankruptcy possibility of each execution case, but to select the execution cases that can be transferred to the bankruptcy procedure and submit the execution-to-bankruptcy procedure, so as to reduce the workload of the execution bureau. Therefore, the design basis of the execution-to-bankruptcy identification model is the exclusion of negative conditions. For the executed persons with final cases, the negative conditions that need to be absolutely excluded are: sensitive public opinion conditions such as labor disputes and guaranteed delivery buildings, historical problems left over from the mixed reform of state-owned enterprises, and risks in the supply of livelihood services. These three situations make the executed persons not only have the profit function of the market economy, but also involve the political risks of local stability maintenance. Once they appear, they must resort to manual judgment. At the same time, there are serious risks for the supply of local employment and finance by enterprises. If the number of employees of the executed person exceeds 200, or the scale of the executed person's enterprise is assessed as medium-sized or above, or the tax amount of the executed person is higher than 0.1% of the city's fiscal revenue, the feasibility of its execution-to-bankruptcy should be temporarily excluded, and the judge shall comprehensively consider the impact of its bankruptcy and then transfer it to bankruptcy. In comparison, negative public opinion is a slightly milder negative condition. If the proportion of negative public opinion exceeds 1%, or the number of negative public opinions is greater than 10, the risk of the enterprise will most likely not be limited to economic difficulties, but will generate externalities. Therefore, enterprises with too many or too high proportions of negative public opinions cannot directly enter bankruptcy proceedings, but require the judge's careful consideration and comprehensive consideration of their risks. In addition, by screening the number of final cases in advance, it is also possible to avoid identifying the debtors who do not have final cases.

[0099] In addition to the above negative conditions, there are four positive conditions involving the profitability of the person subject to execution. The number of patents of the person subject to execution, the number of lawsuits filed by the defendant, the trend of industry changes, and the number of ESG disclosure reports are business-related to its profitability. Therefore, if the person subject to execution can meet two of the four conditions, it can be temporarily determined that it has the possibility of profitability, and the automated execution-to-bankruptcy identification is suspended. The core requirement of the execution-to-bankruptcy case identification method of this application is to provide a higher degree of accuracy, accurately identify cases that can be transferred to bankruptcy, and avoid the erroneous transfer of cases that should not be transferred. In this application, in order to improve the accuracy of the execution-to-bankruptcy case identification method, after obtaining the first execution-to-bankruptcy prediction result, a machine learning model (such as a perceptron) is used to further predict, and the machine learning model mainly focuses on the prediction of the expert system's rejection of the case. The machine learning model performs reflective equilibrium on various factors (feature quantities) in the feature vector of the person subject to execution, obtains the weights of various feature quantities, determines a set of judgment rules close to judicial practice, and realizes the accurate identification of execution-to-bankruptcy. The above 13-dimensional feature vectors are obtained: [name of the person subject to execution, number of final cases, number of patents of the person subject to execution, number of negative public opinions, proportion of negative public opinions, historical problems, risks of people's livelihood service supply, sensitive public opinions, total number of lawsuits in which the person subject to execution is the defendant, number of ESG disclosure reports in the past three years, number of employees, enterprise size, tax value of the person subject to execution in the past three years]. Finally, the prediction result of the second execution-to-bankruptcy is obtained. Among them, the first dimension is the feature value, and the following 12 dimensions are the input of the machine learning model.

[0100] Although the feature quantities of the 12 dimensions are all very important, their publicity varies. In different implementations, only a part of the feature vectors of the 13 dimensions can be selected as the input of the machine learning model. For example, the feature quantities with better publicity generally include: the number of final cases, the number of patents of the person subject to execution, the number of negative public opinions, the proportion of negative public opinions, the situation of historical problems, the risk of supply of people's livelihood services, sensitive public opinions, the total number of lawsuits in which the person subject to execution is the defendant, and the scale of the enterprise. Taking the machine learning model as an example of a perceptron, with these nine factors as input values, the perceptron fitting of the linear function of execution-to-bankruptcy identification is carried out, and the recognition accuracy can be increased on the basis of the expert system of execution-to-bankruptcy identification.

[0101] In this embodiment, the step S300: before inputting the characteristic data corresponding to the information of the person to be executed into the trained machine learning model, further includes the following steps:

[0102] Determine whether the first execution-to-bankruptcy prediction result corresponds to high feasibility or low feasibility;

[0103] When the first execution-to-bankruptcy prediction result corresponds to a high feasibility, determine the execution-to-bankruptcy feasibility annotation of the case to be identified for termination of execution according to the first execution-to-bankruptcy prediction result;

[0104] When the first execution-to-bankruptcy prediction result corresponds to a low feasibility, input the characteristic data corresponding to the information of the person subject to execution into the trained machine learning model.

[0105] Taking the perceptron as an example of the machine learning model, the training process of the perceptron includes: screening multiple real case examples from the execution-to-bankruptcy database as the sample library of the perceptron, including the proportion of cases finally approved for execution-to-bankruptcy and cases rejected for execution-to-bankruptcy, and adding labels to the feature vectors of each case example based on whether each real case example is approved or rejected. Set the initial weight matrix of the perceptron. Multiply the values of each feature quantity in the feature vector of each real case example in the sample library by the initial weight matrix to obtain the prediction result of the perceptron, and adjust the weight matrix according to the label to make the prediction result approach the label. Train the perceptron based on real case examples so that the trained perceptron better meets the usage requirements in the actual case handling scenario.

[0106] In this embodiment, a score can be used to identify the high or low execution-to-bankruptcy feasibility. For example, obtain the score according to the first execution-to-bankruptcy prediction result and / or the second execution-to-bankruptcy prediction result. When the set score is greater than or equal to the first score threshold, the execution-to-bankruptcy feasibility is extremely high. When the score is less than the first score threshold but greater than or equal to the second score threshold (the second score threshold is less than the first score threshold), the execution-to-bankruptcy feasibility is high. When the score is less than the second score threshold, the execution-to-bankruptcy feasibility is low. Taking the full score of the score as 100 as an example, when the first execution-to-bankruptcy prediction result of a person subject to execution corresponds to a high feasibility, the execution-to-bankruptcy score of this person subject to execution is 100. When the first execution-to-bankruptcy prediction result of a person subject to execution corresponds to a low feasibility, calculate the execution-to-bankruptcy score of the person subject to execution according to the first execution-to-bankruptcy prediction result and the second execution-to-bankruptcy prediction result. For example, use the formula K + √((weight matrix * perceptron consideration factor matrix) / 2) to calculate the execution-to-bankruptcy score. K is a preset reference value, such as set to 60, 65, 70, etc. Among them, the "weight matrix * perceptron consideration factor matrix" calculates the comprehensive value of the input features after weight weighting. Dividing by 2 is a scaling operation, taking the square root is a non-linear transformation of the data, and adding the preset reference value introduces a fixed offset to this result. Set that if the execution-to-bankruptcy score is equal to 100, the final feasibility determination result is an extremely high possibility. If the execution-to-bankruptcy score is less than 100 but greater than or equal to 80, the final feasibility determination result is a high feasibility. If the execution-to-bankruptcy score is less than 80, the final feasibility determination result is a low feasibility.

[0107] For example, the following steps are used to determine the feasibility label of converting the execution of a case closed for finalization into bankruptcy:

[0108] When the first prediction result of converting execution into bankruptcy corresponds to high feasibility, the feasibility label of converting the execution of the case closed for finalization to be identified into bankruptcy is determined as extremely high possibility;

[0109] When the first prediction result of converting execution into bankruptcy corresponds to low feasibility, and the final feasibility determination result determined according to the first prediction result of converting execution into bankruptcy and the second prediction result of converting execution into bankruptcy is high feasibility, the feasibility label of converting the execution of the case closed for finalization to be identified into bankruptcy is determined as high possibility;

[0110] When the first prediction result of converting execution into bankruptcy corresponds to low feasibility, and the final feasibility determination result determined according to the first prediction result of converting execution into bankruptcy and the second prediction result of converting execution into bankruptcy is low feasibility, the feasibility label of converting the execution of the case closed for finalization to be identified into bankruptcy is determined as low possibility.

[0111] After obtaining the feasibility label of converting execution into bankruptcy, store the feasibility labels of converting the execution of each case closed for finalization into bankruptcy in the case database. When a user queries and views a case, in response to the case query request, retrieve the case information corresponding to the case query request, display the case information display page, and add the feasibility label of converting execution into bankruptcy to the case information display page. Figure 3 Exemplarily shows the case information display page of an embodiment of the present application. After retrieving the corresponding case according to the filtering conditions selected by the user in the case filtering area, add the feasibility label of converting execution into bankruptcy to the case in the case list. The user can intuitively view the high or low feasibility of converting execution into bankruptcy identified by this method for identifying cases of converting execution into bankruptcy.

[0112] In this embodiment, after determining that the feasibility label of converting the execution of the case closed for finalization to be identified into bankruptcy is low possibility, the following steps are further included:

[0113] Generate a text for the reason for rejection according to the first prediction result of converting execution into bankruptcy of the expert system.

[0114] Specifically, generate a text for the reason for rejection according to the judgment result of the prediction rule and the prediction result generation rule of converting execution into bankruptcy of the expert system. For example, when the prediction rule corresponding to the number of executor patents is not satisfied, the generated text for the reason for rejection is "Having multiple patents may indicate innovative capabilities". Taking the above various characteristic quantities as examples, the factors (characteristic quantities) leading to rejection and the text for the reason for rejection are shown in Table 2 below. When there are multiple characteristic quantities leading to rejection, the text for the reason for rejection may include a combination of the reason texts of multiple characteristic quantities.

[0115] Table 2 Reason text

[0116]

[0117] Furthermore, the generated text of the rejection reason can also add ", the situation is complex and manual judgment is required" based on the reason text in Table 2 to prompt the user to further analyze and judge according to the actual situation of the case. This text of the rejection reason can be directly displayed in the case list on the case information display page, or can be displayed in the form of a pop-up window or floating layer when the user clicks on the possibility of execution conversion to bankruptcy column on the case information display page, or can be displayed by other means convenient for notifying the user, so as to facilitate the user to quickly capture the key information of the case and assist in improving the efficiency of the user's manual analysis and judgment.

[0118] This application combines an expert system and a machine learning model to obtain an execution conversion to bankruptcy case recognition model for comprehensive judgment, connects the expert system and the machine learning model, and realizes the preliminary screening and recheck of execution conversion to bankruptcy cases, so as to meet higher accuracy requirements. Using 566 real cases in the case database for testing, in the final results, TP (true positive) is 533, FP (false positive) is 4, FN (false negative) is 13, TN (true negative) is 16, the accuracy rate reaches 97.00%, the precision rate reaches 99.26%, the recall rate reaches 97.62%, and F1 reaches 98.43%, which can meet the accuracy requirements for the recognition of execution conversion to bankruptcy cases in practical applications.

[0119] As Figure 4 shown, the embodiment of this application also provides an execution conversion to bankruptcy case integration platform for implementing the above-mentioned execution conversion to bankruptcy case recognition method. The platform includes:

[0120] An information collection module M100, configured to call a multi-source data query interface according to the information of the person subject to enforcement of the final case to be recognized, query and obtain the retrieval data of multiple recognition indicators corresponding to the information of the person subject to enforcement, and obtain the feature data corresponding to the information of the person subject to enforcement based on the retrieval data;

[0121] An expert system module M200, configured to call an expert system based on the feature data corresponding to the information of the person subject to enforcement to obtain a first execution conversion to bankruptcy prediction result, and the expert system is configured to predict the feasibility of execution conversion to bankruptcy of the person subject to enforcement based on the preset execution conversion to bankruptcy prediction rules and feature data;

[0122] A result prediction module M300, configured to input the feature data corresponding to the information of the person subject to enforcement into the trained machine learning model, obtain the second execution conversion to bankruptcy prediction result output by the machine learning model, and determine the execution conversion to bankruptcy feasibility annotation of the final case to be recognized according to the first execution conversion to bankruptcy prediction result and / or the second execution conversion to bankruptcy prediction result;

[0123] The user interface module M400 is configured to retrieve the case information corresponding to the case query request in response to the case query request, display the case information display page, and add an annotation on the feasibility of converting enforcement to bankruptcy on the case information display page.

[0124] In the enforcement-to-bankruptcy case integration platform of the present application, the functions of each module can be implemented by the specific implementation manners of the enforcement-to-bankruptcy case identification method described above, which will not be elaborated here.

[0125] An embodiment of the present application further provides an enforcement-to-bankruptcy case identification device, including a processor; a memory storing executable instructions of the processor; wherein, the processor is configured to execute the steps of the enforcement-to-bankruptcy case identification method by executing the executable instructions.

[0126] Those skilled in the art can understand that various aspects of the present application can be implemented as a platform, a method, or a program product. Therefore, various aspects of the present application can be specifically implemented in the following forms, namely: a complete hardware implementation manner, a complete software implementation manner (including firmware, microcode, etc.), or an implementation manner combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "device" here.

[0127] Next, refer to Figure 5 to describe the electronic device 600 according to this embodiment of the present application. Figure 5 The displayed electronic device 600 is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present application.

[0128] As Figure 5 shown, the electronic device 600 is presented in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0129] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present application described in the above enforcement-to-bankruptcy case identification method part of this specification. For example, the processing unit 610 can execute the steps as Figure 1 shown.

[0130] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

[0131] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include, but are not limited to, an operating platform, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0132] The bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

[0133] The electronic device 600 may also communicate with one or more external devices 700 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or may communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through the input / output (I / O) interface 650. Also, the electronic device 600 may communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through the bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID platforms, tape drives, and data backup storage platforms, etc.

[0134] In the execution-to-bankruptcy case identification device, when the program in the memory is executed by the processor, the steps of the execution-to-bankruptcy case identification method are implemented. Therefore, the device can also obtain the technical effects of the above execution-to-bankruptcy case identification method.

[0135] The embodiment of the present application also provides a computer-readable storage medium for storing a program, and when the program is executed by the processor, the steps of the execution-to-bankruptcy case identification method are implemented. In some possible implementation manners, various aspects of the present application may also be implemented in the form of a program product, which includes program code. When the program product is executed on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present application described in the above execution-to-bankruptcy case identification method part of this specification.

[0136] Reference Figure 6As shown, a program product 800 for implementing the above method according to an embodiment of the present application is described. It can be a portable compact disc read-only memory (CD-ROM), include program code, and can be executed on a terminal device, such as a personal computer. However, the program product of the present application is not limited to this. In this document, a readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution platform, device, or component.

[0137] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor platform, device, or component, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0138] The computer-readable storage medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution platform, device, or component. The program code contained on the readable storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.

[0139] The program code for performing the operations of the present application can be written in any combination of one or more programming languages, including object-oriented programming languages - such as Java, C++, etc., and also including conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0140] When the program in the computer storage medium is executed by the processor, the steps of the above-mentioned method for identifying the execution-to-bankruptcy case are implemented. Therefore, the computer storage medium can also achieve the technical effects of the above-mentioned method for identifying the execution-to-bankruptcy case.

[0141] The above content is a further detailed description of the present application in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present application is only limited to these descriptions. For those of ordinary skill in the technical field to which the present application belongs, without departing from the concept of the present application, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present application.

Claims

1. A method for identifying execution-to-bankruptcy cases, characterized in that: The steps include: According to the information of the person subject to execution of the case to be identified and finalized, the multi-source data query interface is called to query and obtain the search data of multiple identification indicators corresponding to the information of the person subject to execution, and the characteristic data corresponding to the information of the person subject to execution is obtained based on the search data; Based on the characteristic data corresponding to the information of the person subject to execution, calling an expert system to obtain a first execution-to-bankruptcy prediction result, wherein the expert system is configured to predict the feasibility of the execution-to-bankruptcy of the person subject to execution based on a preset execution-to-bankruptcy prediction rule and the characteristic data; Inputting the characteristic data corresponding to the information of the person subject to execution into a trained machine learning model, obtaining a second execution-to-bankruptcy prediction result output by the machine learning model, and determining the feasibility labeling of execution-to-bankruptcy of the to-be-identified final case according to the first execution-to-bankruptcy prediction result and / or the second execution-to-bankruptcy prediction result; In response to a case query request, the case information corresponding to the case query request is retrieved, a case information display page is displayed, and the execution-to-bankruptcy feasibility mark is added to the case information display page.

2. The method for identifying execution-to-bankruptcy cases according to claim 1, characterized in that: The calling of the multi-source data query interface to query and obtain the search data of multiple identification indicators corresponding to the information of the person to be executed includes the following steps: Determine the query interface and search keywords for each identification indicator; Based on the query keywords of each identification indicator, the query interface is called to retrieve the data of the identification indicator on the information platform to obtain the retrieval data of multiple identification indicators.

3. The method for identifying execution-to-bankruptcy cases according to claim 2, characterized in that: Acquiring characteristic data corresponding to the information of the person to be executed based on the search data includes the following steps: Determine the identification index and feature generation rule corresponding to each feature; Based on the search data of the identification index corresponding to each feature quantity and the feature quantity generation rule, each feature quantity is generated, and the feature quantities are combined to obtain the feature data corresponding to the information of the person to be executed.

4. The method for identifying execution-to-bankruptcy cases according to claim 1, characterized in that: Before inputting the characteristic data corresponding to the information of the person to be executed into the trained machine learning model, the following steps are also included: Determining whether the first execution-to-bankruptcy prediction result corresponds to high feasibility or low feasibility; When the first execution-to-bankruptcy prediction result corresponds to high feasibility, determining the execution-to-bankruptcy feasibility label of the to-be-identified final case according to the first execution-to-bankruptcy prediction result; When the first execution-to-bankruptcy prediction result corresponds to low feasibility, the feature data corresponding to the information of the person subject to execution is input into a trained machine learning model.

5. The method for identifying execution-to-bankruptcy cases according to claim 4, characterized in that: The following steps are used to determine the feasibility mark of the execution-to-bankruptcy conversion of the final case to be identified: When the first execution-to-bankruptcy prediction result corresponds to high feasibility, the feasibility of execution-to-bankruptcy of the to-be-identified final case is determined to be marked as extremely high possibility; When the first execution-to-bankruptcy prediction result corresponds to low feasibility and is determined to be highly feasible based on the first execution-to-bankruptcy prediction result and the second execution-to-bankruptcy prediction result, the feasibility of execution-to-bankruptcy of the to-be-identified finalized case is marked as high possibility; The first execution-to-bankruptcy prediction result corresponds to low feasibility, and when it is determined to be low feasibility based on the first execution-to-bankruptcy prediction result and the second execution-to-bankruptcy prediction result, the execution-to-bankruptcy feasibility of the case to be identified and finalized is marked as low possibility.

6. The method for identifying execution-to-bankruptcy cases according to claim 1, characterized in that: The method further includes the following steps before calling the multi-source data query interface based on the information of the person subject to execution of the case to be identified and querying and obtaining the search data of multiple identification indicators corresponding to the information of the person subject to execution: Based on procedural pre-selection rules, select cases to be identified and finalized from the execution case database.

7. The method for identifying execution-to-bankruptcy cases according to claim 1, characterized in that: The calling of the expert system to obtain the first execution-to-bankruptcy prediction result comprises the following steps: Based on the preset execution-to-bankruptcy prediction rules, the characteristic quantity corresponding to each execution-to-bankruptcy prediction rule is determined, and the determination result corresponding to each execution-to-bankruptcy prediction rule is obtained; According to the determination results and prediction result generation rules corresponding to the plurality of execution-to-bankruptcy prediction rules, a first execution-to-bankruptcy prediction result is obtained.

8. A platform for integrating execution-to-bankruptcy cases, characterized in that: The platform is used to implement the method for identifying execution-to-bankruptcy cases according to any one of claims 1 to 7, and comprises: An information collection module is used to call a multi-source data query interface based on the information of the person to be executed in the case to be identified, query and obtain retrieval data of multiple identification indicators corresponding to the information of the person to be executed, and obtain feature data corresponding to the information of the person to be executed based on the retrieval data; An expert system module, for invoking an expert system based on the characteristic data corresponding to the information of the person subject to execution to obtain a first execution-to-bankruptcy prediction result, wherein the expert system is configured to predict the feasibility of the execution-to-bankruptcy of the person subject to execution based on a preset execution-to-bankruptcy prediction rule and the characteristic data; A result prediction module is used to input the characteristic data corresponding to the information of the person subject to execution into a trained machine learning model, obtain a second execution-to-bankruptcy prediction result output by the machine learning model, and determine the feasibility label of execution-to-bankruptcy of the to-be-identified final case according to the first execution-to-bankruptcy prediction result and / or the second execution-to-bankruptcy prediction result; The user interface module is used to respond to a case query request, retrieve the case information corresponding to the case query request, display a case information display page, and add the execution-to-bankruptcy feasibility mark in the case information display page.

9. An execution-to-bankruptcy case identification device, characterized in that: include: processor; a memory storing executable instructions of the processor; Wherein, the processor is configured to execute the steps of the execution-to-bankruptcy case identification method described in any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium for storing a program, characterized in that: When the program is executed by a processor, the steps of the method for identifying execution-to-bankruptcy cases described in any one of claims 1 to 7 are implemented.

Images