Case distribution and classification method in mediation stage before financial dispute complaint

Abstract

The invention belongs to the technical field of computers, and particularly relates to a case distribution and classification method in a mediation stage before financial disputes and complaints, which comprises the steps of data cleaning, classification marking, calculation of a classification result interval value and data distribution. The data cleaning comprises data preprocessing, data verification and data processing. The data preprocessing comprises the steps of deleting duplicated items, processing missing values, correcting errors, filling the missing values, recognizing abnormal values and cleaning duplicated data. The data verification comprises jurisdiction verification and evidence integrity verification. The classification mark comprises a first classification standard determined according to the mediation result and a second classification standard determined according to the involved amount. The data distribution result comprises immediate import of a pre-complaint mediation program, signing of a mediation protocol, judicial confirmation of mediation and settlement, forcing notarization and the like. According to the method, a more reasonable case distribution and classification standard is established, and the case data is clearer and more real through data cleaning, so that the efficiency of case distribution and classification work and pre-complaint mediation work is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular relates to a case diversion and classification method in the pre-litigation mediation stage of financial disputes. Background Art

[0002] The diversion and classification of financial contract disputes and credit card disputes (hereinafter collectively referred to as: financial disputes) during the pre-litigation mediation stage is an important basic task for the courts in handling financial dispute cases.

[0003] In judicial practice, the most commonly used method for classifying and diverting financial disputes during the pre-litigation mediation stage is the simple and complex classification method, which categorizes and diverts cases based on factors such as the difficulty, nature, and amount involved, thereby helping the courts to quickly handle financial disputes. However, the method also has obvious flaws, which are specifically reflected in the following five aspects:

[0004] First, the classification standards are not entirely reasonable. For example, some courts overemphasize the impact of the amount involved on the classification of cases (dividing the cases into more than five levels based on the amount involved), but do not pay enough attention to other factors that affect the classification of cases.

[0005] Second, since case diversion must be based on the results of case classification, if the classification standards are not reasonable, the results of case diversion will inevitably have corresponding defects.

[0006] Third, the limitations of manual identification: relying on the experience of individual judges or case officers to identify complex and simple characters is subjective and unstable.

[0007] Fourth, ignoring the individuality of cases: ignoring the particularities of individual cases results in some cases not receiving proper attention and handling.

[0008] Fifth, the procedural conversion is not flexible: In actual operations, the procedural conversion from complex to simplified or from simplified to complex is not flexible enough, which affects the timeliness of case handling.

[0009] In addition to the above-mentioned defects, courts often face the problem of "difficulty in service" when handling case diversion and classification in the pre-trial mediation stage of financial disputes: and the problem of "difficulty in service" will also affect the implementation effect of case diversion and classification. Summary of the invention

[0010] The purpose of the present invention is to overcome the defects of the above-mentioned prior art in the case diversion and classification in the pre-trial mediation stage of financial disputes, and effectively solve the problem of "difficult delivery", so as to quickly and accurately solve the case diversion and classification in the pre-trial mediation stage of financial disputes, and lay the foundation for the court to efficiently handle financial disputes.

[0011] In order to achieve the above object, the present invention adopts the following technical solutions:

[0012] A method for case diversion and classification in the pre-litigation mediation stage of financial disputes. The system implementing the method includes a memory, a processor and a communication device. The memory is communicatively connected to the processor, and the processor is communicatively connected to the communication device. A computer program for the processor to execute the following steps (each step includes a basic step, a specific step and a sub-step) is installed in the memory. The communication device includes an encryption module or a receiving module, and a decryption module matched with the encryption module or the receiving module, or a key matched with the decryption module is also provided in the memory or the processor. The method includes the following four basic steps, namely, step 1, data cleaning, step 2, classification marking, step 3, calculating the classification result interval value (the interval value consists of a quantitative calculation result value and a qualitative identification code, that is, a numerical part and an identification code), and step 4, data diversion; Step 1 uses a computer program in the system (i.e., a computer program installed in the memory) to automatically screen and process data, including three specific steps, namely, step 101, data preprocessing, step 102, data verification, and step 103, data processing; step 101 includes six sub-steps: deleting duplicates, processing missing values, correcting errors, filling missing values, identifying outliers, and cleaning duplicate data; step 102 includes two sub-steps: jurisdiction verification and evidence integrity verification; step 103 includes two sub-steps: AI intelligent outbound calling and lost connection repair; Step 2 includes two specific steps, namely, step 201, determining the first classification standard according to the mediation result, and step 202, determining the second classification standard according to the amount involved; the first classification standard determined in step 201 includes four types: mediatable, no willingness to mediate, completely lost contact, and no civil compensation capacity; the second classification standard determined in step 202 includes two types: large amount and small amount; Step 3 The system uses the formula K=Y×(X1+X2+X3)÷3+C m +R n +Z qCalculate the classification result interval value, where K represents the classification result interval value, Y represents jurisdiction; the value of Y is 0, 0.5 or 1, where 0 represents no jurisdiction, 0.5 represents jurisdiction but there is jurisdictional objection, and 1 represents jurisdiction and no jurisdictional objection; X1 represents the influencing factor of the amount involved, and if the amount involved is within the small amount standard range, X1=1; if the amount involved exceeds the standard range of small amount, X1=1-(D-d)÷2D, where D is the amount involved and d is the lower limit of large amount; X2 represents the guarantee impact factor. If no guarantee is provided, X2=0; if a guarantee is provided, X2=0.85+(P-D)÷2(P+D), where P is the value of the secured property. If there is no secured property but the credit of a third party is used as a guarantee, then X2=0.85; X3 represents the completeness of evidence. The value of X3 is 0 or 1. 0 means that the evidence is incomplete and 1 means that the evidence is complete. If the evidence is incomplete, the system prompts the party to complete the evidence and calculates it after the evidence is complete; C m is the first identification code, and the value of its subscript m is 1 or 2. When m=1, C m That is, C1, indicating that the debtor has the ability to pay civil compensation. When m=2, C m That is C2, indicating that the debtor has no civil compensation capacity; R n is the second identification code, the value of its subscript n is 1 or 2. When n=1, R n That is, R1, indicating that the system can get in touch with both parties. When n=2, R n That is R2, indicating that one party is completely out of contact; Z q is the third identification code, the value of its subscript q is 1 or 2. When q=1, Z q That is Z1, which means it can be adjusted. When q=2, Z q That is Z2, indicating no willingness to mediate; Step 4 performs case diversion based on the above steps; the case diversion results include two basic types: immediate introduction to the pre-litigation mediation procedure and direct transfer to the litigation procedure. Among them, the conditions for immediate introduction to the pre-litigation mediation procedure are: K>0.5, the first identification code is C1, the second identification code is R1, and the third identification code is Z1; the conditions for direct transfer to the litigation procedure are: K>0.5, the first identification code is C1, the second identification code is R2 or the third identification code is Z2. It should also be noted that if 0<K<0.5, it means that there is a jurisdictional objection, and the system prompts to resolve the jurisdictional objection first, and then proceed with the subsequent procedures. If K=0, it means that the debtor has no civil compensation capacity. In addition, according to the calculation formula and assignment rules of the present invention, the situation of K=0.5 will not actually occur (excluding the application of the rounding rule). Therefore, as long as 0<K<0.5, it is because there is a jurisdictional objection.

[0013] On the basis of the above technical solution, the present invention can add the following technical means to better solve the technical problem to be solved by the present invention:

[0014] If the numerical part of the classification result interval value K is ≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z1, the system prompts the court to include the case in the fast mediation channel of the pre-trial mediation procedure.

[0015] Furthermore, for multiple cases that meet the statutory conditions for consolidated trial, if the numerical part of the classification result interval value K of each case is ≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z2, the system prompts the court to give priority to consolidated trial.

[0016] Furthermore, the case diversion results of implementing step 4 also include four specific types: signing a mediation agreement, judicial confirmation of mediation settlement, compulsory notarization, and classification as having no civil compensation capacity (case K=0). It should be noted that the signing of a mediation agreement is a positive result produced after the immediate introduction of the pre-trial mediation procedure. After the introduction of the pre-trial mediation procedure, the parties may not be able to reach a mediation agreement due to differences of opinion, and they still need to enter the litigation procedure. In addition, during the mediation process, the debtor may also provide evidence to prove that he has no civil compensation capacity, in which case the case needs to be classified as having no civil compensation capacity. As for judicial confirmation of mediation settlement and compulsory notarization, they are two ways of handling after the signing of the mediation agreement, and the parties can choose one of them.

[0017] Furthermore, when implementing step 1, the system's cooperative inquiry function is used to inquire about the party's mobile phone number with the three major operators, the household registration address registered with the public security agency, and the permanent address for more than six months.

[0018] Furthermore, after executing the immediate introduction of the pre-litigation mediation procedure described in step 4, the system prompts the parties to be served with a mediation notice.

[0019] Compared with the prior art, the present invention has the following beneficial effects: First, the present invention applies computer technology, artificial intelligence technology, and big data technology to the case diversion and classification work in the pre-trial mediation stage of financial disputes, greatly improving the efficiency of case diversion and classification work and pre-trial mediation work.

[0020] Second, by cleaning case data, deleting duplicate items, erroneous items, filling missing values, identifying outliers, cleaning duplicate data, and using intelligent applications such as AI outbound calls and lost connection repair to update and repair case data, the present invention makes case data clearer and more authentic.

[0021] Third, a more reasonable case classification standard has been established. This standard not only takes into account the impact of the amount involved on case classification, but also pays more attention to the impact of mediation results (referring to the statistically significant mediation results obtained by the applicant in handling a large number of pre-trial mediation work on financial disputes in the past) on case classification. On the basis of case classification, corresponding classification marks and operational case diversion measures are adopted.

[0022] Fourth, by calculating the interval values ​​of the classification results and determining three identification codes, the system can better provide a basis for case diversion in a quantitative and qualitative manner.

[0023] Fifth, in the system of the present invention, the communication device includes an encryption module or a receiving module, and a decryption module matching the encryption module or the receiving module, or a key matching the decryption module is also provided in the memory or the processor, which makes the present invention safer than the existing system in terms of the transmission of case information data. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a system structure block diagram of an embodiment of the present invention; Figure 2 is a schematic diagram of a security structure of a communication device in the system; Figure 3 is a schematic diagram of another security structure of the communication device; Figure 4 It is a logic block diagram of the basic steps of this embodiment; Figure 5 is a logic block diagram of step 1 (data cleaning) in this embodiment; Figure 6 FIG. 4 is a logic block diagram of step 2 (classification labeling) in this embodiment. DETAILED DESCRIPTION

[0025] In order to facilitate those skilled in the art to fully understand the technical solution, an embodiment of the present invention is described below in conjunction with the accompanying drawings.

[0026] First of all, it should be explained that, in terms of hardware, the existing systems (including the systems previously developed by the applicant and recorded in the invention patent specification CN114201951B and the invention patent specification CN114238418B) can in principle be used to implement the present invention. Of course, in order to improve the security of case information data transmission and thus better implement the technical solution of the present invention, the applicant has developed an intelligent system (hereinafter referred to as: system) for implementing the case diversion and classification method in the pre-litigation mediation stage of financial disputes based on the existing system. Figure 1As shown, the system includes a memory, a processor and a communication device. The memory is connected to the processor in communication, and the processor is connected to the communication device in communication. The memory is installed with a computer program for implementing the following steps of the present invention and stores relevant case information data (including case information data submitted by the parties, and case information data obtained after the processor processes the case information data submitted by the parties according to the computer program). The processor is used to execute the computer program, and the communication device is used to transmit the case information data obtained after the processor processes the case information data. In this embodiment, the communication device can adopt two security structures: one is, Figure 2 As shown, the communication device includes a receiving module and a decryption module matched with the receiving module, and the communication device is used to transmit case data information that has been encrypted by a memory or a processor (that is, it is applicable to the case where the memory or the processor is provided with a corresponding key); secondly, as Figure 3 As shown, the communication device includes an encryption module and a decryption module matched with the encryption module, and the communication device is used to transmit unencrypted case data information. Compared with the existing system that also includes a memory, a processor and a communication device, the present invention improves the security of case data information transmission by encrypting the case data information. As for the specific encryption processing method, the existing technical means in the prior art can be used, which will not be repeated here. Taking the system in the present invention as the material technical means, the present invention provides a case diversion and classification method for the pre-trial mediation stage of financial disputes, such as Figure 4 Combined with Figure 5 , Figure 6 As shown, the method includes four basic steps, namely, step 1, data cleaning, step 2, classification marking, step 3, calculating the interval value of the classification result, and step 4, data diversion.

[0027] Step 1 includes three specific steps, namely, step 101, data preprocessing; step 102, data verification; step 103, data processing.

[0028] Step 101 includes the following six sub-steps:

[0029] (1) Deleting duplicate items. For example, if multiple lawsuits are filed for the same case, the case data of duplicate lawsuits should be deleted;

[0030] (2) Processing missing values ​​refers to discovering and confirming data that should have been filled in but was not filled in (partial missing data). For example, the delivery address was only filled in Chancheng District, Foshan City, Guangdong Province, without the specific street number, etc.;

[0031] (3) Correcting errors; For example, if the registered permanent address of the party has changed, it is necessary to enter the changed registered permanent address in the system. In this embodiment, the system's cooperative inquiry function can be used to query the party's mobile phone number with the three major operators, the registered permanent address registered with the public security agency, and the party's permanent address for more than six months confirmed by the public security agency or postal express.

[0032] (4) Fill in missing values; input some missing data into the system to make the data complete.

[0033] (5) Identify outliers: Filter out outliers by inconsistencies between data, such as principal, interest and other data. If there are inconsistent data in the litigation request, claim form and element table, the system will automatically filter them out. In principle, the system will notify the financial institutions and agents to correct and confirm the relevant data through inquiries. The data will not be overwritten, but a new piece of data will be added and marked. When using data, the latest data must be used.

[0034] (6) Clean up duplicate data, that is, find and filter repeated data, such as duplicate phone numbers.

[0035] Step 102 includes two sub-steps: jurisdiction verification and evidence integrity verification.

[0036] Step 103 includes two sub-steps: AI intelligent outbound calling and lost connection repair. Since the results of intelligent outbound calling belong to the content of data cleaning, this embodiment incorporates AI intelligent outbound calling into the sub-step of step 101 (data cleaning). The purpose of intelligent outbound calling is to collect the opinions of the parties. The parties usually have different opinions, which is an important basis for case classification (classification marking). Through AI intelligent outbound calling, it is possible that one party (debtor) cannot be contacted. At this time, it is necessary to use the means of lost connection repair to contact the party. If the party still cannot be contacted through lost connection repair, it means that the party cannot be contacted at all. When classifying and marking, the case needs to be classified as a completely lost connection type.

[0037] It should be noted that when implementing step 1, the present invention uses a computer program in the system (i.e., a computer program installed in the memory) to automatically screen and process case data, which not only reduces the workload of judges and relevant case handlers, but also makes the data more real and objective, greatly reducing the probability of data errors. In addition, the present invention also delivers mediation notices to the parties through the court's centralized delivery platform, so that the parties can understand the situation in a timely manner.

[0038] Step 2 includes two specific steps, namely, step 201, determining the first classification standard according to the mediation results (the mediation results here refer to the statistically significant mediation results obtained by the applicant in handling a large number of pre-litigation mediation work on financial disputes in the past), and step 202, determining the second classification standard according to the amount involved. The first classification standard determined in step 201 includes four types: mediatable, no willingness to mediate (the reason why it is called no willingness to mediate is that after the case is transferred to the litigation procedure, it is still possible to settle the case by mediation), completely lost contact, and no civil compensation ability; the second classification standard determined in step 202 includes two types: large amount and small amount. It should be noted that the standard for dividing large amount and small amount is a dynamic standard that changes with the change of the average annual income of residents. In this embodiment, the applicant uses 50% of the average annual income of local urban residents in the previous year as the standard for distinguishing large amount and small amount. If the amount involved is greater than or equal to the amount corresponding to the standard, it is determined as a large amount, and if the amount involved is less than (excluding the principal amount) the amount corresponding to the standard, it is determined as a small amount. Empirical facts show that when the amount involved is less than 50 percent of the local urban residents' average annual income in the previous year, more than 80 percent of the debtors have the ability to repay immediately.

[0039] Step 3 The system uses the formula K=Y×(X1+X2+X3)÷3+C m +R n +Z q Calculate the classification result interval value.

[0040] In the above formula, K represents the classification result interval value, Y represents jurisdiction, and the value of Y is 0, 0.5 or 1, where 0 represents no jurisdiction, 0.5 represents jurisdiction but there is a jurisdictional objection, and 1 represents jurisdiction and no jurisdictional objection. It should be noted that according to the statutory civil procedure, the jurisdictional objection is raised by the defendant during the submission of the reply after receiving the copy of the complaint. However, based on the batch and typological characteristics of financial dispute cases and the need to quickly resolve disputes, the present invention actually considers the jurisdictional objection issue before the court transfers the copy of the complaint to the defendant (this is also a distinguishing technical feature of the present invention that is different from other inventions involving evidence verification). Therefore, the jurisdictional objection mentioned in the present invention actually refers to the high probability of the existence of jurisdictional objection. For example, when multiple courts, including the District A Court, have jurisdiction over a case in accordance with the law, if the parties have agreed in the contract on specific applicable conditions for the jurisdiction of the District A Court, but the creditor does not attach evidence that the specific applicable conditions have been met when filing a lawsuit with the District A Court, the system will treat this situation as a case of objection to jurisdiction and prompt the creditor to submit evidence that the specific applicable conditions have been met.

[0041] The meanings of the other letters in the above formula are as follows: X1 represents the influencing factor of the amount involved in the case. If the amount involved in the case is within the standard range of small amounts, X1=1. If the amount involved in the case exceeds the standard range of small amounts, X1=1-(D-d)÷2D, where D is the amount involved in the case and d is the lower limit of the large amount. For example, assuming that the lower limit of the large amount d is 40,000 yuan and the amount involved D is 80,000 yuan, then: X1=1-(80000-40000)÷160000=0.75 The larger the value of the factor affecting the amount involved in the case (the maximum value is 1), the easier it is for the debtor to repay the debt.

[0042] X2 represents the guarantee impact factor. If no guarantee is provided, X2=0. If a guarantee is provided, X2=0.85+(P-D)÷2(P+D). In the formula, P is the value of the secured property. If the value of the secured property exceeds the amount involved in the case and a separate guarantee is provided, P-D=0 is treated; if P÷D>1.85, P÷D=1.85 is treated, that is, it is presumed that the value of the secured property exceeds 1.85 times the amount involved in the case and a separate guarantee is provided. For example, assuming that the amount involved in the case is 100,000 yuan and the value of the secured property is 150,000 yuan, then: X2=0.85+(150000-100000)÷2(150000+100000)=0.95; Assuming that the amount involved in the case is RMB 100,000 and the value of the secured property is RMB 200,000, then the result of P÷D is greater than 1.85: then P÷D = 1.85, that is, P = 1.85 × 100,000 = 185,000 (RMB), thus obtaining: X2=0.85+(185000-100000)÷2(185000+100000)=0.999 If the value of the secured property is lower than the amount involved in the case, then X2 < 0.85 (generally speaking, the value of the secured property should be equal to or higher than the amount involved in the case, but the actual value of the secured property may be lower than the amount involved in the case due to improper custody or changes in market supply and demand). If there is no secured property but the credit of a third party is used as collateral, the value of the guarantee impact factor is set to 0.85, that is, X2 = 0.85.

[0043] In general, under the premise that there is no separate guarantee, the higher the value of the secured property, the greater the value of the guarantee impact factor X2, and the higher the probability that the creditor will realize his creditor's rights. However, in order to balance the interests between the creditor and the debtor, the present invention presumes that the value of the secured property exceeds 1.85 times the amount involved in the case and is separately secured. Of course, when specifically determining the value of the secured property, the parties (first of all, the creditor as the plaintiff) should provide evidence. If the evidence provided does not meet the requirements, when the system executes step 102, the evidence related to the secured property cannot pass the evidence integrity verification. It should be emphasized that in actual work, if the secured property belongs to the important means of production that the debtor is currently using, and both parties agree to mediation, the applicant will try to persuade the creditor to agree to the debtor's appropriate extension of repayment so as to give full play to the role of the means of production, and at the same time, it is also convenient for the debtor to create new value by using the secured property, thereby improving its debt repayment ability.

[0044] X3 indicates the completeness of evidence. The value of X3 is 0 or 1. 0 indicates that the evidence is incomplete, and 1 indicates that the evidence is complete. In the case of incomplete evidence, the system will prompt the parties to complete the evidence, and the calculation will be performed again after the evidence is complete.

[0045] C m is the first identification code. The identification code itself is not assigned a value. The value of its subscript m is 1 or 2. When m=1, C m That is, C1, indicating that the debtor has the ability to pay civil compensation. When m=2, C m That is C2, which means that the debtor has no civil compensation capacity. For example, if there is evidence that the debtor has lost his ability to work, his property can only meet his basic living needs, and he has no other source of income, then the debtor can be deemed to have no civil compensation capacity. It is worth noting that if the debtor has no civil compensation capacity, if the debtor provides property guarantee when borrowing from a financial institution, then the debtor can be deemed to have civil compensation capacity, except that the property is necessary to meet his basic living needs.

[0046] R n is the second identification code. The identification code itself is not assigned a value. The value of its subscript n is 1 or 2. When n=1, R n That is R1, indicating that the system can get in touch with both parties. When n=2, R n That is R2, which means that one party is completely out of touch. Generally speaking, the party that is likely to be completely out of touch is the debtor who is the defendant, while the creditor usually does not have the problem of being out of touch.

[0047] Z q is the third identification code. This identification code itself is not assigned a value. The value of its subscript q is 1 or 2. When q=1, Z qThat is Z1, which means it can be adjusted. When q=2, Z q That is Z2, which means there is no willingness to mediate for the time being.

[0048] From the 130 cases that the applicant has randomly counted, the numerical part of the classification result interval value K of 129 cases is between 0.502 and 0.999 (12 of them originally had the problem of incomplete evidence, resulting in the numerical part of K being only 0.251. After the evidence was improved, the numerical part of K reached 0.502. This treatment can ensure that the numerical part of K is distinguished from the situation of disputed jurisdiction), and only 1 case caused the numerical part of the classification result interval value K to be only 0.333 due to jurisdictional objection. As for the situation where there is no jurisdiction and the numerical part of the classification result interval value K is 0, the applicant has not actually encountered it in the preliminary work of the present invention, but this possibility is completely possible. It should also be noted that according to the calculation formula and assignment rules of the present invention, without considering rounding, the situation of K=0.5 will not actually occur. Therefore, as long as 0<K<0.5, it is because there is a jurisdictional objection. On the contrary, as long as K>0.5, it is sufficient to show that the court has jurisdiction.

[0049] It should be emphasized that if the numerical part of the classification result interval value K is ≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z1, the probability of successful mediation is very high, so the system prompts the court to include the case in the fast mediation channel of the pre-litigation mediation procedure (that is, give priority to mediating such cases and resolve disputes between the parties as soon as possible). For multiple cases that meet the statutory conditions for consolidated trial, if the numerical part of the classification result interval value K is ≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z2, the parties will usually agree to consolidate the trial, and the probability of obeying the first-instance judgment is very high, so the system will prompt the court to give priority to consolidated trial for such cases.

[0050] Step 4: Case diversion is performed on the basis of the above steps. The case diversion results include immediate introduction to the pre-litigation mediation procedure (referred to as: pre-litigation mediation), signing of a mediation agreement, judicial confirmation of the settlement of the mediation (referred to as: judicial confirmation), mandatory notarization, transfer to the litigation procedure (referred to as: litigation), and classification as having no civil compensation capacity. Among them, immediate introduction to the pre-litigation mediation procedure and transfer to the litigation procedure are two basic types of case diversion, and signing of a mediation agreement, judicial confirmation, mandatory notarization, and classification as having no civil compensation capacity belong to specific types of case diversion. In this embodiment, pre-litigation mediation is used to divert litigation, and the result of pre-litigation mediation determines whether the case will go through judicial confirmation or mandatory notarization (successful mediation), or litigation (unable to mediate) or classification as having no civil compensation capacity (returned to the bank for self-disposal). It should also be noted that if it is indeed impossible to contact the party (debtor), the case needs to be directly transferred to the litigation procedure, and the court will try to resolve the problem of contacting the party during the litigation procedure.

[0051] The result of case diversion directly corresponding to the case classification standard of no civil compensation capacity is to classify the case as no civil compensation capacity. In some cases, after entering the pre-litigation mediation procedure, based on the evidence provided by the debtor, it can be confirmed that the debtor does not have the civil compensation capacity. In this case, it is necessary to classify the case as no civil compensation capacity and hand it over to the creditor (financial institution) to handle it on its own, in order to avoid wasting judicial resources.

[0052] The above describes in detail the working steps in the implementation process of the method for case diversion and classification (hereinafter referred to as case diversion and classification) in the pre-litigation mediation stage of financial disputes of the present invention through an embodiment. The following briefly introduces the relevant matters in the implementation process of the present invention:

[0053] First, the element data in the present invention originates from financial institutions, and can be directly extracted from the data submitted by financial institutions to the court system, or manually searched and filled in from the paper materials sorted out by financial institutions. In short, the sorting of element data is completed by the financial institutions, rather than the court extracting element data on demand from the paper materials after the financial institutions submit the paper materials.

[0054] Second, in accordance with relevant laws, judicial interpretations and other normative documents, normative documents and element tables for element data shall be formulated, and financial institutions shall organize and submit evidence materials (paper materials and scanned electronic materials) and evidence element data in accordance with the regulations.

[0055] Third, only by standardizing factor data can automated evidence verification be possible, reducing errors in manual verification and improving verification efficiency.

[0056] Fourth, data elementization is a necessary condition to simplify the intensity of compiling various documents (outbound call script matching).

[0057] Fifth, the court's artificial intelligence case-handling system includes subsystems such as the court's pre-trial mediation system and the court's element approval system. The court's pre-trial mediation system is a prerequisite system for the court's element approval system. This system completes statutory procedures such as evidence verification, evidence exchange, evidence examination, evidence collection, confirmation of delivery methods, address confirmation, and recording of undisputed facts in advance.

[0058] It should be noted that in the process of implementing the present invention, the applicant also verified the technical effect of the present invention in the classification of financial dispute cases. The specific verification method is: a panel of three experienced judges who are familiar with financial laws and related businesses evaluates the classification results of 100 randomly selected cases made by the artificial intelligence case handling system according to the above steps. During the evaluation process, the panel evaluates the correctness of the classification results of the cases according to the principle of minority obeys majority. Among the classification results of these 100 cases, the classification results of 98 cases are unanimously determined to be correct by the three members of the panel, and the classification results of the remaining two cases are determined to be correct by two members of the panel, and only one member of the panel has a different opinion (the reason for the different opinions is that the collateral properties involved in the two cases are all intangible properties. Unlike the relatively stable value of tangible objects, the value of intangible properties varies greatly due to different evaluation methods, and the value evaluated by the same method also has a large volatility). According to the principle of minority obeys majority, the classification results of the remaining two cases are also correct. It can be seen that the present invention has a fairly high credibility in the classification of financial dispute cases. Correct classification lays a solid foundation for the diversion of cases.

Claims

1. A method for classifying and sorting cases in the pre-litigation mediation stage of financial disputes, wherein a system implementing the method comprises a memory, a processor and a communication device, wherein the memory is communicatively connected to the processor, and the processor is communicatively connected to the communication device, and wherein the method is characterized in that: A computer program for the processor to execute the following steps is installed in the memory, the communication device includes an encryption module or a receiving module, and a decryption module matched with the encryption module or the receiving module, or a key matched with the decryption module is also provided in the memory or the processor, and the method includes the following four basic steps, namely, step 1, data cleaning, step 2, classification marking, step 3, calculating the interval value of the classification result, and step 4, data diversion; Step 1 uses the computer program in the system to automatically screen and process data, including three specific steps, namely step 101, data preprocessing, step 102, data verification, and step 103, data processing; step 101 includes six sub-steps: deleting duplicates, processing missing values, correcting errors, filling missing values, identifying outliers, and cleaning duplicate data; step 102 includes two sub-steps: jurisdiction verification and evidence integrity verification; step 103 includes two sub-steps: AI intelligent outbound calling and lost connection repair; Step 2 includes two specific steps, namely, step 201, determining the first classification standard according to the mediation result, and step 202, determining the second classification standard according to the amount involved; the first classification standard determined in step 201 includes four types: mediatable, no willingness to mediate, completely lost contact, and no civil compensation capacity; the second classification standard determined in step 202 includes two types: large amount and small amount; Step 3 The system uses the formula K=Y×(X1+X2+X3)÷3+C m +R n +Z q Calculate the classification result interval value, where K represents the classification result interval value, Y represents jurisdiction; the value of Y is 0, 0.5 or 1, where 0 represents no jurisdiction, 0.5 represents jurisdiction but there is jurisdictional objection, and 1 represents jurisdiction and no jurisdictional objection; X1 represents the influencing factor of the amount involved, and if the amount involved is within the small amount standard range, X1=1; if the amount involved exceeds the standard range of small amount, X1=1-(D-d)÷2D, where D is the amount involved and d is the lower limit of large amount; X2 represents the guarantee impact factor. If no guarantee is provided, X2=0; if a guarantee is provided, X2=0.85+(P-D)÷2(P+D), where P is the value of the secured property. If there is no secured property but the credit of a third party is used as a guarantee, then X2=0.85; X3 represents the completeness of evidence. The value of X3 is 0 or 1. 0 means that the evidence is incomplete and 1 means that the evidence is complete. If the evidence is incomplete, the system prompts the party to complete the evidence and calculates it after the evidence is complete; C m is the first identification code, and the value of its subscript m is 1 or 2. When m=1, C m That is, C1, indicating that the debtor has the ability to pay civil compensation. When m=2, C m That is C2, indicating that the debtor has no civil compensation capacity; R n is the second identification code, the value of its subscript n is 1 or 2. When n=1, R n That is R1, indicating that the system can get in touch with both parties. When n=2, R n That is R2, indicating that one party is completely out of touch; Z q is the third identification code, the value of its subscript q is 1 or 2. When q=1, Z q That is Z1, which means it can be adjusted. When q=2, Z q That is Z2, indicating no willingness to mediate; Step 4 performs case diversion based on the above steps; the case diversion results include two basic types: immediate introduction to pre-trial mediation procedures and direct transfer to litigation procedures. The condition for immediate introduction to pre-trial mediation procedures is: K>0.5, the first identification code is C1, the second identification code is R1, and the third identification code is Z1; the condition for direct transfer to litigation procedures is: K>0.5, the first identification code is C1, the second identification code is R2 or the third identification code is Z2.

2. The case classification method for pre-litigation mediation of financial disputes as claimed in claim 1 is characterized by: If the classification result interval value K≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z1, the system prompts to include the case in the fast mediation channel of the pre-litigation mediation procedure.

3. The case classification method for pre-litigation mediation of financial disputes as claimed in claim 1 is characterized by: For multiple cases that meet the statutory conditions for consolidated trial, if the numerical part of the classification result interval value K of each case is ≥0.85, and the first identification code is C1, the second identification code is R1, and the third identification code is Z2, the system prompts to give priority to consolidated trial.

4. The case diversion and classification method for the pre-litigation mediation stage of financial disputes as claimed in any one of claims 1 to 3, characterized in that: After executing the immediate introduction of the pre-litigation mediation procedure described in step 4, the system prompts the parties to be served with a mediation notice.

Images