Intelligent case processing method and system for litigation mediation and medium

Abstract

The invention relates to the technical field of judicial informatization and artificial intelligence, and particularly provides an intelligent case processing method and system for litigation mediation and a medium. The method comprises the steps that litigation data of appeals are obtained and filled into an appeal form, the compliance state of the appeal form is judged, if the appeal form is compliant, keyword extraction is conducted on the appeal form, vector conversion processing is conducted, an appeal semantic vector is obtained, the appeal semantic vector is input into a preset appeal classification model to be processed, and the appeal data of the appeal form is obtained. Obtaining complaint and call category feature data, generating a mediation scheme according to the complaint and call category feature data, performing mediation, and if the mediation is successful, generating a mediation document and sending the mediation document to a case party; according to the method, the complaint form is automatically generated, the compliance state verification is performed, the complaint category feature data is determined through keyword extraction and vector representation, and then the conciliation scheme is generated, so that the whole-process intelligent processing of the litigation conciliation case is realized, and the efficiency and success rate of litigation conciliation are improved.

Description

Technical Field

[0001] This application relates to the fields of judicial informatization and artificial intelligence technologies. Specifically, it relates to an intelligent case processing method, system, and medium for litigation mediation. Background Art

[0002] Litigation mediation helps improve the case closing efficiency of the court and reduce the number of accumulated cases, thereby saving judicial resources. Currently, traditional litigation mediation cases mainly rely on manual processing. Case mediators need to manually collect case-related information, such as pleadings and evidence materials submitted by the parties, and read, analyze, and organize them one by one. This case processing method is inefficient and tests the professional qualities of case mediators. There is a strong subjectivity in identifying the focus of disputes in cases, formulating case mediation plans, and conducting case mediation. At the same time, it is difficult to achieve the optimal allocation of judicial resources. There is an urgent need for an automated intelligent processing method in the full-process management of litigation mediation cases.

[0003] In response to the above problems, effective technical solutions are urgently needed. Summary of the Invention

[0004] The purpose of this application is to provide an intelligent case processing method, system, and medium for litigation mediation, which can automatically generate a pleading form and perform compliance status verification, determine the category feature data of litigation mediation through keyword extraction and vector representation, and then generate a mediation plan, realizing the full-process intelligent processing of litigation mediation cases and improving the efficiency and success rate of litigation mediation.

[0005] In a first aspect, this application provides an intelligent case processing method for litigation mediation, including the following steps: Obtain the litigation data of the pleading, fill the litigation data into the pleading form, and determine the compliance status of the pleading form; If it is compliant, extract keywords according to the pleading form and perform vector conversion processing to obtain a litigation mediation semantic vector; Input the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain litigation mediation category feature data; Generate a mediation plan according to the litigation mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve it to the case parties.

[0006] Optionally, in the intelligent case processing method for litigation mediation described in this application, the step of obtaining the litigation data of the pleading, filling the litigation data into the pleading form, and determining the compliance status of the pleading form includes: Obtain the litigation data of the pleading, including the identity data of the parties and the semantic data of the pleading evidence; Fill the identity data of the parties and the semantic data of the pleading evidence into the pleading form to obtain a pleading form; Perform integrity verification on the complaint form; If the integrity verification fails, activate the first-link outbound call response; If the integrity verification passes, perform logical verification on the complaint form through a preset complaint knowledge graph; If the logical verification fails, activate the first-link outbound call response; If the logical verification passes, determine that the compliance status of the complaint form is compliant.

[0007] Optionally, in the intelligent case processing method for litigation mediation described in this application, if it is compliant, keyword extraction is performed according to the complaint form, and vector conversion processing is performed to obtain a litigation mediation semantic vector, including: If it is compliant, match according to the complaint form through a preset keyword library to obtain a litigation mediation keyword set; Perform vector conversion on the litigation mediation keyword set through a preset language model to obtain a litigation mediation semantic vector.

[0008] Optionally, in the intelligent case processing method for litigation mediation described in this application, inputting the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain litigation mediation category feature data includes: Input the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain initial litigation mediation category feature data and corresponding probability values; Arrange the probability values in descending order, and determine the initial litigation mediation category feature data corresponding to the maximum probability value as the litigation mediation category feature data.

[0009] Optionally, in the intelligent case processing method for litigation mediation described in this application, generating a mediation plan according to the litigation mediation category feature data and conducting mediation. If the mediation is successful, generate a mediation document and serve it to the case parties, including: Query a preset mediation plan database according to the litigation mediation category feature data, combine it with the complaint form to generate multiple initial mediation plans, and extract corresponding mediation plan feature data; Obtain the sufficiency of evidence for the litigation mediation case; Input the evidence sufficiency and mediation plan feature data into a preset mediation plan success rate prediction model for processing to obtain the corresponding mediation success rate, and arrange the mediation success rates of all initial mediation plans in descending order; Determine the initial mediation plan with the highest mediation success rate as the mediation plan and conduct mediation; If the mediation is successful, generate a mediation document and serve it to the case parties.

[0010] Optionally, in the method for intelligent processing of litigation mediation cases described in this application, the step of generating a mediation document and serving it to the parties to the case if the mediation is successful includes: Obtain the contact information of the parties to the case and perform semantic analysis; If the semantic analysis is normal, verify the contact information through a preset household registration data platform; If the verification passes, obtain the service risk assessment data of the parties to the case, process it to obtain the service difficulty level, and determine the case service method according to the service difficulty level.

[0011] In a second aspect, this application provides a system for intelligent processing of litigation mediation cases, which includes: A preprocessing module for litigation mediation case information, used to obtain the litigation data of the complaint and perform preprocessing for integrity and logical verification. If the verification fails, trigger the first-link outbound call response; An intelligent analysis module for litigation mediation cases, used to extract litigation mediation keywords, perform litigation mediation semantic vector conversion, and obtain litigation mediation category feature data after passing the integrity and logical verification; An intelligent auxiliary mediation module for litigation mediation cases, used to generate multiple initial mediation plans according to the litigation mediation category feature data in combination with the complaint form, and evaluate the mediation success rate. If the mediation is successful, perform semantic analysis and verification on the contact information of the parties to the case. If both pass, determine the case service method according to the service difficulty level. If not, activate the manual confirmation response; A mediation document generation module for litigation mediation cases, used to generate a mediation document according to a preset mediation document template; A data transmission module, used to encrypt the data of the intelligent analysis module for litigation mediation cases, the intelligent auxiliary mediation module for litigation mediation cases, and the mediation document generation module for litigation mediation cases and transmit them to the storage space.

[0012] Optionally, in a system for intelligent processing of litigation mediation cases described in this application, the system further includes: a memory and a processor. The memory includes a program for a method for intelligent processing of litigation mediation cases. When the program for the method for intelligent processing of litigation mediation cases is executed by the processor, the following steps are implemented: Obtain the litigation data of the complaint, fill the litigation data into the complaint form, and determine the compliance status of the complaint form; If it is compliant, extract keywords according to the complaint form and perform vector conversion processing to obtain a litigation mediation semantic vector; Input the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain litigation mediation category feature data; Generate a mediation plan based on the above-mentioned lawsuit mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve it to the parties involved in the case.

[0013] Optionally, in a case intelligent processing system for lawsuit mediation described in this application, the obtaining of the lawsuit data of the complaint, filling the lawsuit data into the complaint form, and determining the compliance status of the complaint form include: Obtain the lawsuit data of the complaint, including the identity data of the parties and the semantic data of the evidence in the complaint; Fill the identity data of the parties and the semantic data of the evidence in the complaint into the complaint form to obtain a complaint form; Conduct integrity verification on the complaint form; If the integrity verification fails, activate the first-link outbound call response; If the integrity verification passes, conduct logical verification on the complaint form through a preset complaint knowledge graph; If the logical verification fails, activate the first-link outbound call response; If the logical verification passes, determine that the compliance status of the complaint form is compliant.

[0014] Thirdly, this application also provides a computer-readable storage medium. A program for a case intelligent processing method for lawsuit mediation is stored in the computer-readable storage medium. When the program for the case intelligent processing method for lawsuit mediation is executed by a processor, the steps of a case intelligent processing method for lawsuit mediation as described in any one of the above are implemented.

[0015] As can be seen from the above, a case intelligent processing method, system, and medium for lawsuit mediation provided by this application realize the full-process intelligent processing of lawsuit mediation cases by automatically generating a complaint form and conducting compliance status verification, determining lawsuit mediation category feature data through keyword extraction and vector representation, and then generating a mediation plan, improving the efficiency and success rate of lawsuit mediation.

[0016] Other features and advantages of this application will be described in the subsequent specification. Moreover, some of them will become obvious from the specification, or can be understood by implementing the embodiments of this application. The objectives and other advantages of this application can be achieved and obtained through the structures specifically pointed out in the written specification and the drawings. Description of the Drawings

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the following will briefly introduce the drawings required to be used in the embodiments of this application. It should be understood that the following drawings only show some embodiments of this application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can also be obtained based on these drawings without creative efforts.

[0018] Figure 1 Flowchart of an intelligent case processing method for litigation mediation provided by an embodiment of the present application; Figure 2 Flowchart of obtaining the compliance status of a pleading form in an intelligent case processing method for litigation mediation provided by an embodiment of the present application; Figure 3 Flowchart of obtaining a litigation mediation semantic vector in an intelligent case processing method for litigation mediation provided by an embodiment of the present application; Figure 4 High-level flowchart of the methods of various embodiments of the present application, which can be used in the intelligent case processing method for litigation mediation; Figure 5 System diagram of an intelligent case processing system for litigation mediation provided by an embodiment of the present application. Detailed implementation manners

[0019] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0020] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for descriptive distinction and cannot be understood as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 is a flowchart of an intelligent case processing method for litigation mediation in some embodiments of the present application. This intelligent case processing method for litigation mediation is used in terminal devices, such as computers, mobile phone terminals, etc. This intelligent case processing method for litigation mediation includes the following steps: S11. Obtain the litigation data of the pleading, fill the litigation data into the pleading form, and determine the compliance status of the pleading form; S12. If it is compliant, extract keywords according to the pleading form, and perform vector conversion processing to obtain a litigation mediation semantic vector; S13. Input the mediation semantic vector into a preset mediation classification model for processing to obtain mediation category feature data; S14. Generate a mediation plan based on the mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve it to the case parties.

[0022] It should be noted that after the case parties submit the complaint and evidence materials, the system conducts intelligent recognition according to the demand information of the complaint form, such as automatically recognizing through OCR technology, extracting the corresponding litigation data and filling it into the complaint form, and conducting integrity and logic verification, reducing the workload of data entry and improving data accuracy. To further understand the litigation mediation requirements, keywords are extracted through matching with a preset keyword library to identify the mediation category and obtain mediation category feature data, such as the right of contract rescission and the determination of liability for breach of contract. Based on the determined mediation category feature data, the system generates multiple initial mediation plans in combination with historical litigation mediation cases and predicts the success rate in combination with a preset model. The initial mediation plan with the highest success rate is determined as the mediation plan to improve the success rate of litigation mediation. Finally, litigation mediation is conducted according to the determined mediation plan. If the mediation is successful, a mediation document is automatically generated according to a preset template and served to the case parties to complete the current litigation mediation. If it is not successful, case adjudication is processed.

[0023] Please refer to Figure 2 , Figure 2 is a flowchart for obtaining the compliance status of a complaint form in an intelligent case processing method for litigation mediation in some embodiments of the present application. According to an embodiment of the present invention, the obtaining of litigation data of a complaint, filling the litigation data into the complaint form, and determining the compliance status of the complaint form include: S21. Obtain litigation data of the complaint, including party identity data and semantic data of complaint evidence; S22. Fill the party identity data and semantic data of complaint evidence into the complaint form to obtain the complaint form; S23. Conduct integrity verification on the complaint form; S24. If the integrity verification fails, activate the first-link outbound call response; S25. If the integrity verification passes, conduct logic verification on the complaint form through a preset complaint knowledge graph; S26. If the logic verification fails, activate the first-link outbound call response; S27. If the logic verification passes, determine the compliance status of the complaint form as compliant.

[0024] It should be noted that in order to improve the efficiency of the preliminary processing of litigation mediation cases and reduce the workload of case handlers, after receiving the complaint from the case parties, the OCR technology is used to automatically identify the complaint data including the party identity data and the semantic data of the complaint evidence, and automatically fill in the complaint form. Among them, the party identity data includes name, contact information, plaintiff or defendant, and the semantic data of the complaint evidence includes time, amount, interest, and evidence type. And integrity verification is carried out according to the preset elements of the complaint form. If the elements are incomplete, such as the lack of transfer record evidence materials, the first-link outbound call response is activated, that is, the case parties are called for the first time by phone, WeChat, etc., and the parties are required to complete the missing elements. If the verification passes, further logical verification is carried out according to the preset complaint knowledge graph, such as whether the lending rate exceeds the upper limit. If any logical verification fails, the first-link outbound call response is activated. If all pass, it means that the complaint expression meets the requirements. Among them, the preset complaint knowledge graph is constructed by those skilled in the art through obtaining the entities and relationship edges of a large number of historical case samples.

[0025] Please refer to Figure 3 , Figure 3 is a flowchart for obtaining the litigation mediation semantic vector of an intelligent case processing method for litigation mediation in some embodiments of the present application. According to an embodiment of the present invention, if it is compliant, keyword extraction is performed according to the complaint form, and vector conversion processing is performed to obtain the litigation mediation semantic vector, including: S31. If it is compliant, match according to the complaint form through a preset keyword library to obtain a litigation mediation keyword set; S32. Vectorize the litigation mediation keyword set through a preset language model to obtain the litigation mediation semantic vector.

[0026] It should be noted that litigation mediation cases involve different mediation types, such as liability for breach of contract determination, right of contract rescission. In order to accurately identify the category of litigation mediation, keyword matching extraction is performed according to the generated complaint form. The keywords are matched according to a preset keyword library, and the preset keyword library is preset by those skilled in the art according to historical cases. The obtained litigation mediation keyword set is vectorized through a preset language model to obtain the litigation mediation semantic vector, and the keywords are structurally represented. Among them, the preset language model is trained by those skilled in the art through the litigation mediation keyword sets and the corresponding litigation mediation semantic vectors of a large number of historical case samples using a neural network.

[0027] According to an embodiment of the present invention, inputting the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain litigation mediation category feature data includes: Input the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain initial litigation mediation category feature data and the corresponding probability value; Arrange the probability values in descending order, and determine the initial litigation mediation category feature data corresponding to the maximum probability value as the litigation mediation category feature data.

[0028] It should be noted that in order to accurately identify the litigation mediation category, the structured litigation mediation semantic vector is input into a preset litigation mediation classification model for processing to obtain the initial litigation mediation category feature data and the corresponding probability values. Among them, the preset litigation mediation classification model is trained by those skilled in the art using a neural network with Softmax regression as the output layer based on the litigation mediation semantic vectors of a large number of historical case samples, the corresponding initial litigation mediation category feature data, and the calibrated probability values. Then, by arranging the probability values in descending order, the initial litigation mediation category feature data with the largest probability value is determined as the litigation mediation category feature data. For example, if the probability value of liability for breach of contract determination is 0.92 and the probability value of the right to rescind the contract is 0.81, then the liability for breach of contract determination is determined as the litigation mediation category feature data.

[0029] According to an embodiment of the present invention, generating a mediation plan based on the litigation mediation category feature data and conducting mediation. If the mediation is successful, a mediation document is generated and served to the parties to the case, including: Query the preset mediation plan database according to the litigation mediation category feature data and generate multiple initial mediation plans in combination with the complaint form, and extract the corresponding mediation plan feature data; Obtain the sufficiency of evidence in the litigation mediation case; Input the evidence sufficiency and the mediation plan feature data into a preset mediation plan success rate prediction model for processing to obtain the corresponding mediation success rate, and arrange the mediation success rates of all initial mediation plans in descending order; Determine the initial mediation plan with the highest mediation success rate as the mediation plan and conduct mediation; If the mediation is successful, generate a mediation document and serve it to the parties to the case.

[0030] It should be noted that after determining the characteristic data of the litigation - mediation category, the mediation process is entered. In order to provide mediation guidance to the mediator and improve the success rate of mediation, multiple initial mediation plans are generated by querying the preset mediation plan database based on the characteristic data of the litigation - mediation category and combining with the specific case situation extracted from the complaint form. Among them, the preset mediation plan database is constructed by technicians in the field according to the intelligent analysis module of litigation - mediation cases, extracting specific elements such as the amount involved in the case, the performance time, and the details of the dispute focus from the complaint forms of historical cases, and combining with historical litigation - mediation cases of the same category and their corresponding mediation plans. In order to determine the most effective plan, characteristic data of the mediation plan is extracted, such as the compensation amount and the performance method, and then combined with the sufficiency degree of evidence obtained in this case. The sufficiency degree of evidence is rated by technicians in the field according to the complaint form, divided into level 1, level 2, level 3, level 4, and level 5, and level 5 evidence is the most sufficient. Finally, it is processed according to the preset prediction model of the success rate of the mediation plan to obtain the corresponding success rate of mediation. The preset prediction model of the success rate of the mediation plan is obtained by training a neural network using the sufficiency degree of evidence, the characteristic data of the mediation plan, and the corresponding success rate of a large number of historical samples, and the success rates of mediation are arranged in descending order. The initial mediation plan with the highest success rate of mediation is determined as the mediation plan. Then, the mediator conducts case mediation according to the mediation plan. If the mediation is successful, the system automatically generates a mediation document according to the mediation plan and serves it to the case parties. If the mediation fails, the adjudication process is carried out.

[0031] According to an embodiment of the present invention, if the mediation is successful, generating a mediation document and serving it to the case parties includes: Obtaining the contact information of the case parties and performing semantic analysis; If the semantic analysis is normal, verifying the contact information through a preset household registration data platform; If the verification passes, obtaining the service risk assessment data of the case parties, processing it to obtain the service difficulty level, and determining the case service method according to the service difficulty level.

[0032] It should be noted that in order to serve the mediation document to the parties in a timely and successful manner, first, the contact information of the parties is obtained, such as the contact phone number, contact address, and contact email, and semantic analysis is performed. Semantic analysis refers to determining whether the contact information of the parties meets the format requirements, such as whether the number of digits of the contact phone number is accurate and whether the email format is correct. If the analysis is normal, it is further verified through a preset household registration data platform to check whether they are consistent. If they are inconsistent, manual confirmation is required. If the verification passes, then the service risk is analyzed to evaluate the service difficulty level, and different service methods are determined according to different service difficulty levels, such as on - site service, postal service, or email service.

[0033] Please refer to Figure 4 ,Figure 4 This is a high-level flowchart of the methods of various embodiments of the present application, and these methods can be used for the intelligent processing method of litigation mediation cases. According to the embodiments of the present invention, for example, mediation is carried out according to the mediation plan. If the mediation is successful, a mediation document is automatically generated and the service of the mediation document is executed. If the mediation fails, the adjudication procedure is executed for processing.

[0034] It is worth mentioning that according to the embodiments of the present invention, if the verification passes, the service risk evaluation data of the case parties is obtained and processed to obtain the service difficulty level, and the case service method is determined according to the service difficulty level, including: Obtain the service risk evaluation data of the case parties, including the historical rejection rate, service address location data, and preset service method characteristic data; Input the historical rejection rate, service address location data, and preset service method characteristic data into a preset case service risk evaluation model for processing to obtain case service risk parameters; Compare the case service risk parameters with a preset case service risk control threshold to obtain the case service difficulty level, including low risk, medium risk, or high risk.

[0035] It should be noted that the specific method for evaluating the case service difficulty level is as follows: First, obtain the historical rejection rate, address matching data, service address location data, and preset service method characteristic data of the case parties. Among them, the service method refers to the ordinary service method, electronic service method, or service by leaving at the place. The preset service method characteristic data is represented by a numerical value and can be dynamically adjusted by those skilled in the art according to specific circumstances. For example, the ordinary service method is represented by the numerical value 1, and the electronic service method is represented by the numerical value 2. Through a preset case service risk evaluation model for processing, case service risk parameters are obtained. Among them, the preset case service risk evaluation model is trained by using a neural network with the historical rejection rate, service address location data, and preset service method characteristic data of a large number of example case samples and the corresponding case service risk parameters. Then, the case service risk parameters are compared with the preset case service risk control threshold. Among them, the preset case service risk control threshold includes a first preset case service risk control threshold and a second preset case service risk control threshold, and the first preset case service risk control threshold is less than the second preset case service risk control threshold. If the case service risk parameter is less than or equal to the first preset case service risk control threshold, it is a low risk. If it is greater than the first preset case service risk control threshold and less than or equal to the second preset case service risk control threshold, it is a medium risk. If it is greater than the second preset case service risk control threshold, it is a high risk. For example, door-to-door service is selected for high risk, postal service is selected for medium risk, and email service is selected for low risk.

[0036] It is worth mentioning that according to the embodiments of the present invention, it further includes: Obtain the service instruction of the mediation document; Obtain the confirmation of service instructions for the case parties at a preset time point according to the service instructions; If the service is confirmed, record the case service time; If the service is not confirmed, activate the outbound call confirmation response.

[0037] It should be noted that after the case is served, obtain the confirmation feedback of the case parties in a timely manner to achieve the closed-loop of litigation mediation for the case. First, obtain the service instructions of the mediation document, including the receipt of the shipping logistics, the email has been sent or the on-site service. Then, at a predetermined time point, such as 24 hours after the service instructions, obtain the confirmation of service instructions from the case parties. If the service is confirmed, record the case service time, complete the closed-loop of the litigation mediation case, and conduct the case file closing process. If the service is not confirmed, activate the outbound call confirmation response to promptly urge the case parties to confirm.

[0038] Please refer to Figure 5 , Figure 5 which is the system diagram of an intelligent case processing system for litigation mediation in some embodiments of the present application.

[0039] In a second aspect, the present invention also discloses an intelligent case processing system 5 for litigation mediation, including: A preprocessing module 51 for litigation and mediation case information, which is used to obtain the litigation data of the complaint and perform preprocessing of integrity and logical verification. If the verification fails, trigger the first-link outbound call response; An intelligent analysis module 52 for litigation and mediation cases, which is used to extract litigation and mediation keywords, convert litigation and mediation semantic vectors, and obtain litigation and mediation category feature data after passing the integrity and logical verification; An intelligent auxiliary mediation module 53 for litigation and mediation cases, which is used to generate multiple initial mediation plans according to the litigation and mediation category feature data in combination with the complaint form, and evaluate the mediation success rate. If the mediation is successful, perform semantic analysis and verification on the contact information of the case parties. If all pass, determine the case service method according to the service difficulty level. If not, activate the manual confirmation response; A mediation document generation module 54 for litigation and mediation cases, which is used to generate mediation documents according to a preset mediation document template; A data transmission module 55, which is used to encrypt the data of the intelligent analysis module for litigation and mediation cases, the intelligent auxiliary mediation module for litigation and mediation cases, and the mediation document generation module for litigation and mediation cases and transmit them to the storage space.

[0040] It should be noted that the preprocessing module for lawsuit mediation case information corresponds to the steps of "obtaining the litigation data of the complaint and determining the compliance status", and is responsible for data acquisition and compliance verification. The intelligent analysis module for lawsuit mediation cases corresponds to the steps of "keyword extraction, vector conversion, and obtaining lawsuit mediation category feature data", and is executed based on the verification result of the preprocessing module. If the preprocessing module passes the verification, it enters the analysis process. If the verification fails (integrity or logical verification fails), it triggers the first-link outbound call response, pauses the execution of the subsequent analysis module, and resumes after supplementation and improvement. The intelligent assisted mediation module for lawsuit mediation cases corresponds to the steps of "generating a mediation plan and conducting mediation", and the mediation document generation module for lawsuit mediation cases corresponds to the step of "generating a mediation document". The data transmission module runs through the entire process and is responsible for encrypting and transmitting the data of each module. For example, it encrypts and transmits litigation data including party identity data and semantic data of complaint evidence to the storage space.

[0041] According to an example of the present invention, the intelligent case processing system for litigation mediation further includes a memory and a processor. The memory includes a program for the intelligent case processing method for litigation mediation. When the program for the intelligent case processing method for litigation mediation is executed by the processor, the following steps are implemented: Obtain the litigation data of the complaint, fill the litigation data into the complaint form, and determine the compliance status of the complaint form; If it is compliant, extract keywords according to the complaint form and perform vector conversion processing to obtain a lawsuit mediation semantic vector; Input the lawsuit mediation semantic vector into a preset lawsuit mediation classification model for processing to obtain lawsuit mediation category feature data; Generate a mediation plan according to the lawsuit mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve it to the case parties.

[0042] It should be noted that after the case parties submit the complaint and evidence materials, the system performs intelligent recognition according to the requirement information of the complaint form, such as automatic recognition by OCR technology, extracts the corresponding litigation data and fills it into the complaint form, and conducts integrity and logic verification, reducing the data entry workload and improving data accuracy. To further understand the litigation mediation requirements, keywords are extracted by matching with a preset keyword library, and the lawsuit mediation category is identified to obtain lawsuit mediation category feature data, such as the right of contract rescission and liability for breach of contract determination. According to the determined lawsuit mediation category feature data, the system generates multiple initial mediation plans in combination with historical litigation mediation cases and predicts the success rate in combination with a preset model. The initial mediation plan with the highest success rate is determined as the mediation plan to improve the success rate of litigation mediation. Finally, litigation mediation is conducted according to the determined mediation plan. If the mediation is successful, a mediation document is automatically generated according to a preset template and served to the case parties to complete this litigation mediation. If it is not successful, case adjudication processing is carried out.

[0043] According to an embodiment of the present invention, obtaining the litigation data of the complaint, filling the litigation data into the complaint form, and determining the compliance status of the complaint form includes: Obtaining the litigation data of the complaint, including party identity data and semantic data of complaint evidence; Filling the party identity data and the semantic data of complaint evidence into the complaint form to obtain a complaint form; Performing integrity verification on the complaint form; If the integrity verification fails, activate the first-link outbound call response; If the integrity verification passes, perform logical verification on the complaint form through a preset complaint knowledge graph; If the logical verification fails, activate the first-link outbound call response; If the logical verification passes, determine that the compliance status of the complaint form is compliant.

[0044] It should be noted that in order to improve the efficiency of the preliminary processing of litigation mediation cases and reduce the workload of case handlers, after receiving the complaint from the case parties, the OCR technology is used to automatically identify the complaint data including the party identity data and the semantic data of complaint evidence, and automatically fill in the complaint form. Among them, the party identity data includes name, contact information, plaintiff or defendant, and the semantic data of complaint evidence includes time, amount, interest, and evidence type. And perform integrity verification according to the preset elements of the complaint form. If the elements are incomplete, such as the lack of transfer record evidence materials, activate the first-link outbound call response, that is, make the first call by phone, WeChat, etc. to the case parties, and require them to supplement the missing elements. If the verification passes, further perform logical verification according to the preset complaint knowledge graph, such as whether the lending rate exceeds the upper limit. If any one of the logical verifications fails, activate the first-link outbound call response. If all pass, it means that the complaint expression meets the requirements. Among them, the preset complaint knowledge graph is constructed by those skilled in the art by obtaining the entities and relationship edges of a large number of historical case samples.

[0045] According to an embodiment of the present invention, if it is compliant, keyword extraction is performed according to the complaint form, and vector conversion processing is performed to obtain a litigation mediation semantic vector, including: If it is compliant, match according to the complaint form through a preset keyword library to obtain a litigation mediation keyword set; Perform vector conversion on the litigation mediation keyword set through a preset language model to obtain a litigation mediation semantic vector.

[0046] It should be noted that litigation mediation cases involve different mediation types, such as liability for breach of contract determination and the right of contract rescission. In order to accurately identify the categories of litigation mediation, keyword matching extraction is performed according to the generated complaint form. The keywords are matched based on a preset keyword library, which is preset by those skilled in the art based on historical cases. The obtained set of litigation mediation keywords is vectorized through a preset language model to obtain litigation mediation semantic vectors, and the keywords are structurally represented. Among them, the preset language model is trained by those skilled in the art using a neural network with a set of litigation mediation keywords and corresponding litigation mediation semantic vectors from a large number of historical case samples.

[0047] According to an embodiment of the present invention, inputting the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain litigation mediation category feature data includes: Inputting the litigation mediation semantic vector into a preset litigation mediation classification model for processing to obtain initial litigation mediation category feature data and corresponding probability values; Sorting the probability values in descending order, and determining the initial litigation mediation category feature data corresponding to the maximum probability value as the litigation mediation category feature data.

[0048] It should be noted that in order to accurately identify the litigation mediation category, the structurally represented litigation mediation semantic vector is input into a preset litigation mediation classification model for processing to obtain initial litigation mediation category feature data and corresponding probability values. Among them, the preset litigation mediation classification model is trained by those skilled in the art using a neural network with Softmax regression as the output layer based on the litigation mediation semantic vectors, corresponding initial litigation mediation category feature data, and calibrated probability values from a large number of historical case samples. Then, by sorting the probability values in descending order, the initial litigation mediation category feature data with the largest probability value is determined as the litigation mediation category feature data. For example, if the probability value of liability for breach of contract determination is 0.92 and the probability value of the right of contract rescission is 0.81, then liability for breach of contract determination is determined as the litigation mediation category feature data.

[0049] According to an embodiment of the present invention, generating a mediation plan based on the litigation mediation category feature data and conducting mediation. If the mediation is successful, generating a mediation document and serving it to the case parties includes: Querying a preset mediation plan database based on the litigation mediation category feature data and combining it with the complaint form to generate multiple initial mediation plans, and extracting corresponding mediation plan feature data; Obtaining the sufficiency of evidence in the litigation mediation case; Inputting the evidence sufficiency and mediation plan feature data into a preset mediation plan success rate prediction model for processing to obtain the corresponding mediation success rate, and sorting the mediation success rates of all initial mediation plans in descending order; Determining the initial mediation plan with the highest mediation success rate as the mediation plan and conducting mediation; If the mediation is successful, a mediation document shall be generated and served on the parties to the case.

[0050] It should be noted that after determining the category feature data of litigation mediation, the mediation process is entered. In order to provide mediation guidance to the mediator and improve the success rate of mediation, multiple initial mediation plans are generated by querying the preset mediation plan database based on the category feature data of litigation mediation and combining with the specific case situation extracted from the complaint form. Among them, the preset mediation plan database is constructed by those skilled in the art based on the intelligent analysis module of litigation mediation cases by extracting specific elements such as the amount involved in the case, the performance time, and the details of the dispute focus from the complaint forms of historical cases, and combining historical litigation mediation cases of the same category and the corresponding mediation plans. In order to determine the most effective plan, the feature data of the mediation plan is extracted from the initial mediation plan, such as the compensation amount and the performance method, and then combined with the sufficiency degree of evidence obtained in this case. Among them, the sufficiency degree of evidence is rated by those skilled in the art according to the complaint form, divided into level 1, level 2, level 3, level 4, and level 5, and level 5 evidence is the most sufficient. Finally, it is processed according to the preset success rate prediction model of the mediation plan to obtain the corresponding mediation success rate. Among them, the preset success rate prediction model of the mediation plan is obtained by training a neural network using the sufficiency degree of evidence, the feature data of the mediation plan, and the corresponding mediation success rate of a large number of historical samples, and the mediation success rates are sorted in descending order. The initial mediation plan with the highest mediation success rate is determined as the mediation plan. Then, the mediator conducts the case mediation according to the mediation plan. If the mediation is successful, the system automatically generates a mediation document according to the mediation plan and serves it on the parties to the case. If the mediation fails, the adjudication process is carried out.

[0051] According to the embodiments of the present invention, the step of if the mediation is successful, generating a mediation document and serving it on the parties to the case includes: Obtain the contact information of the parties to the case and conduct semantic analysis; If the semantic analysis is normal, verify the contact information through the preset household registration data platform; If the verification is passed, obtain the service risk evaluation data of the parties to the case, process it to obtain the service difficulty level, and determine the case service method according to the service difficulty level.

[0052] It should be noted that, in order to timely and successfully serve the mediation documents to the parties, the contact information of the parties, such as contact phone number, contact address, and contact email, is first obtained and semantic analysis is carried out. Semantic analysis refers to determining whether the contact information of the parties meets the format requirements, such as whether the number of digits of the contact phone number is accurate and whether the email format is correct. If the analysis is normal, further verification is carried out through a preset household registration data platform to check whether they are consistent. If they are inconsistent, manual confirmation is required. If the verification passes, the service risk is further analyzed to evaluate the service difficulty level, and different service methods are determined according to different service difficulty levels, such as on-site service, mail service, or email service.

[0053] According to an embodiment of the present invention, for example, mediation is carried out according to the mediation plan. If the mediation is successful, the mediation document is automatically generated and the service of the mediation document is executed. If the mediation fails, the adjudication procedure is executed for handling.

[0054] It is worth mentioning that, according to an embodiment of the present invention, if the verification passes, the service risk evaluation data of the case parties is obtained and processed to obtain the service difficulty level, and the case service method is determined according to the service difficulty level, including: Obtaining the service risk evaluation data of the case parties, including the historical rejection rate, service address location data, and preset service method feature data; Inputting the historical rejection rate, service address location data, and preset service method feature data into a preset case service risk evaluation model for processing to obtain the case service risk parameter; Comparing the case service risk parameter with a preset case service risk control threshold to obtain the case service difficulty level, including low risk, medium risk, or high risk.

[0055] It should be noted that the specific method for evaluating the difficulty level of case service is as follows: First, obtain the historical rejection rate, address matching data, service address location data, and preset service method characteristic data of the case parties. Among them, the service method refers to the ordinary service method, electronic service method, or service by leaving a copy. The preset service method characteristic data is represented by a numerical value and can be dynamically adjusted by those skilled in the art according to specific circumstances. For example, the ordinary service method is represented by the numerical value 1, and the electronic service method is represented by the numerical value 2. Process through the preset case service risk evaluation model to obtain the case service risk parameter. Among them, the preset case service risk evaluation model is trained by using a neural network with the historical rejection rate, service address location data, preset service method characteristic data, and the corresponding case service risk parameters of a large number of example case samples. Then, compare the case service risk parameter with the preset case service risk control threshold. Among them, the preset case service risk control threshold includes the first preset case service risk control threshold and the second preset case service risk control threshold, and the first preset case service risk control threshold is less than the second preset case service risk control threshold. If the case service risk parameter is less than or equal to the first preset case service risk control threshold, it is a low risk. If it is greater than the first preset case service risk control threshold and less than or equal to the second preset case service risk control threshold, it is a medium risk. If it is greater than the second preset case service risk control threshold, it is a high risk. For example, door-to-door service is selected for high risk, postal service is selected for medium risk, and email service is selected for low risk.

[0056] It is worth mentioning that according to the embodiments of the present invention, it further includes: Obtain the service instruction of the mediation document; Obtain the confirmed service instruction of the case parties at a preset time point according to the service instruction; If the service is confirmed, record the case service time; If the service is not confirmed, activate the outbound call confirmation response.

[0057] It should be noted that after the case is served, obtain the confirmed service feedback of the case parties in a timely manner to achieve the closed-loop of case litigation mediation. First, obtain the service instruction of the mediation document, including the receipt of the sending logistics, the email has been sent, or door-to-door service. Then, at a predetermined time point, such as 24 hours after the service instruction, obtain the confirmed service instruction of the case parties. If the service is confirmed, record the case service time, complete the closed-loop of the litigation mediation case, and perform the file closing process. If the service is not confirmed, activate the outbound call confirmation response to promptly urge the case parties to confirm.

[0058] The third aspect of the present invention provides a readable storage medium, in which a program for an intelligent processing method of litigation mediation cases is stored. When the program for an intelligent processing method of litigation mediation cases is executed by a processor, the steps of an intelligent processing method of litigation mediation cases as described in any one of the above are realized.

[0059] An intelligent processing method, system and medium for lawsuit mediation cases disclosed by the present invention realize the full-process intelligent processing of lawsuit mediation cases and improve the efficiency and success rate of lawsuit mediation by automatically generating a complaint form and performing compliance status verification, determining lawsuit mediation category feature data through keyword extraction and vector representation, and then generating a mediation plan.

[0060] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed with each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0061] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0062] In addition, in each embodiment of the present invention, each functional unit can be all integrated in a processing unit, or each unit can be separately used as a unit, or two or more units can be integrated in a unit; the above-mentioned integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0063] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media that can store program codes such as removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs.

[0064] Alternatively, if the above integrated units of the present invention are implemented in the form of software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as removable storage devices, ROM, RAM, magnetic disks, or optical discs.

Claims

1. An intelligent processing method for cases of litigation mediation, characterized in that, It includes the following steps: Obtain the litigation data of the complaint, fill the litigation data into the complaint form, and determine the compliance status of the complaint form; If it is compliant, extract keywords according to the complaint form, and perform vector conversion processing to obtain the semantic vector for lawsuit mediation; Input the semantic vector for lawsuit mediation into a preset lawsuit mediation classification model for processing to obtain lawsuit mediation category feature data; Generate a mediation plan according to the lawsuit mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve it to the case parties.

2. The intelligent processing method for lawsuit mediation cases according to claim 1, wherein The step of obtaining the litigation data of the complaint, filling the litigation data into the complaint form, and determining the compliance status of the complaint form includes: Obtain the litigation data of the complaint, including the identity data of the parties and the semantic data of the evidence in the complaint; Fill the identity data of the parties and the semantic data of the evidence in the complaint into the complaint form to obtain the complaint form; Conduct integrity verification on the complaint form; If the integrity verification fails, activate the first-link outbound call response; If the integrity verification passes, conduct logical verification on the complaint form through a preset complaint knowledge graph; If the logical verification fails, activate the first-link outbound call response; If the logical verification passes, determine that the compliance status of the complaint form is compliant.

3. The intelligent processing method for lawsuit mediation cases according to claim 2, wherein The step of if it is compliant, extract keywords according to the complaint form, and perform vector conversion processing to obtain the semantic vector for lawsuit mediation includes: If it is compliant, match according to the complaint form through a preset keyword library to obtain a set of lawsuit mediation keywords; Perform vector conversion on the set of lawsuit mediation keywords through a preset language model to obtain the semantic vector for lawsuit mediation.

4. The intelligent processing method for lawsuit mediation cases according to claim 3, wherein The step of inputting the semantic vector for lawsuit mediation into a preset lawsuit mediation classification model for processing to obtain lawsuit mediation category feature data includes: Input the semantic vector for lawsuit mediation into a preset lawsuit mediation classification model for processing to obtain initial lawsuit mediation category feature data and corresponding probability values; Arrange the probability values in descending order, and determine the initial lawsuit mediation category feature data corresponding to the maximum probability value as the lawsuit mediation category feature data.

5. The intelligent processing method for cases of litigation mediation according to claim 4, characterized in that, The step of generating a mediation plan according to the lawsuit mediation category feature data and conducting mediation. If the mediation is successful, generate a mediation document and serve it to the case parties includes: Query a preset mediation plan database according to the lawsuit mediation category feature data, combine it with the complaint form to generate multiple initial mediation plans, and extract the corresponding mediation plan feature data; Obtain the sufficiency of evidence for the lawsuit mediation case; Input the sufficiency of evidence and the mediation plan feature data into a preset mediation plan success rate prediction model for processing to obtain the corresponding mediation success rate, and arrange the mediation success rates of all initial mediation plans in descending order; Determine the initial mediation plan with the highest mediation success rate as the mediation plan and conduct mediation; If the mediation is successful, generate a mediation document and serve it to the case parties.

6. The intelligent processing method for litigation mediation cases according to claim 5, characterized in that, The step of if the mediation is successful, generate a mediation document and serve it to the case parties includes: Obtain the contact information of the case parties and conduct semantic analysis; If the semantic analysis is normal, verify the contact information through a preset household registration data platform; If the verification passes, obtain the service risk assessment data of the case parties, process it to obtain the service difficulty level, and determine the case service method according to the service difficulty level.

7. An intelligent case processing system for litigation mediation, the intelligent case processing system for litigation mediation implements the intelligent case processing method for litigation mediation described in claims 1-6, characterized in that, Including: A preprocessing module for lawsuit and mediation case information, used to obtain the lawsuit data of the complaint and perform preprocessing on the integrity and logical verification. If the verification fails, trigger the first-link outbound call response; An intelligent analysis module for lawsuit and mediation cases, used to extract lawsuit and mediation keywords, perform lawsuit and mediation semantic vector conversion, and obtain lawsuit and mediation category feature data after the integrity and logical verification pass; An intelligent auxiliary mediation module for lawsuit and mediation cases, used to generate multiple initial mediation plans based on the lawsuit and mediation category feature data in combination with the complaint form, and evaluate the mediation success rate. If the mediation is successful, perform semantic analysis and verification on the contact information of the case parties. If all pass, determine the case service method according to the service difficulty level. If not, activate the manual confirmation response; A mediation document generation module for lawsuit and mediation cases, used to generate mediation documents according to a preset mediation document template; A data transmission module, used to encrypt and transmit the data of the intelligent analysis module for lawsuit and mediation cases, the intelligent auxiliary mediation module for lawsuit and mediation cases, and the mediation document generation module for lawsuit and mediation cases to the storage space.

8. An intelligent case processing system for litigation mediation, characterized in that, It also includes a memory and a processor. The memory stores the program of the intelligent processing method for lawsuit mediation cases. When the program of the intelligent processing method for lawsuit mediation cases is executed by the processor, the following steps are implemented: Obtain the lawsuit data of the complaint, fill the lawsuit data into the complaint form, and determine the compliance status of the complaint form; If it is compliant, extract keywords according to the complaint form and perform vector conversion processing to obtain the lawsuit and mediation semantic vector; Input the lawsuit and mediation semantic vector into a preset lawsuit and mediation classification model for processing to obtain lawsuit and mediation category feature data; Generate a mediation plan according to the lawsuit and mediation category feature data and conduct mediation. If the mediation is successful, generate a mediation document and serve the case parties.

9. The intelligent case processing system for litigation mediation according to claim 8, characterized in that, The step of obtaining the lawsuit data of the complaint, filling the lawsuit data into the complaint form, and determining the compliance status of the complaint form includes: Obtain the lawsuit data of the complaint, including the party identity data and the semantic data of the evidence in the complaint; Fill the party identity data and the semantic data of the evidence in the complaint into the complaint form to obtain the complaint form; Perform integrity verification on the complaint form; If the integrity verification fails, activate the first-link outbound call response; If the integrity verification passes, perform logical verification on the complaint form through a preset complaint knowledge graph; If the logical verification fails, activate the first-link outbound call response; If the logical verification passes, determine that the compliance status of the complaint form is compliant.

10. A computer-readable storage medium, characterized in that, The program of the intelligent processing method for lawsuit mediation cases is stored in the computer-readable storage medium. When the program of the intelligent processing method for lawsuit mediation cases is executed by the processor, the steps of an intelligent processing method for lawsuit mediation cases as described in any one of claims 1 to 6 are implemented.

Images