Judicial case quality evaluation-based trial auxiliary data management system and method

Abstract

This invention discloses a trial auxiliary data management system and method based on judicial case quality evaluation, belonging to the field of data processing technology. By decomposing and reorganizing core and auxiliary cases, a cache of related cases is constructed. This allows for further mining of related cases from both core and auxiliary case perspectives, outputting the most similar related cases to the submitted cases to be evaluated. Simultaneously, using the optimal solution as a reference, cases requiring random checks can be quickly identified, significantly reducing errors and inaccuracies caused by manual operation, improving trial quality and efficiency. Furthermore, the optimal solution found can provide positive or negative preventative references for trial retrospective analysis.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a trial auxiliary data management system and method based on judicial case quality evaluation. Background Technology

[0002] At present, research on the outsourcing of judicial support services is still relatively rudimentary. Due to the professional and special nature of judicial work, there is currently no blueprint for courts nationwide to refer to. Many problems exist in the exploration process. For example, the multi-layered principal-agent relationship in the outsourcing of judicial support services contracts often leads to overlapping handling of case affairs. This results in a chaotic triple relationship of "processes and processes, personnel and personnel, and personnel and processes" in the judicial support services process, which is not conducive to improving the quality and efficiency of trials and makes it impossible to effectively prevent risks in judicial cases. Summary of the Invention

[0003] The purpose of this invention is to provide a trial auxiliary data management system and method based on judicial case quality evaluation, so as to solve the problems mentioned in the background art.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0005] The trial support data management system based on judicial case quality evaluation includes: a data acquisition module, a data processing module, a case screening module, a storage and display module, and a case database module.

[0006] The data acquisition module is used to collect submitted cases and to decompose submitted cases.

[0007] The data processing module is used to calculate case label values ​​and to correct case label values.

[0008] The case filtering module is used to judge the tag values ​​of submitted cases; to obtain auxiliary examples of submitted cases, to establish a case filtering model, and to output the optimal filtering solution.

[0009] The storage and display module is used to obtain the optimal solution. If the optimal solution is a successful case, the submitted case is directly stored in the case database. Otherwise, the submitted case is sent to the case display front end for random inspection and display.

[0010] The case database module is used to store cases;

[0011] The output of the data acquisition module is connected to the input of the data processing module, the output of the data processing module is connected to the input of the case filtering module, the output of the case filtering module is connected to the input of the storage and display module, and the output of the storage and display module is connected to the input of the case database module.

[0012] Furthermore, the data acquisition module also includes an acquisition unit and a decomposition unit; the output terminal of the acquisition unit is connected to the input terminal of the decomposition unit.

[0013] The collection unit is used to randomly collect submitted cases;

[0014] The decomposition unit is used to extract core cases from any submitted case according to the process of case filing, mediation, service of process, preservation, execution, and case file scanning and entry; it is used to decompose submitted cases with different levels of principal-agent relationships into core cases and auxiliary cases, decomposing them into K core cases and W auxiliary cases, where K and W are constants; and it is used to combine the extracted submitted cases according to the correspondence between core cases and auxiliary cases, wherein the combination method is that one core case corresponds to at least one auxiliary case.

[0015] Furthermore, the data processing module also includes a tag value calculation unit and a tag value correction unit; the output of the tag value calculation unit is connected to the input of the tag value correction unit.

[0016] The tag value calculation unit is used to extract feature information of core cases, retrieve related cases from the case database according to the similarity matching algorithm model, and construct a case association cache library; it is used to extract all cases in the case association cache library, mine the core cases of each case, and deduplicate and integrate the mined core cases; it is used to encode the deduplicated and integrated core cases to generate an encoding set, which is denoted as {I1, I2, ..., I...} n}, where each code corresponds to a core case; used to generate a separate set of core cases for cases based on the code set, the separate set containing the submitted case set and each associated case set; used to denote a separate set as a set method P, and to calculate the tag value of any case. Where S represents the label value of any case, a represents any code in a single set of cases, and L a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation.

[0017] The tag value correction unit is used to obtain the code of each core case in the submitted cases, calculate the weight of any core case code in the case association database, and if the weight of a core case in the submitted cases is less than the weight threshold, remove the core case from the code and re-encode the core case; it is also used to iterate the case association cache library and calculate the correction value of the submitted case tag value. , where X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1 Let M represent the label value of the commit case in the (q-1)th iteration, where q represents the iteration number; the actual label value used to calculate the commit case in the qth iteration is M=M q +X q .

[0018] Furthermore, the case filtering module also includes a tag value judgment unit and a case filtering unit; the output of the tag value judgment unit is connected to the input of the case filtering unit.

[0019] The tag value judgment unit is used to calculate the average tag value of the associated cases. When the tag value of the submitted case is less than the average value, the tag value of the submitted case is corrected. When the tag value of the submitted case is greater than or equal to the average value, the auxiliary case of the submitted case is obtained.

[0020] The case selection unit is used to select a core case from the submitted cases as a search dimension and an auxiliary case from the corresponding auxiliary case as a search source, thus obtaining a total of K dimensions and W search sources. It is also used to preset the number of searches and, using a randomized trial approach, calculate the efficiency of any search source across different cases. Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,j G represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases. i，jThis represents the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among the related cases, where L is a random number in the interval [-1, 1]. The judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension. It is used to calculate the energy efficiency value of the auxiliary cases included in each core case in the submitted case in each dimension of the related cases, and to sum the calculated energy efficiency values ​​to obtain the comprehensive energy efficiency value of the extracted case relative to the related cases. It is used to filter the related cases where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches its maximum value in the number of searches, and to output the optimal solution when the maximum value is found.

[0021] A method for managing trial support data based on judicial case quality evaluation, comprising the following steps:

[0022] Step S100: The data acquisition module obtains the submitted cases and decomposes them into K core cases and W auxiliary cases, where each core case corresponds to at least one auxiliary case; where K and W are constants.

[0023] Step S200: Based on the core cases, retrieve the case database and establish a case association model; encode the core cases and calculate the case tag values ​​based on the encoded data;

[0024] Step S300: Calculate the average value of the label values ​​of the associated cases. When the label value of a submitted case is less than the average value, correct the label value of the submitted case and return to step S200.

[0025] Step S400: When the label value of the submitted case is greater than or equal to the average value, obtain auxiliary cases of the submitted case, establish a case screening model, and output the optimal screening solution;

[0026] Step S500: If the optimal solution is a successful case, the submitted case is directly stored in the case database; otherwise, the submitted case is sent to the case display front-end for random inspection and display.

[0027] Furthermore, the specific implementation process of decomposing the submitted case into K core cases and W auxiliary cases in step S100 includes:

[0028] Step S101: Based on the process of case filing, mediation, service of process, preservation, enforcement, and case file scanning and entry, the core cases of the submitted cases are extracted;

[0029] Step S102: Decompose the submitted cases with different levels of principal-agent relationships into core cases and auxiliary cases, decomposing them into K core cases and W auxiliary cases, where K and W are constants; according to the correspondence between core cases and auxiliary cases, combine the submitted cases after stripping them, wherein the combination method is that one core case corresponds to at least one auxiliary case.

[0030] Furthermore, the specific implementation process of calculating the case label value in step S200 includes:

[0031] Step S201: Extract the feature information of core cases; based on the similarity matching algorithm model, retrieve related cases from the case database to construct a case association cache library; extract all cases from the case association cache library, perform core case mining for each case, and deduplicate and integrate the mined core cases; encode the deduplicated and integrated core cases to generate an encoding set, denoted as {I1, I2, ..., I...} n Each code corresponds to a core case.

[0032] Step S202: Based on the encoding set, generate a separate set of core cases for each case. This separate set includes the submitted case set and each associated case set. Denote each separate set as a set type P. Calculate the tag value for any case using the following formula:

[0033]

[0034] Where S represents the label value of any case, a represents any code in a single set of cases, and L a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation.

[0035] According to the above method, assessing the quality of a case involves evaluating every operational step and handling of the case. This process often requires human intervention and comprehensive evaluation based on past cases, which can lead to errors in judgment. Furthermore, it wastes considerable time searching for similar cases for comparison, significantly reducing trial efficiency. The above method first separates the core and supporting cases, focusing on all core cases within the case to be evaluated. Through mining these core cases, similar cases are selected from a vast database. By comparing with successful cases, a tag value is calculated for the case to be submitted for evaluation. A higher tag value indicates a lower probability of error and greater compliance with regulations.

[0036] Furthermore, the specific implementation process of correcting the label values ​​of the submitted cases in step S300 includes:

[0037] Step S301: Obtain the code of each core case in the submitted cases. In the case association database, calculate the weight of any core case code. If the weight of a core case in the submitted cases is less than the weight threshold, remove the core case from the code and re-encode the core case. Return to step S200 and iterate through the case association cache until the tag value of the submitted cases is greater than or equal to the average value. The iteration stops then.

[0038] Step S302: Calculate the correction value of the tag value of the submitted case. The specific calculation formula is as follows:

[0039]

[0040] Among them, X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1 This represents the label value of the submitted case in the (q-1)th iteration, where q represents the iteration number;

[0041] Step S303: Calculate the actual label value of the submitted case in the q-th iteration: M = M q +X q ;

[0042] The purpose of adding corrections to the cases according to the above method is to improve the accuracy of related cases, narrow down the scope, and strive for excellence through continuous iteration.

[0043] Furthermore, the specific implementation process of outputting and filtering the optimal solution in step S400 includes:

[0044] Step S401: Take a core case of the submitted case as a search dimension, and take an auxiliary case of the auxiliary case corresponding to the core case as a search source, then a total of K dimensions and W search sources are obtained;

[0045] Step S402: Preset the number of searches, and use the idea of ​​randomized experimentation to calculate the efficiency value of any search source in different cases. The specific calculation formula is as follows:

[0046]

[0047] Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,jG represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases. i，j This represents the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among related cases, where L is a random number in the interval [-1, 1]; the judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension.

[0048] Based on the above method, under the condition that the value is greater than or equal to the average value of the label, it is necessary to further mine each auxiliary case in these related cases. By using the idea of ​​random experimentation, the evaluation can be made more fair, and the most similar cases can be selected as the basis for comparison.

[0049] Step S403: According to step S402, calculate the energy efficiency value of the auxiliary cases included in each core case in the submitted case in each dimension of the related cases, sum the calculated energy efficiency values, and obtain the comprehensive energy efficiency value of the extracted case relative to the related cases.

[0050] Step S404: Complete all preset search counts, filter the related cases in the search counts where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches the maximum value, and take the related cases where the maximum value occurs as the optimal solution and output the optimal solution.

[0051] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In the trial auxiliary data management system and method based on judicial case quality evaluation provided by the present invention, the core cases and auxiliary cases are decomposed and reorganized to construct a related case cache library, thereby further mining related cases from the two aspects of core cases and auxiliary cases, outputting the related cases that are closest to the cases submitted for evaluation, and using the optimal solution as a reference, the cases that need to be checked and verified can be quickly found, which greatly reduces the errors and inaccuracies caused by manual operation, improves the quality and efficiency of trials, and the optimal solution found can also provide positive or negative preventive references for trial retrospective. Attached Figure Description

[0052] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0053] Figure 1 This is a schematic diagram of the trial auxiliary data management system based on judicial case quality evaluation of the present invention;

[0054] Figure 2 This is a schematic diagram illustrating the steps of the trial auxiliary data management method based on judicial case quality evaluation of the present invention;

[0055] Figure 3 This is a flowchart illustrating the trial auxiliary data management method based on judicial case quality evaluation of the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Please see Figures 1-3 The present invention provides the following technical solution:

[0058] Please see Figure 1 In this first embodiment: a trial assistance data management system based on judicial case quality evaluation is provided. The system includes: a data acquisition module, a data processing module, a case screening module, a storage and display module, and a case database module.

[0059] The data acquisition module is used to collect submitted cases and to break down the submitted cases.

[0060] The data processing module is used to calculate case label values ​​and to correct those values.

[0061] The case filtering module is used to judge the tag values ​​of submitted cases; to obtain auxiliary examples of submitted cases, to build a case filtering model, and to output the optimal filtering solution.

[0062] The storage and display module is used to obtain the optimal solution. If the optimal solution is a successful case, the submitted case is directly stored in the case database. Otherwise, the submitted case is sent to the case display front-end for random inspection and display.

[0063] The case database module is used to store cases;

[0064] The output of the data acquisition module is connected to the input of the data processing module, the output of the data processing module is connected to the input of the case screening module, the output of the case screening module is connected to the input of the storage and display module, and the output of the storage and display module is connected to the input of the case database module.

[0065] The data acquisition module includes an acquisition unit and a decomposition unit; the output of the acquisition unit is connected to the input of the decomposition unit.

[0066] The collection unit is used to randomly collect submitted cases;

[0067] The decomposition unit is used to extract core cases from any submitted case according to the process of case filing, mediation, service of process, preservation, enforcement, and case file scanning and entry; it is used to decompose submitted cases with different levels of principal-agent relationships into core cases and auxiliary cases, decomposing them into K core cases and W auxiliary cases, where K and W are constants; it is used to combine the extracted submitted cases according to the correspondence between core cases and auxiliary cases, wherein the combination method is that one core case corresponds to at least one auxiliary case.

[0068] The data processing module also includes a tag value calculation unit and a tag value correction unit; the output of the tag value calculation unit is connected to the input of the tag value correction unit.

[0069] The tag value calculation unit is used to extract feature information of core cases, retrieve related cases from the case database according to the similarity matching algorithm model, and build a case association cache library; it is used to extract all cases in the case association cache library, mine the core cases of each case, and deduplicate and integrate the mined core cases; it is used to encode the deduplicated and integrated core cases to generate an encoding set, which is denoted as {I1, I2, ..., I...} n}, where each code corresponds to a core case; used to generate a separate set of core cases from the code set, where each separate set contains a set of submitted cases and a set of each associated case; used to denote a separate set as a set method P, and to calculate the tag value of any case. Where S represents the label value of any case, a represents any code in a single set of cases, and L a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation.

[0070] The tag value correction unit is used to obtain the code of each core event in the submitted cases, calculate the weight of any core event code in the case association database, and if the weight of a core event in the submitted cases is less than the weight threshold, the core event is removed from the code and re-coded. It is also used to iterate through the case association cache library to calculate the correction value of the submitted case tag value. , where X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1Let M represent the label value of the commit case in the (q-1)th iteration, where q represents the iteration number; the actual label value used to calculate the commit case in the qth iteration is M=M q +X q .

[0071] The case filtering module includes a tag value judgment unit and a case filtering unit; the output of the tag value judgment unit is connected to the input of the case filtering unit.

[0072] The tag value judgment unit is used to calculate the average tag value of the associated cases. When the tag value of the submitted case is less than the average value, the tag value of the submitted case is corrected. When the tag value of the submitted case is greater than or equal to the average value, the auxiliary case of the submitted case is obtained.

[0073] The case selection unit is used to select a core case from the submitted cases as a search dimension and a corresponding auxiliary case as a search source, resulting in a total of K dimensions and W search sources. It is also used to preset the number of searches and, using a randomized trial approach, calculate the efficiency of any search source across different cases. Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,j G represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases. i，j This represents the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among the related cases, where L is a random number in the interval [-1, 1]. The judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension. This is used to calculate the energy efficiency value of the auxiliary cases included in each core case in the submitted case in each dimension of the related cases, and to sum the calculated energy efficiency values ​​to obtain the comprehensive energy efficiency value of the extracted case relative to the related cases. This is used to filter the related cases where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches its maximum value in the number of searches, and to output the optimal solution when the maximum value is found.

[0074] Please see Figures 2-3 In this second embodiment: a method for managing trial auxiliary data based on judicial case quality evaluation is provided, which includes the following steps:

[0075] Based on the process of case filing, mediation, service of process, preservation, enforcement, and case file scanning and entry, the core cases are extracted from the submitted cases.

[0076] For submitted cases with different levels of principal-agent relationships, core cases and auxiliary cases are decomposed into K core cases and W auxiliary cases, where K and W are constants. Based on the correspondence between core cases and auxiliary cases, the submitted cases are decomposed and combined, where one core case corresponds to at least one auxiliary case.

[0077] Extract feature information from core cases, retrieve related cases from the case database based on a similarity matching algorithm model, and construct a case association cache library; extract all cases from the case association cache library, perform core case mining for each case, and deduplicate and integrate the mined core cases; encode the deduplicated and integrated core cases to generate an encoding set, denoted as {I1, I2, ..., I...} n Each code corresponds to a core case.

[0078] Based on the encoding set, a separate set of core cases is generated for each case. This separate set includes the submitted case set and each associated case set. Each separate set is denoted as a set type P. The label value for any case is calculated using the following formula:

[0079]

[0080] Where S represents the label value of any case, a represents any code in a single set of cases, and L a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation.

[0081] Calculate the average value of the tag values ​​of the related cases, and correct the tag values ​​of the submitted cases when the tag values ​​of the submitted cases are less than the average value;

[0082] The specific implementation process for correcting the label values ​​of submitted cases includes:

[0083] The code for each core case in the submitted cases is obtained. In the case association database, the weight of any core case code is calculated. If the weight of a core case in the submitted cases is less than the weight threshold, the core case is removed from the code and re-coded. The case association cache is iterated until the label value of the submitted cases is greater than or equal to the average value, at which point the iteration stops. For example, if the weights of I1 to I4 are 0.36, 0.27, 0.18, and 0.18 respectively, and the weight threshold is set to 0.3, then core case I4 will be removed from the code. The smaller the weight of a core case, the less impact it has on the case analysis, while consuming a lot of resources and making the data analysis inaccurate.

[0084] The formula for calculating the corrected value of the tag for the submitted case is as follows:

[0085]

[0086] Among them, X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1 This represents the label value of the submitted case in the (q-1)th iteration, where q represents the iteration number;

[0087] Calculate the actual label value of the submitted case in the q-th iteration: M = M q +X q ;

[0088] When the tag value of the submitted case is greater than or equal to the average value, obtain auxiliary cases of the submitted case, build a case screening model, and output the optimal screening solution;

[0089] The specific implementation process of outputting and filtering the optimal solution includes:

[0090] If we take a core case as a search dimension and an auxiliary case as a search source, we will get a total of K dimensions and W search sources.

[0091] The number of searches is preset, and the efficiency of any search source in different cases is calculated using the concept of randomized experimentation. The specific calculation formula is as follows:

[0092]

[0093] Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,j G represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases.i，j Let L represent the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among the related cases, where L is a random number in the interval [-1, 1]; the judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension.

[0094] For each core case in the submitted cases, the energy efficiency value of the auxiliary cases included is calculated in each dimension of the related cases. The calculated energy efficiency values ​​are summed to obtain the comprehensive energy efficiency value of the extracted case relative to the related cases.

[0095] Complete all preset number of searches, filter the related cases in the search count where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches the maximum value, and take the related cases where the maximum value occurs as the optimal solution and output the optimal solution;

[0096] If the optimal solution is a successful case, the submitted case will be stored directly in the case database; otherwise, the submitted case will be sent to the case display front-end for random checks and display.

[0097] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0098] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for managing trial auxiliary data based on judicial case quality evaluation, characterized in that: The method includes the following steps: Step S100: The data acquisition module obtains the submitted cases and decomposes them into K core cases and W auxiliary cases, where each core case corresponds to at least one auxiliary case; where K and W are constants. Step S200: Based on the core cases, retrieve the case database and establish a case association model; encode the core cases and calculate the case tag values ​​based on the encoded data; Step S300: Calculate the average value of the label values ​​of the associated cases. When the label value of a submitted case is less than the average value, correct the label value of the submitted case and return to step S200. Step S400: When the label value of the submitted case is greater than or equal to the average value, obtain auxiliary cases of the submitted case, establish a case screening model, and output the optimal screening solution; Step S500: If the optimal solution is a successful case, the submitted case is directly stored in the case database; otherwise, the submitted case is sent to the case display front end for random inspection and display. The specific implementation process for calculating the case label value in step S200 includes: Step S201: Extract the feature information of core cases; based on the similarity matching algorithm model, retrieve related cases from the case database to construct a case association cache library; extract all cases from the case association cache library, perform core case mining for each case, and deduplicate and integrate the mined core cases; encode the deduplicated and integrated core cases to generate an encoding set, denoted as {I1, I2, ..., I...} n Each code corresponds to a core case. Step S202: Based on the encoding set, generate a separate set of core cases for each case. This separate set includes the submitted case set and each associated case set. Denote each separate set as a set type P. Calculate the tag value for any case using the following formula: ； Where S represents the label value of any case, a represents any code in a single set of cases, and y a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation. The specific implementation process of correcting the label value of the submitted case in step S300 includes: Step S301: Obtain the code of each core case in the submitted cases. In the case association database, calculate the weight of any core case code. If the weight of a core case in the submitted cases is less than the weight threshold, remove the core case from the code and re-encode the core case. Return to step S200 and iterate through the case association cache until the tag value of the submitted cases is greater than or equal to the average value. The iteration stops then. Step S302: Calculate the correction value of the tag value of the submitted case. The specific calculation formula is as follows: ； Among them, X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1 This represents the label value of the submitted case in the (q-1)th iteration, where q represents the iteration number; Step S303: Calculate the actual label value of the submitted case in the q-th iteration: M = M q +X q ; The specific implementation process of outputting and filtering the optimal solution in step S400 includes: Step S401: Take a core case of the submitted case as a search dimension, and take an auxiliary case of the auxiliary case corresponding to the core case as a search source, then a total of K dimensions and W search sources are obtained; Step S402: Preset the number of searches, and use the idea of ​​randomized experimentation to calculate the efficiency value of any search source in different cases. The specific calculation formula is as follows: ； Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,j G represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases. i，j This represents the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among related cases, where L is a random number in the interval [-1, 1]; the judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension. Step S403: According to step S402, calculate the energy efficiency value of the auxiliary cases included in each core case in the submitted case in each dimension of the related cases, sum the calculated energy efficiency values, and obtain the comprehensive energy efficiency value of the extracted case relative to the related cases. Step S404: Complete all preset search counts, filter the related cases in the search counts where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches the maximum value, and take the related cases where the maximum value occurs as the optimal solution and output the optimal solution.

2. The method for managing trial auxiliary data based on judicial case quality evaluation according to claim 1, characterized in that, The specific implementation process of decomposing the submitted case into K core cases and W auxiliary cases in step S100 includes: Step S101: Based on the process of case filing, mediation, service of process, preservation, enforcement, and case file scanning and entry, the core cases of the submitted cases are extracted; Step S102: Decompose the submitted cases with different levels of principal-agent relationships into core cases and auxiliary cases, decomposing them into K core cases and W auxiliary cases, where K and W are constants; according to the correspondence between core cases and auxiliary cases, combine the submitted cases after stripping them, wherein the combination method is that one core case corresponds to at least one auxiliary case.

3. A trial support data management system based on judicial case quality evaluation, characterized in that: The system includes: a data acquisition module, a data processing module, a case screening module, a storage and display module, and a case database module; The data acquisition module is used to acquire submitted cases and decompose them into K core cases and W auxiliary cases, where each core case corresponds to at least one auxiliary case; where K and W are constants. The data processing module is used to calculate case label values ​​and to correct case label values. The case filtering module is used to judge the tag values ​​of submitted cases; to obtain auxiliary examples of submitted cases, to establish a case filtering model, and to output the optimal filtering solution. The storage and display module is used to obtain the optimal solution. If the optimal solution is a successful case, the submitted case is directly stored in the case database. Otherwise, the submitted case is sent to the case display front end for random inspection and display. The case database module is used to store cases; The output of the data acquisition module is connected to the input of the data processing module, the output of the data processing module is connected to the input of the case filtering module, the output of the case filtering module is connected to the input of the storage and display module, and the output of the storage and display module is connected to the input of the case database module. The data processing module further includes a tag value calculation unit and a tag value correction unit; the output of the tag value calculation unit is connected to the input of the tag value correction unit. The tag value calculation unit is used to extract feature information of core cases, retrieve related cases from the case database according to the similarity matching algorithm model, and construct a case association cache library; it is used to extract all cases in the case association cache library, mine the core cases of each case, and deduplicate and integrate the mined core cases; it is used to encode the deduplicated and integrated core cases to generate an encoding set, which is denoted as {I1, I2, ..., I...} n }, where each code corresponds to a core case; used to generate a separate set of core cases for cases based on the code set, the separate set containing the submitted case set and each associated case set; used to denote a separate set as a set method P, and to calculate the tag value of any case. Where S represents the label value of any case, a represents any code in a single set of cases, and y a Y represents the number of individual sets of successful cases in the case association cache for any given code 'a'. a This represents the number of individual sets of all cases in the case association cache library for any given code 'a', where successful cases are those that have been recognized by the judicial case quality evaluation. The tag value correction unit is used to obtain the code of each core case in the submitted cases, calculate the weight of any core case code in the case association database, and if the weight of a core case in the submitted cases is less than the weight threshold, remove the core case from the code and re-encode the core case; it is also used to iterate the case association cache library and calculate the correction value of the submitted case tag value. , where X q This represents the adjusted value of the label value of the submitted case, where Q represents the adjustment factor and M represents the adjustment factor. q M represents the tag value of the commit case in the q-th iteration. q-1 Let M represent the label value of the commit case in the (q-1)th iteration, where q represents the iteration number; the actual label value used to calculate the commit case in the qth iteration is M=M q +X q ; The case filtering module further includes a tag value judgment unit and a case filtering unit; the output of the tag value judgment unit is connected to the input of the case filtering unit. The tag value judgment unit is used to calculate the average tag value of the associated cases. When the tag value of the submitted case is less than the average value, the tag value of the submitted case is corrected. When the tag value of the submitted case is greater than or equal to the average value, the auxiliary case of the submitted case is obtained. The case selection unit is used to select a core case from the submitted cases as a search dimension and an auxiliary case from the corresponding auxiliary case as a search source, thus obtaining a total of K dimensions and W search sources. It is also used to preset the number of searches and, using a randomized trial approach, calculate the efficiency of any search source across different cases. Among them, B i，j C represents the energy efficiency value of the i-th search source in the j-th dimension obtained from any random trial within a preset number of searches. i,j G represents the judicial case quality evaluation value of the i-th search source in the j-th dimension of the submitted cases. i，j This represents the judicial case quality evaluation value of the i-th retrieval source in the j-th dimension among the related cases, where L is a random number in the interval [-1, 1]. The judicial case quality evaluation value is obtained by assessing the performance of the i-th retrieval source in the j-th dimension. It is used to calculate the energy efficiency value of the auxiliary cases included in each core case in the submitted case in each dimension of the related cases, and to sum the calculated energy efficiency values ​​to obtain the comprehensive energy efficiency value of the extracted case relative to the related cases. It is used to filter the related cases where the comprehensive energy efficiency value of the submitted case relative to the related cases reaches its maximum value in the number of searches, and to output the optimal solution when the maximum value is found.

4. The trial auxiliary data management system based on judicial case quality evaluation according to claim 3, characterized in that: The data acquisition module further includes an acquisition unit and a decomposition unit; the output end of the acquisition unit is connected to the input end of the decomposition unit. The collection unit is used to randomly collect submitted cases; The decomposition unit is used to extract core cases from any submitted case based on the process of case filing, mediation, service of process, preservation, execution, and case file scanning and entry. This is used to decompose submitted cases with different levels of principal-agent relationships into core cases and auxiliary cases, resulting in K core cases and W auxiliary cases, where K and W are constants. This is used to combine submitted cases after separating them according to the correspondence between core cases and auxiliary cases, wherein the combination method is that one core case corresponds to at least one auxiliary case.

Images