Management system and method for quality evaluation system of judicial cases based on multi-source data

Abstract

The application discloses a multi-source data-based judicial case quality evaluation system management system and method, and belongs to the technical field of data processing. Through source discussion on multi-source data factors influencing judicial case quality, labeling on entity parts, and division into two categories of reaching standards and exceeding standards for procedure parts, a coded case label set under unified rules is generated, a reaching standard association set and an exceeding standard association set are generated, and further in-depth mining and analysis are carried out from social scales, the possibility of reaching standards in non-reached standards, and the possibility of non-reached standards in reached standards. Whether the case can be stored in a judicial case database as a high-quality judicial case is judged and analyzed in multiple aspects of static and dynamic, internal and external, and subjective and objective, so that the systematization and accuracy of the construction of the judicial case database are ensured. Meanwhile, in the process of judgment and analysis, the error and inaccuracy caused by human operation are greatly reduced, and the auditing of the quality of the judicial case has a comprehensive auxiliary guiding effect.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a management system and method for evaluating the quality of judicial cases based on multi-source data. Background Technology

[0002] Case studies serve as vivid and engaging legal education materials, the best textbooks for the rule of law. Guiding cases play a crucial role in unifying judicial application standards, standardizing judicial conduct, clarifying complex and difficult issues in legal application, conducting legal education through cases, and promoting the rule of law. Currently, the court system often conducts judicial quality assessments through open peer review, followed by a collective determination of case rankings. This process is susceptible to subjective influence, making it time-consuming and laborious, and lacking accuracy in case evaluation. Furthermore, the evaluation process often overlooks key points, failing to achieve precise assessment of specific cases and issues. Consequently, many barriers exist in the process of perfecting and developing high-quality, guiding cases. Summary of the Invention

[0003] The purpose of this invention is to provide a management system and method for evaluating the quality of judicial cases based on multi-source data, 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 judicial case quality evaluation system management system based on multi-source data includes: a data acquisition module, an association set module, a data processing module, and a case database module.

[0006] The data acquisition module is used to collect text information of cases to be processed; to split the case text information into entity parts and procedural parts; and to encode all retrieved judicial cases in a unified manner according to the case number of the judicial case.

[0007] The association set module is used to generate a compliant set from compliant procedures and an excess set from non-compliant procedures that have exceeded the execution time. The compliant set and the excess set are then traversed and searched in the judicial case database to obtain the compliant association set and the excess association set associated with the compliant set and the excess set, respectively. It is also used to obtain valid association sets and invalid association sets.

[0008] The data processing module is used to calculate the impact value of judicial cases, the first evaluation value of cases to be processed, the second evaluation value of cases to be processed, and the comprehensive evaluation value of cases to be processed.

[0009] The case database module is used to store the case in the judicial case database if the comprehensive evaluation value of the case to be processed is greater than the average comprehensive evaluation value of the judicial cases stored in the judicial case database; otherwise, it submits the case to the superior court for review.

[0010] Furthermore, the data acquisition module also includes an acquisition unit, a decomposition unit, and an encoding unit;

[0011] The acquisition unit is used to acquire the text information of the case to be processed and to identify the text information of the case.

[0012] The decomposition unit is used to split the case text information into an entity part and a procedural part; the entity part is the factual part of the case, and the procedural part is the necessary procedures for handling the case; it is used to strip the coded judicial cases of social standard influencing factors, the social standard influencing factors being public opinion heat and public evaluation value; the public opinion heat includes the increase in attention brought about by the dissemination of judicial cases and the attention drag-derived quantity brought about by the progress of the judicial cases themselves, the attention drag-derived quantity is at least one other point of contention derived from a certain ongoing point of contention in the judicial case, wherein the way in which at least one other point of contention is derived from a certain ongoing point of contention is recorded as a derivation method, and the number of all derivation methods is recorded as the attention drag-derived quantity; the public evaluation value is the degree of public acceptance of the judicial case judgment result; the acceptance degree is the ratio of the number of public acceptances of the judicial case judgment result to the total number of participants in the evaluation;

[0013] The encoding unit is used to obtain tags for each entity part of the case to be processed according to tagging rules. The tagging rules involve generating a tag for each entity part in a judicial case database, where the judicial case database stores high-quality judicial cases that have received judicial case quality evaluation and recognition. All tags for the case to be processed are traversed and retrieved from the judicial case database, and all retrieved judicial cases are statistically analyzed. All retrieved judicial cases are then encoded in a unified manner according to their case numbers. Based on the encoding, the retrieval results for any tag y of the case to be processed are generated into a set, denoted as y = {I1, I2, ..., I...}. n}, where I n This indicates the nth encoding retrieved for the tag y.

[0014] Furthermore, the association set module also includes an association set generation unit and an association set calculation unit;

[0015] The association set generation unit is used to obtain the program parts of the case to be processed after the case information text recognition and segmentation, obtain the execution time of each program, and if the execution time of a program is within the specified time range required for program execution, then the program execution meets the standard; otherwise, the program execution does not meet the standard. The compliant programs are generated into a compliant set, and the non-compliant programs are generated into an excess set that exceeds the execution time. The compliant set and the excess set are respectively traversed and searched in the judicial case database to obtain the compliant association set and the excess association set associated with the compliant set and the excess set, respectively.

[0016] The association set calculation unit is used to denot the compliant set as λ1, the compliant association set as λ2, the excess set as β1, and the excess association set as β2; if any non-compliant program exists in the compliant association set, then the non-compliant program is marked as a valid association; if any compliant program exists in the excess association set, then the compliant program is marked as an invalid association; the valid association set and the invalid association set are obtained respectively, wherein the valid association set φ1 = β1 ∩ λ2, and the invalid association set φ2 = λ1 ∩ β2.

[0017] Furthermore, the data processing module also includes an impact value calculation unit, a first evaluation value calculation unit, a second evaluation value calculation unit, and a comprehensive evaluation value calculation unit;

[0018] The impact value calculation unit is used to strip social standard influencing factors from any coded judicial case, and to obtain the change in attention increment and the change in acceptance under any derivative method; it is also used to obtain the impact value of the coded judicial case based on the social standard influencing factors. Among them, E(I i ) represents the impact value of any coded judicial case, I i ∈{I1, I2, ..., I n}, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method;

[0019] The first evaluation value calculation unit is used to preset the impact value threshold of coded judicial cases, and mark coded judicial cases that exceed the threshold as related cases; obtain the label value B(y) = a / n of any label set, where a represents the number of related cases in the label set; and calculate the first evaluation value of the case to be processed based on the label value, with V1 = ∑ y∈Y B(y), where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed;

[0020] The second evaluation value calculation unit is used to calculate the second evaluation value of the case to be processed. Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets that meet the criteria λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets that exceed the criteria λ1.

[0021] The comprehensive evaluation value calculation unit is used to obtain the comprehensive evaluation value V = V1 × V2 of the case to be processed based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed.

[0022] A management method for the quality evaluation system of judicial cases based on multi-source data, which includes the following steps:

[0023] Step S100: Obtain the case text information to be processed, identify and split the case text information into an entity part and a procedure part; the entity part is the factual part of the case, and the procedure part is the necessary procedure for handling the case;

[0024] Based on the above methods, judicial cases often involve various factors that may affect judicial quality during the execution process. These include litigation costs, seized funds and property, management of fines and confiscated property, recusal, case handling time limits, case acceptance, filing procedures, litigation preservation measures, court proceedings, execution procedures, judicial supervision, indictments, case statements and defenses, and evidence presentation. Tracing and identifying these factors is beneficial for evaluating and analyzing case quality. In the judicial process, these factors can be divided into two main categories: substantive and procedural. The substantive part includes relevant legal provisions, case facts, and judgment results; it is static and objective. However, the procedural part is a necessary process in the execution of cases. The entire process from case filing to trial has a corresponding procedural flow, which is executed by court personnel. Due to factors such as human subjectivity, inconsistent judgments in similar cases often occur; therefore, the procedural part is dynamic and subjective. Therefore, this invention analyzes both the substantive and procedural parts, tracing the source of multi-source data.

[0025] Step S200: According to the tagging rules, obtain the tags for each entity part of the case to be processed, wherein the tagging rules generate a tag for each entity part in the judicial case database, and the judicial case database stores high-quality judicial cases that have obtained judicial case quality evaluation and recognition; traverse and search all tags of the case to be processed in the judicial case database, count all the searched judicial cases, and encode all the searched judicial cases in a unified way according to the case number; according to the encoding, generate a set of the tag search results of the case to be processed, with one tag corresponding to one set;

[0026] Step S300: Isolate the social standard influencing factors from the coded judicial cases. These factors are public opinion heat and public evaluation value. Public opinion heat includes the increase in attention brought about by the dissemination of the judicial case and the attention drag-out derived from the progress of the judicial case itself. Attention drag-out is defined as at least one additional point of contention derived from a particular ongoing point of contention in the judicial case. The manner in which at least one additional point of contention is derived from a particular ongoing point of contention is recorded as one derivation method, and the total number of all derivation methods is recorded as the attention drag-out derived quantity. Public evaluation value is the degree of public acceptance of the judicial case's judgment. Acceptance is the ratio of the number of public acceptances of the judicial case's judgment to the total number of participants in the evaluation. Based on the social standard influencing factors, obtain the influence value of the coded judicial cases.

[0027] Based on the above methods, the evaluation standards for judicial cases are often difficult to unify due to the subjective initiative of human beings. Therefore, it is necessary to establish a judicial case database to assist in case handling and improve the efficiency and quality of case handling. Improving judicial quality evaluation standards cannot be separated from both legal and social standards. Legal standards provide the internal legal basis, while social standards are the external supervision of the public. Combining internal and external circumstances, it is necessary to first search, compare, and classify the substantive parts of the law in the judicial case database, and then, based on the characteristics of different judicial cases, to separate the social standard influencing factors of judicial cases. The most direct manifestation of the public's influence on cases is whether they agree or disagree with the court's judgment, and the degree of different views and attention they pay to a particular viewpoint.

[0028] Step S400: Based on the impact value, obtain the label value of each entity part of the case to be processed; based on the label value, calculate the first evaluation value of the case to be processed;

[0029] Step S500: Obtain the program portion of the case to be processed after text recognition and segmentation of case information; obtain the execution time of each program; if the execution time of a program is within the specified time range required for program execution, the program execution is qualified; otherwise, the program execution is unqualified; generate a qualified set for qualified programs and an unqualified set for programs that exceed the execution time; traverse and search the judicial case database for the qualified set and the unqualified set respectively to obtain the qualified association set and the unqualified association set associated with the qualified set and the unqualified set respectively; calculate the second evaluation value of the case to be processed.

[0030] Step S600: Based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed, obtain the comprehensive evaluation value V = V1 × V2 of the case to be processed. If the comprehensive evaluation value of the case to be processed is greater than the average value of the comprehensive evaluation values ​​of judicial cases stored in the judicial case database, then the case to be processed is stored in the judicial case database; otherwise, the case to be processed is submitted to the superior court for review.

[0031] Furthermore, the specific implementation process of step S200 includes:

[0032] Step S201: Retrieve judicial cases from the judicial case database, identify and split the case text information for each judicial case, count all the entity parts after splitting, and generate a label for each entity part; according to the labeling rules, label each entity part of the case to be processed, and obtain a total of K labels, and record any label as y;

[0033] Step S202: For each tag of the case to be processed, perform a traversal search in the judicial case database to complete the traversal search of all tags of the case to be processed; encode all the retrieved judicial cases in a unified way according to the case number, resulting in W codes; based on the codes, generate a set of search results for any tag y, denoted as y = {I1, I2, ..., I...} n}, where I n This indicates the nth encoding retrieved for the tag y.

[0034] Furthermore, the specific implementation process of step S300 includes:

[0035] Step S301: Extract the influencing factors of social standards from any coded judicial case, and obtain the changes in attention increment and recognition under any derivative method;

[0036] Step S302: Based on the influencing factors of social standards, obtain the impact value of the coded judicial cases. The specific calculation formula is as follows:

[0037]

[0038] Among them, E(I i ) represents the impact value of any coded judicial case, I i ∈{I1, I2, ..., I n}, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method;

[0039] According to the above method, the ratio of the change in acceptance to the change in attention can intuitively reflect the change in public acceptance of judicial cases as the level of attention changes. This ratio often fluctuates due to changes in the mode of development, and eventually tends to converge; E(I i The larger the value of ), the more the judicial case can withstand public scrutiny as the derivation method changes, meaning that it is more convincing and accurate to store as a high-quality judicial case in the judicial case database.

[0040] Furthermore, the specific implementation process of calculating the first evaluation value of the case to be processed based on the label value in step S400 includes:

[0041] Step S401: Preset the impact value threshold of coded judicial cases, and mark coded judicial cases that exceed the threshold as related cases; obtain the label value B(y) = a / n for any label set, where a represents the number of related cases in the label set;

[0042] Step S402: Calculate the first evaluation value of the case to be processed based on the label value. The specific calculation formula is as follows:

[0043]

[0044] Where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed;

[0045] According to the above method, the purpose of searching each tag of the cases to be processed is to avoid the judgment scope being too small, which would lead to the lack of universality and balance in the data comparison. At the same time, the purpose of encoding all judicial cases after searching all tags in a unified way is to ensure that the data calculation is carried out under a unified dimension and standard, making the data calculation more accurate and objective, and enabling the data of related cases to be carried out in the whole domain. Therefore, when calculating the first evaluation value, the tag values ​​are summed.

[0046] Furthermore, the specific implementation process of calculating the second evaluation value of the case to be processed in step S500 includes:

[0047] Step S501: Denote the compliant set as λ1, the compliant association set as λ2, the excess set as β1, and the excess association set as β2; if any non-compliant program exists in the compliant association set, then mark the non-compliant program as a valid association; if any compliant program exists in the excess association set, then mark the compliant program as an invalid association; obtain the valid association set and the invalid association set respectively, where the valid association set φ1 = β1 ∩ λ2, and the invalid association set φ2 = λ1 ∩ β2;

[0048] Step S502: Calculate the second evaluation value of the case to be processed. The specific calculation formula is as follows:

[0049]

[0050] Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets that meet the criteria λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets that exceed the criteria set β1.

[0051] Based on the above methods, when analyzing case data, the execution time of each procedure reflects case-handling efficiency. Execution times that are too fast or too slow may indicate abnormal subjective initiative in the execution process. We must not blindly pursue speed at the expense of detail and quality, nor should we delay procedures, causing negative emotions for the parties involved and hindering the smooth progress of other procedures, leading to a series of chain reactions such as insufficient manpower in handling other cases by the court. Conversely, when analyzing a specific procedure, we must not ignore the execution of other procedures; a holistic view is necessary. Therefore, it is essential to analyze both the compliant set and the excess sets. The system comprehensively evaluates cases based on the big data foundation of the judicial case database, enabling effective analysis of both compliant and non-compliant cases. The formula M(φ1) / M(λ1) represents the possibility of compliance within non-compliant procedures, while the formula M(φ2) / M(β1) represents the possibility of non-compliance within compliant procedures. This comprehensive evaluation yields a second evaluation value, thus avoiding the existence of "extra-legal standards" and ensuring objectivity while avoiding subjectivity in individual cases. A higher second evaluation value indicates greater persuasiveness and accuracy in storing the case in the judicial case database as a high-quality judicial case.

[0052] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The judicial case quality evaluation system management system and method based on multi-source data provided by this invention analyzes cases at two levels: the substantive part and the procedural part. It traces and discusses the multi-source data factors affecting the quality of judicial cases, labels the substantive part, and classifies the procedural part into two categories: meeting standards and exceeding standards. This generates a set of coded case labels under unified rules, as well as sets of meeting standards and exceeding standards. Furthermore, it conducts in-depth analysis from social scales, the possibility of non-compliance within meeting standards, and the possibility of meeting standards within non-compliance. Thus, it performs static and dynamic, internal and external, and subjective and objective multi-level evaluation and analysis on whether a case can be stored as a high-quality judicial case in the judicial case database, ensuring the systematicness and accuracy of the judicial case database construction. At the same time, in the evaluation and analysis process, it greatly reduces the errors and inaccuracies caused by manual operation, and has a comprehensive auxiliary and guiding role in the review of judicial case quality. Attached Figure Description

[0053] 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:

[0054] Figure 1 This is a schematic diagram of the structure of the judicial case quality evaluation system management system based on multi-source data of the present invention;

[0055] Figure 2 This is a schematic diagram illustrating the steps of the judicial case quality evaluation system management method based on multi-source data 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 1-2 The present invention provides the following technical solution:

[0058] Please see Figure 1 In this first embodiment: a judicial case quality evaluation system management system based on multi-source data is provided. The system includes: a data acquisition module, an association set module, a data processing module, and a case database module.

[0059] The data acquisition module is used to collect text information of cases to be processed; to split the case text information into entity and procedural parts; and to encode all retrieved judicial cases in a unified manner according to the case number of the judicial case.

[0060] The association set module is used to generate a compliant set for compliant procedures and an excess set for non-compliant procedures that have exceeded the execution time. The compliant set and the excess set are then traversed and searched in the judicial case database to obtain the compliant association set and the excess association set associated with the compliant set and the excess set, respectively. It is also used to obtain the valid association set and the invalid association set.

[0061] The data processing module is used to calculate the impact value of judicial cases, the first evaluation value of cases to be processed, the second evaluation value of cases to be processed, and the comprehensive evaluation value of cases to be processed.

[0062] The case database module is used to store a case in the judicial case database if the comprehensive evaluation value of the case to be processed is greater than the average comprehensive evaluation value of judicial cases stored in the judicial case database; otherwise, the case to be processed will be submitted to the superior court for review.

[0063] The data acquisition module also includes an acquisition unit, a decomposition unit, and an encoding unit;

[0064] The acquisition unit is used to acquire the text information of the case to be processed and to identify the text information of the case.

[0065] The decomposition unit is used to break down case text information into substantive and procedural parts. The substantive part is the factual part of the case, and the procedural part is the necessary procedures for handling the case. It is used to strip the coded judicial cases of social standard influencing factors, which are public opinion intensity and public evaluation value. Public opinion intensity includes the incremental attention brought about by the dissemination of judicial cases and the attention drag-derived quantity brought about by the progress of the judicial case itself. Attention drag-derived quantity refers to at least one additional point of contention derived from a certain ongoing point of contention in the judicial case. The way in which at least one additional point of contention is derived from a certain ongoing point of contention is recorded as a derivation method, and the number of all derivation methods is recorded as the attention drag-derived quantity. Public evaluation value is the public's acceptance of the judicial case's judgment result; acceptance is the ratio of the number of public acceptances of the judicial case's judgment result to the total number of participants in the evaluation.

[0066] The encoding unit is used to obtain the tags for each entity part of the case to be processed according to the tagging rules. The tagging rules generate a tag for each entity part in the judicial case database, which stores high-quality judicial cases that have received judicial case quality evaluation and recognition. All tags for the case to be processed are traversed and retrieved from the judicial case database, and all retrieved judicial cases are statistically analyzed. All retrieved judicial cases are then encoded in a unified manner according to their case numbers. Based on the encoding, the retrieval results for any tag y of the case to be processed are generated into a set, denoted as y = {I1, I2, ..., I...}. n}, where I n This indicates the nth encoding retrieved for the tag y.

[0067] The association set module also includes an association set generation unit and an association set calculation unit;

[0068] The association set generation unit is used to obtain the program parts of the case to be processed after the case information text recognition and segmentation, obtain the execution time of each program, and if the execution time of a program is within the specified time range required for program execution, the program execution is qualified; otherwise, the program execution is unqualified. The qualified programs are generated into a qualified set, and the unqualified programs are generated into an unqualified set that exceeds the execution time. The qualified set and the unqualified set are respectively traversed and searched in the judicial case database to obtain the qualified association set and the unqualified association set associated with the qualified set and the unqualified set, respectively.

[0069] The association set calculation unit is used to denote the compliant set as λ1, the compliant association set as λ2, the excess set as β1, and the excess association set as β2. If any non-compliant program exists in the compliant association set, then the non-compliant program is marked as a valid association; if any compliant program exists in the excess association set, then the compliant program is marked as an invalid association. The valid association set and the invalid association set are obtained respectively, where the valid association set φ1 = β1 ∩ λ2 and the invalid association set φ2 = λ1 ∩ β2.

[0070] The data processing module also includes an impact value calculation unit, a first evaluation value calculation unit, a second evaluation value calculation unit, and a comprehensive evaluation value calculation unit.

[0071] The impact value calculation unit is used to strip away the social standard influencing factors from any coded judicial case, obtaining the change in attention increment and the change in acceptance under any derivative method; it is also used to obtain the impact value of the coded judicial case based on the social standard influencing factors. Among them, E(I i ) represents the impact value of any coded judicial case, I i ∈{I1, I2, ..., I n}, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method;

[0072] The first evaluation value calculation unit is used to preset the impact value threshold of coded judicial cases, and mark coded judicial cases that exceed the threshold as related cases; obtain the label value B(y) = a / n of any label set, where a represents the number of related cases in the label set; and calculate the first evaluation value of the case to be processed based on the label value, with V1 = ∑ y∈Y B(y), where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed;

[0073] The second evaluation value calculation unit is used to calculate the second evaluation value of the case to be processed. Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets that meet the criteria λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets that exceed the criteria λ1.

[0074] The comprehensive evaluation value calculation unit is used to obtain the comprehensive evaluation value V = V1 × V2 of the case to be processed based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed.

[0075] Please see Figure 2In this second embodiment: a management method for a judicial case quality evaluation system based on multi-source data is provided, which includes the following steps:

[0076] The text information of the case to be processed is obtained, and after identification, it is split into an entity part and a procedural part. The entity part is the factual part of the case, and the procedural part is the necessary procedure for handling the case.

[0077] Retrieve judicial cases from the judicial case database, identify and segment the text information of each judicial case, count all the entity parts after segmentation, and generate a label for each entity part; according to the labeling rules, label each entity part of the case to be processed, and obtain a total of K labels, and denote any label as y;

[0078] Each tag of the case to be processed is traversed and searched in the judicial case database to complete the traversal and retrieval of all tags of the case to be processed; all judicial cases retrieved are coded in a unified way according to the case number, resulting in W codes; based on the codes, a set is generated for the search results of any tag y, denoted as y = {I1, I2, ..., I...} n}, where I n This represents the nth encoding obtained by retrieving the tag y;

[0079] The social standard influencing factors of coded judicial cases are stripped away. These factors are public opinion intensity and public evaluation. Public opinion intensity includes the incremental attention brought about by the dissemination of judicial cases and the attention drag generated by the progress of the judicial cases themselves. Attention drag is defined as at least one additional point of contention derived from a certain ongoing point of contention in a judicial case. The way in which at least one additional point of contention is derived from a certain ongoing point of contention is recorded as one type of derivation, and the total number of all derivation types is recorded as the attention drag. Public evaluation is the degree of public acceptance of the judicial case's judgment. Acceptance is the ratio of the number of public acceptances of the judicial case's judgment to the total number of participants in the evaluation. Based on the influencing factors of social standards, the impact value of the coded judicial cases is obtained.

[0080] By stripping away the social standard influencing factors from any coded judicial case, we can obtain the changes in attention increment and acceptance under any derivative method.

[0081] Based on the influencing factors of social standards, the impact value of coded judicial cases is obtained, and the specific calculation formula is as follows:

[0082]

[0083] Among them, E(I i) represents the impact value of any coded judicial case, I i ∈{I1, I2, ..., I n}, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method;

[0084] Based on the impact value, obtain the label value of each entity part of the case to be processed; based on the label value, calculate the first evaluation value of the case to be processed.

[0085] A threshold for the impact value of coded judicial cases is preset, and coded judicial cases that exceed the threshold are marked as related cases; the label value B(y) = a / n is obtained for any label set, where a represents the number of related cases in the label set;

[0086] Based on the tag value, calculate the first evaluation value of the case to be processed. The specific calculation formula is as follows:

[0087]

[0088] Where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed;

[0089] For example, the entity labels of the cases to be processed are {label 1, label 2}; for label 1, the retrieved set is {case I1, case I2, case I3}; for label 2, the retrieved set is {case I1, case I3, case I4, case I5}; where case I1 and case I3 are related cases, then the first evaluation value V1 of the cases to be processed is V1 = 2 / 3 + 2 / 4 = 1.17;

[0090] The process involves obtaining the program segments of the pending case after text recognition and segmentation of the case information, and obtaining the execution time of each program. If the execution time of a program is within the specified time range required for program execution, the program is considered to have met the execution standard; otherwise, the program is considered to have failed the execution standard. A set of qualified programs is generated, and a set of unqualified programs that exceed the execution time is generated. The qualified and unqualified sets are then traversed and searched in the judicial case database to obtain the qualified and unqualified association sets, respectively. Finally, a second evaluation value for the pending case is calculated.

[0091] Let λ1 denote the set of compliant programs, λ2 denote the set of compliant associations, β1 denote the set of non-compliant programs, and β2 denote the set of non-compliant associations. If any non-compliant program exists in the set of compliant associations, then the non-compliant program is marked as a valid association. If any compliant program exists in the set of non-compliant associations, then the compliant program is marked as an invalid association. Obtain the set of valid associations and the set of invalid associations, where the set of valid associations φ1 = β1 ∩ λ2 and the set of invalid associations φ2 = λ1 ∩ β2.

[0092] The second evaluation value for the case to be processed is calculated using the following formula:

[0093]

[0094] Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets that meet the criteria λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets that exceed the criteria set β1.

[0095] For example, the set of cases that meet the criteria is {Program 1, Program 2, Program 4, Program 5}, and the set that exceeds the criteria is {Program 3, Program 6, Program 7}. The sets of cases that meet the criteria and the set that exceed the criteria are traversed and searched in the judicial case database, resulting in the set of cases that meet the criteria being {Program 1, Program 2, Program 3, Program 4}, and the set of cases that exceed the criteria being {Program 1, Program 2, Program 3, Program 6, Program 7}. Further, the set of cases that meet the criteria is {Program 3}, and the set of cases that exceed the criteria is {Program 1, Program 2}. Then, the second evaluation value V2 of the case to be processed is V2 = 1 / 4 + 2 / 3 = 0.92.

[0096] Based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed, the comprehensive evaluation value V = V1 × V2 of the case to be processed is obtained; if the comprehensive evaluation value of the case to be processed is greater than the average value of the comprehensive evaluation values ​​of judicial cases stored in the judicial case database, the case to be processed is stored in the judicial case database; otherwise, the case to be processed is submitted to the superior court for review.

[0097] For example, if the first evaluation value of a case to be processed is 1.17 and the second evaluation value is 0.92, then the comprehensive evaluation value of the case to be processed is V = 1.17 * 0.92 = 1.07; and the average comprehensive evaluation value of judicial cases is 0.9, then the case to be processed will be stored in the judicial case database.

[0098] 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.

[0099] 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 management method for a judicial case quality evaluation system based on multi-source data, characterized in that: The method includes the following steps: Step S100: Obtain the case text information to be processed, identify and split the case text information into an entity part and a procedure part; the entity part is the factual part of the case, and the procedure part is the necessary procedure for handling the case; Step S200: According to the tagging rules, obtain the tags for each entity part of the case to be processed, wherein the tagging rules generate a tag for each entity part in the judicial case database, and the judicial case database stores high-quality judicial cases that have obtained judicial case quality evaluation and recognition; traverse and search all tags of the case to be processed in the judicial case database, count all the searched judicial cases, and encode all the searched judicial cases in a unified way according to the case number; according to the encoding, generate a set of the tag search results of the case to be processed, with one tag corresponding to one set; Step S300: Isolate the social standard influencing factors from the coded judicial cases. These factors are public opinion heat and public evaluation value. Public opinion heat includes the increase in attention brought about by the dissemination of the judicial case and the attention drag-out derived from the progress of the judicial case itself. Attention drag-out is defined as at least one additional point of contention derived from a particular ongoing point of contention in the judicial case. The manner in which at least one additional point of contention is derived from a particular ongoing point of contention is recorded as one derivation method, and the total number of all derivation methods is recorded as the attention drag-out derived quantity. Public evaluation value is the degree of public acceptance of the judicial case's judgment. Acceptance is the ratio of the number of public acceptances of the judicial case's judgment to the total number of participants in the evaluation. Based on the social standard influencing factors, obtain the influence value of the coded judicial cases. Step S400: Based on the impact value, obtain the label value of each entity part of the case to be processed; based on the label value, calculate the first evaluation value of the case to be processed; Step S500: Obtain the program portion of the case to be processed after text recognition and segmentation of case information; obtain the execution time of each program; if the execution time of a program is within the specified time range required for program execution, the program execution is qualified; otherwise, the program execution is unqualified; generate a qualified set for qualified programs and an unqualified set for programs that exceed the execution time; traverse and search the judicial case database for the qualified set and the unqualified set respectively to obtain the qualified association set and the unqualified association set associated with the qualified set and the unqualified set respectively; calculate the second evaluation value of the case to be processed. Step S600: Based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed, obtain the comprehensive evaluation value V = V1 × V2 of the case to be processed; if the comprehensive evaluation value of the case to be processed is greater than the average value of the comprehensive evaluation values ​​of judicial cases stored in the judicial case database, then store the case to be processed in the judicial case database; otherwise, submit the case to be processed to the superior court for review. The specific implementation process of calculating the first evaluation value of the case to be processed based on the label value in step S400 includes: Step S401: Preset the impact value threshold of the coded judicial cases, and mark the coded judicial cases that exceed the threshold as related cases; obtain the label value B(y)=a / n for any label set, where a represents the number of related cases in the label set; Step S402: Calculate the first evaluation value of the case to be processed based on the label value. The specific calculation formula is as follows: ； Where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed; The specific implementation process for calculating the second evaluation value of the case to be processed in step S500 includes: Step S501: Denote the compliant set as λ1, the compliant association set as λ2, the excess set as β1, and the excess association set as β2; if any non-compliant program exists in the compliant association set, then mark the non-compliant program as a valid association; if any compliant program exists in the excess association set, then mark the compliant program as an invalid association; obtain the valid association set and the invalid association set respectively, where the valid association set φ1 = β1 ∩ λ2, and the invalid association set φ2 = λ1 ∩ β2; Step S502: Calculate the second evaluation value of the case to be processed. The specific calculation formula is as follows: ； Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets that meet the criteria λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets that exceed the criteria λ1.

2. The management method for the judicial case quality evaluation system based on multi-source data according to claim 1, characterized in that, The specific implementation process of step S200 includes: Step S201: Retrieve judicial cases from the judicial case database, identify and split the case text information for each judicial case, count all the entity parts after splitting, and generate a label for each entity part; according to the labeling rules, label each entity part of the case to be processed, and obtain a total of K labels, and record any label as y; Step S202: For each tag of the case to be processed, perform a traversal search in the judicial case database to complete the traversal search of all tags of the case to be processed; encode all the retrieved judicial cases in a unified way according to the case number, resulting in W codes; based on the codes, generate a set of search results for any tag y, denoted as y={I1, I2, ..., I... n }, where I n This indicates the nth encoding retrieved for the tag y.

3. The management method for the judicial case quality evaluation system based on multi-source data according to claim 2, characterized in that, The specific implementation process of step S300 includes: Step S301: Extract the social standard influencing factors from any coded judicial case and obtain the change in attention increment and the change in acceptance under any derivative method; Step S302: Based on the influencing factors of social standards, obtain the impact value of the coded judicial cases. The specific calculation formula is as follows: ； Among them, E(I i ) represents the impact value of any coded judicial case, I i ϵ{I1,I2,...,I n }, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method.

4. A judicial case quality evaluation system management system based on multi-source data, used to implement the judicial case quality evaluation system management method based on multi-source data as described in claim 1, characterized in that, The system includes: a data acquisition module, an association set module, a data processing module, and a case database module; The data acquisition module is used to collect text information of cases to be processed; to split the case text information into entity parts and procedural parts; and to encode all retrieved judicial cases in a unified manner according to the case number of the judicial case. The association set module is used to generate a compliant set from compliant procedures and an excess set from non-compliant procedures that have exceeded the execution time. The compliant set and the excess set are then traversed and searched in the judicial case database to obtain the compliant association set and the excess association set associated with the compliant set and the excess set, respectively. It is also used to obtain valid association sets and invalid association sets. The data processing module is used to calculate the impact value of judicial cases, the first evaluation value of cases to be processed, the second evaluation value of cases to be processed, and the comprehensive evaluation value of cases to be processed. The case database module is used to store the case in the judicial case database if the comprehensive evaluation value of the case to be processed is greater than the average comprehensive evaluation value of the judicial cases stored in the judicial case database; otherwise, it submits the case to the superior court for review.

5. The judicial case quality evaluation system management system based on multi-source data according to claim 4, characterized in that: The data acquisition module further includes an acquisition unit, a decomposition unit, and an encoding unit; The acquisition unit is used to acquire the text information of the case to be processed and to identify the text information of the case. The decomposition unit is used to split the case text information into an entity part and a procedural part; the entity part is the factual part of the case, and the procedural part is the necessary procedures for handling the case; it is used to strip the coded judicial cases of social standard influencing factors, the social standard influencing factors being public opinion heat and public evaluation value; the public opinion heat includes the increase in attention brought about by the dissemination of judicial cases and the attention drag-derived quantity brought about by the progress of the judicial cases themselves, the attention drag-derived quantity is at least one other point of contention derived from a certain ongoing point of contention in the judicial case, wherein the way in which at least one other point of contention is derived from a certain ongoing point of contention is recorded as a derivation method, and the number of all derivation methods is recorded as the attention drag-derived quantity; the public evaluation value is the degree of public acceptance of the judicial case judgment result; the acceptance degree is the ratio of the number of public acceptances of the judicial case judgment result to the total number of participants in the evaluation; The encoding unit is used to obtain tags for each entity part of the case to be processed according to tagging rules. The tagging rules involve generating a tag for each entity part in a judicial case database, where the judicial case database stores high-quality judicial cases that have received judicial case quality evaluation and recognition. All tags for the case to be processed are traversed and retrieved from the judicial case database, and all retrieved judicial cases are statistically analyzed. All retrieved judicial cases are then encoded in a unified manner according to their case numbers. Based on the encoding, the retrieval results for any tag y of the case to be processed are generated into a set, denoted as y = {I1, I2, ..., I...}. n }, where I n This indicates the nth encoding retrieved for the tag y.

6. The judicial case quality evaluation system management system based on multi-source data according to claim 5, characterized in that: The association set module also includes an association set generation unit and an association set calculation unit; The association set generation unit is used to obtain the program parts of the case to be processed after the case information text recognition and segmentation, obtain the execution time of each program, and if the execution time of a program is within the specified time range required for program execution, then the program execution meets the standard; otherwise, the program execution does not meet the standard. The compliant programs are generated into a compliant set, and the non-compliant programs are generated into an excess set that exceeds the execution time. The compliant set and the excess set are respectively traversed and searched in the judicial case database to obtain the compliant association set and the excess association set associated with the compliant set and the excess set, respectively. The association set calculation unit is used to denot the compliant set as λ1, the compliant association set as λ2, the excess set as β1, and the excess association set as β2. If any non-compliant procedure exists in the compliant association set, then the non-compliant procedure is marked as a valid association; if any compliant procedure exists outside the association set, then the compliant procedure is marked as an invalid association. Obtain the valid association set and the invalid association set respectively, where the valid association set φ1=β1∩λ2 and the invalid association set φ2=λ1∩β2.

7. The judicial case quality evaluation system management system based on multi-source data according to claim 6, characterized in that: The data processing module further includes an impact value calculation unit, a first evaluation value calculation unit, a second evaluation value calculation unit, and a comprehensive evaluation value calculation unit; The impact value calculation unit is used to strip the social standard impact factors from any coded judicial case and obtain the change in attention increment and the change in recognition under any derivative method. Used to obtain the impact value of coded judicial cases based on social standard influencing factors. , among which, E(I i ) represents the impact value of any coded judicial case, I i ϵ{I1,I2,...,I n }, where N represents the total number of derivative methods, and d j G represents the change in the acceptance of the j-th derivative method. j This represents the change in the attention increment for the j-th derivative method; The first evaluation value calculation unit is used to preset the impact value threshold of coded judicial cases, and mark coded judicial cases that exceed the threshold as related cases; obtain the label value B(y)=a / n of any label set, where a represents the number of related cases in the label set; and calculate the first evaluation value of the case to be processed based on the label value. Where V1 represents the first evaluation value of the case to be processed, and Y represents the set of labels of the case to be processed; The second evaluation value calculation unit is used to calculate the second evaluation value of the case to be processed. Where V2 represents the second evaluation value of the case to be processed, M(φ1) represents the number of sets of valid association sets φ1, M(λ1) represents the number of sets of sets of compliant sets λ1, M(φ2) represents the number of sets of invalid association sets φ2, and M(β1) represents the number of sets of sets exceeding set λ1. The comprehensive evaluation value calculation unit is used to obtain the comprehensive evaluation value V = V1 × V2 of the case to be processed based on the first evaluation value V1 and the second evaluation value V2 of the case to be processed.

Images