Legal business management distribution system and method

Abstract

The invention discloses a legal business management and distribution system and method, relates to the technical field of legal business management, and solves the technical problems of time consumption, low efficiency and easiness in making mistakes when legal cases are distributed manually at present. The method comprises the following steps: analyzing the type of a to-be-processed case according to case information of the to-be-processed case; extracting a standard complexity attribute set of the case, and analyzing the complexity level of the to-be-processed case; determining a category attribute of the to-be-processed case based on the type and the complexity; analyzing the processing capability of lawyers for processing cases with the same category attribute as the to-be-processed case to obtain matched lawyers of the to-be-processed case; evaluating the workload of the current matched lawyers, and allocating the lawyers to the to-be-processed case; according to the method, refined classification of the law cases is realized, and reasonable distribution of the cases is also realized, so that lawyers can be promoted; the operation efficiency and the service quality of the office are improved, and the satisfaction degree of customers is improved.

Description

Technical Field

[0001] The present invention belongs to the field of legal business management, and specifically relates to a legal business management distribution system and method. Background Art

[0002] With the rapid development of social economy and the popularization of legal awareness, the demand for legal services is growing, and the number of cases is rising. Legal services cover a variety of cases, including contract disputes, intellectual property disputes, labor disputes, criminal defense, etc. Each case has its own specific legal terms and handling procedures. There is an uneven distribution of lawyer resources in different regions and different professional fields, resulting in a shortage of lawyer resources in some fields and a relative surplus in some fields.

[0003] Traditional legal business management allocation methods mainly rely on manual judgment and experience allocation, which may lead to uneven quality of case handling and affect customer satisfaction.

[0004] Legal cases are of many types and each has its own complexity, and each lawyer usually specializes in a different field of law. When a law firm is faced with a large number of legal case commissions, the traditional practice is to first go through a cumbersome case filing process, and then rely on manual labor to manually assign the case to the appropriate lawyer. This allocation process is not only time-consuming and inefficient, but it is also often difficult to ensure the accuracy of the allocation. What is even more difficult is that due to the subjectivity and uncertainty in the allocation process, cases often need to be reallocated, which undoubtedly further increases the operating burden and cost of the law firm. In short, the current manual case allocation method is time-consuming, inefficient, and error-prone, and is in urgent need of improvement. Therefore, the present invention provides a legal business management allocation system and method. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a legal business management allocation system and method for solving the technical problem that the current manual allocation of legal cases is time-consuming, inefficient, and prone to errors.

[0006] To achieve the above-mentioned purpose, the first aspect of the present invention provides a legal business management allocation system and method, including a case processing module and a case matching module;

[0007] Case classification module: used to determine the category attributes of the cases to be processed; the category attributes include type and complexity level;

[0008] Analyze the types of pending cases according to the case information of the pending cases; wherein the case information includes the title and cause of the case;

[0009] Extract the standard complexity attribute set of the case and analyze the complexity level of the case to be processed;

[0010] Case matching module: used to analyze the ability of lawyers to handle cases of the same category and attributes as the pending case, and obtain matching lawyers for the pending case;

[0011] Assess the workload of currently matched attorneys and assign attorneys to pending cases.

[0012] Preferably, the determining of the category attributes of the case to be processed includes:

[0013] The cases to be processed are first classified into the primary category and then into the secondary category to determine the type of the cases to be processed;

[0014] Based on the standard complexity attribute set, analyze the complexity ratio of pending cases; classify pending cases into their corresponding complexity levels; among which, the complexity attribute is an evaluation indicator of the difficulty of case handling;

[0015] The pending cases are divided into corresponding case groups; the case group structure is divided into primary, secondary and tertiary categories; the tertiary category is divided according to complexity.

[0016] The present invention determines the type of case by classifying it into primary and secondary categories, and then divides the case into more detailed categories by analyzing the complexity level, thereby achieving fine division of cases, which not only makes case management more orderly and efficient, but also provides a solid basis for the allocation of subsequent cases. Managers can more reasonably allocate human resources, time resources and other related resources according to the complexity and type of the case, ensuring that each case can be handled in the most appropriate and effective way.

[0017] Preferably, the case group structure is primary, secondary and tertiary; the secondary category is a subordinate category of the primary category, and the tertiary category is a subordinate category of the secondary category; the secondary category is further subdivided into types based on the primary category, and the tertiary category is subdivided into complexity based on the secondary category.

[0018] Preferably, the case to be processed is first classified into a primary category, and then classified into a secondary category to determine the type of the case to be processed, including:

[0019] Extract the title and cause of action of pending cases;

[0020] Extracting a number of keywords from the title of the case to be processed to obtain a title keyword sequence; and extracting keywords of the first-level category to obtain a number of first-level category keyword sequences; calculating the cosine similarity between the title keyword sequence and each first-level keyword sequence to obtain a number of cosine similarities between the two, and if the cosine similarity is greater than a similarity threshold, the first-level category is matched with the first-level category;

[0021] Extract the second-level categories to which the matching first-level category belongs, and perform semantic similarity analysis on the case cause of the pending case and the overview content of each second-level category to obtain the semantic similarity between the two. If the semantic similarity is greater than the semantic similarity threshold, the second-level category is the matching second-level category;

[0022] The pending cases are classified into the matching secondary category under the matching primary category.

[0023] The present invention extracts the title and the cause of action from the pending cases. The title is a brief summary of the case content, and the cause of action clearly points out the main nature and controversial focus of the case. The two together constitute an important basis for case classification. By extracting keywords from the title, a title keyword sequence is formed, and a primary category keyword sequence is further refined. This step can quickly narrow the possible classification range of the case and provide strong support for subsequent classification work. By calculating the cosine similarity between the title keyword sequence and the primary category keyword sequence, the degree of association between the case and each primary category can be objectively evaluated, thereby determining the matching primary category; after determining the matching primary category, the method further extracts all secondary categories under the primary category, and performs semantic similarity analysis on the cause of action of the pending case and the overview content of these secondary categories. This step fully considers the complexity and diversity of the cause of action of the case, and through a deep comparison at the semantic level, the secondary category that best fits the case can be found more accurately; finally, the pending case is successfully classified into the matching secondary category under the matching primary category, which improves the accuracy of case type classification.

[0024] Preferably, the analysis of the complexity ratio of the cases to be processed based on the standard complexity attribute set includes:

[0025] Input the standard complexity attribute set and the case to be processed into the semantic analysis model, and the semantic analysis model outputs the complexity attributes of the case to be processed; wherein the semantic analysis model is built based on the BERT model;

[0026] The number of complexity attributes of the cases to be processed is counted, and the ratio is processed with the number of elements contained in the standard complexity attribute set to obtain the complexity ratio of the cases to be processed.

[0027] The semantic analysis model of the present invention is constructed based on the BERT model, making full use of the powerful capabilities of the BERT model in the field of natural language processing. The BERT model has accumulated rich language knowledge and context understanding capabilities through pre-training on a large-scale corpus, which enables it to perform well in processing complex and diverse case texts. Therefore, the semantic analysis model built based on BERT can more accurately capture the key information in the case text, thereby more accurately evaluating the complexity attributes of the case, improving the efficiency of case processing, and reducing the subjectivity and errors caused by human intervention.

[0028] Preferably, the complexity proportion is divided into different complexity levels based on a proportion threshold range.

[0029] Preferably, the analysis of the lawyer's ability to handle cases of the same category and attributes as the case to be handled includes:

[0030] Calculate the number of successful cases handled by each lawyer with the same category and attributes as the pending cases, marked as Pi, and extract the years of practice of each lawyer, marked as Si, and calculate the processing capacity coefficient of each lawyer in handling pending cases through the formula CLi = Pi / [Si×Σ(Pi / Si)]; where i is the lawyer's label, and Σ is the sum of I;

[0031] The processing capacity coefficient of each lawyer is compared with the preset coefficient. If it is greater than the preset coefficient, then the lawyer is the matching lawyer for the case to be handled.

[0032] The present invention can more comprehensively evaluate the actual ability and experience of lawyers in handling specific types of cases by comprehensively considering the number of successful cases and years of practice of lawyers in handling cases of the same category. The number of successful cases reflects the lawyer's winning rate or resolution effect in similar cases, which is an important indicator for evaluating his professional ability; while the years of practice reflect the lawyer's accumulation and depth of experience in the legal industry. Combining the two can more accurately portray the lawyer's potential in handling pending cases, screen out matching lawyers, and improve the accuracy and reliability of matching.

[0033] Preferably, the step of evaluating the workload of currently matched lawyers and assigning lawyers to pending cases includes:

[0034] Analyze the current workload of the matched lawyers and sort the matched lawyers according to their current workload; the lawyer ranked first is the lawyer with the heaviest workload;

[0035] Assign pending cases to matching attorneys with the lightest workload.

[0036] The present invention takes into account the current workload of the matching lawyers and ensures the rationality and fairness of case allocation. By sorting the workload of the matching lawyers, it is possible to clearly understand the current workload and stress of each lawyer, avoid the situation where the lawyer's heavy workload leads to reduced work efficiency or damaged case handling quality, and help optimize the efficiency of the use of lawyer resources; and avoid lawyers from being overworked and burnout.

[0037] Preferably, the current workload is determined based on the caseload that the lawyer currently needs to handle.

[0038] Preferably, the second aspect of the present invention provides a legal business management allocation method, comprising the following steps:

[0039] Step 1: Analyze the types and complexity levels of pending cases and classify them;

[0040] Step 2: Get matching lawyers for pending cases;

[0041] Step 3: Assess the workload of currently matched lawyers and assign lawyers to pending cases.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. The present invention determines the type of case based on the title and cause of the case to be processed, and divides the complexity level of the case to be processed through a semantic analysis model according to the standard complexity attribute of the case; based on the hierarchical division of case types and complexity levels, the case can be divided into detailed categories, which improves the orderliness and efficiency of case management and provides a reliable basis for the allocation of subsequent cases;

[0044] 2. The present invention integrates two dimensions: the number of successful cases that lawyers have handled in the same category as the pending case and their years of experience, so as to comprehensively and deeply evaluate the lawyers' practical ability and experience accumulation in specific case types. By taking these two indicators into consideration, it is possible to more accurately screen out the cases that the matching lawyers should handle, thereby improving the accuracy and reliability of the matching. In addition, the matching lawyers can be matched with suitable lawyers for the pending cases in combination with their workload, thus avoiding the situation where too many cases are assigned to the same lawyer, resulting in reduced work efficiency or impaired case handling quality.

[0045] 3. The present invention analyzes the types and complexity levels of pending cases and classifies the pending cases; obtains matching lawyers for the pending cases; evaluates the workload of the currently matched lawyers and assigns lawyers to the pending cases; and achieves refined management and reasonable allocation of legal services through effective classification of pending cases and evaluation of each lawyer's ability to handle pending cases, thereby promoting smoother and more efficient operations of law firms. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 This is a schematic diagram of the process of analyzing the types of cases to be processed in the present invention;

[0048] Figure 2It is a schematic diagram of the complexity level analysis process of the case to be processed in the present invention;

[0049] Figure 3 A schematic diagram of the matching lawyer assignment process for the present invention;

[0050] Figure 4 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The first aspect of the present invention provides a legal business management allocation system and method, including a case processing module and a case matching module;

[0053] The case classification module classifies the pending cases into first-level categories and then second-level categories according to the case information of the pending cases, and determines the types of the pending cases; wherein the case information includes the title and cause of the case;

[0054] It should be noted that the content of a legal case includes the case title, party information, cause of action, facts and reasons, request, list of evidence, and conclusion; the cause of action mainly briefly describes the nature and background of the case.

[0055] The first-level category is the major category of case types, such as civil cases, criminal cases, administrative cases, etc.; the second-level category is the subordinate classification of the first-level category, such as civil cases include contract disputes, tort disputes, inheritance disputes, etc.; criminal cases include intentional injury, theft, fraud, etc.; administrative cases include administrative penalties, administrative levies, administrative enforcement, etc.

[0056] See also Figure 1 Specifically, extract the title and cause of the case to be processed; extract several keywords from the title of the case to be processed to obtain a title keyword sequence; and extract the keywords of the first-level category to obtain several first-level category keyword sequences; calculate the cosine similarity between the title keyword sequence and each first-level keyword sequence respectively, and obtain several cosine similarities between the two. If the cosine similarity is greater than the similarity threshold, the first-level category is matched with the first-level category;

[0057] Extract the second-level categories to which the matching first-level category belongs, and perform semantic similarity analysis on the case cause of the pending case and the overview content of each second-level category to obtain the semantic similarity between the two. If the semantic similarity is greater than the semantic similarity threshold, the second-level category is the matching second-level category;

[0058] The pending cases are classified into the matching secondary category under the matching primary category.

[0059] For example, the title of a pending case is "Sales Contract Dispute", and the case is "The plaintiff Zhang San and the defendant Li Si have a dispute arising from the performance of the sales contract. They now file a lawsuit and request the court to make a judgment in accordance with the law." The keyword sequence extracted from the title is ["sales", "contract", "dispute"]; the keywords of each first-level category are extracted to obtain the first-level category keyword sequence, and the keyword sequence is subjected to cosine similarity analysis, where the first-level category corresponding to the similarity threshold is the matching first-level category of the pending case; the matching first-level category of the pending case is a civil case;

[0060] The subordinate secondary classifications of civil cases are extracted, and the case causes of the pending case are analyzed for semantic similarity with the overview contents of each secondary classification. The secondary classification corresponding to the semantic similarity greater than the semantic similarity threshold is matched as the secondary classification, i.e., contract dispute. The pending case belongs to the contract dispute category under civil cases.

[0061] And, extract the standard complexity attribute set of the case, analyze the complexity level of the pending cases; classify the pending cases based on the type and complexity level;

[0062] See also Figure 2 Specifically, the standard complexity attribute set and the case to be processed are input into the semantic analysis model, and the semantic analysis model outputs the complexity attributes of the case to be processed; wherein the semantic analysis model is constructed based on the BERT model;

[0063] The number of complexity attributes of the cases to be processed is counted, and the ratio is processed with the number of elements contained in the standard complexity attribute set to obtain the complexity ratio of the cases to be processed.

[0064] Classify pending cases into their corresponding complexity levels; compare the complexity ratio threshold range; if it is less than the ratio threshold range, it is level one complexity; if it is within the ratio threshold range, it is level two complexity; if it exceeds the ratio threshold range, it is level three complexity; complexity level: level one complexity < level two complexity < level three complexity.

[0065] Among them, the complexity attribute is an indicator of the difficulty of case handling;

[0066] The cases to be processed are divided into corresponding case groups; among them, the case group structure is divided into primary, secondary and tertiary categories; the tertiary category is divided according to complexity; the secondary category is a subordinate category of the primary category, and the tertiary category is a subordinate category of the secondary category; the secondary category is further subdivided into types based on the primary category, and the tertiary category is subdivided according to complexity based on the secondary category.

[0067] It should be noted that the standard complexity attributes include the diversity of legal issues, the complexity of facts, the complexity of legal relations, the historical background of the case, etc., which reflect the difficulty of handling legal cases, and these factors constitute the standard complexity attribute set of the case.

[0068] For example: collect several different types of cases and the complexity attributes of each case, use several different types of cases and standard complexity attribute sets as the input data of the BERT model, use the complexity attributes of the cases as the output data of the BERT model, and use the input data and output data to repeatedly train the BERT model to obtain a semantic analysis model.

[0069] Assuming that the type of the pending case has been determined to be a contract dispute of a civil case, the pending case is input into the semantic analysis model. If the complexity attribute of the pending case is output as "diversity of legal issues, complexity of facts, complexity of legal relations, and historical background of the case", the pending case has four complexity attributes. Assuming that the number of elements in the standard complexity attribute set is N, then 4 / N is the complexity ratio of the pending case, and is less than the ratio threshold range, then the pending case is classified into the first-level complexity case category under the contract dispute category.

[0070] The case matching module analyzes the lawyers’ ability to handle cases with the same attributes as the pending case, and obtains the matching lawyers for the pending case;

[0071] See also Figure 3 Specifically, the number of successful cases handled by each lawyer with the same category and attributes as the pending case is calculated, marked as Pi, and the years of practice of each lawyer is extracted, marked as Si, and the processing capacity coefficient of each lawyer in handling pending cases is calculated by the formula CLi = Pi / [Si×Σ(Pi / Si)]; where i is the lawyer's label, and Σ is the sum of I;

[0072] It should be noted that Pi / Si is the number of cases successfully handled by each lawyer each year; Σ(Pi / Si) is the total number of cases successfully handled by all lawyers each year; the ratio of the two is used to obtain the processing capacity coefficient CLi of each lawyer in handling pending cases.

[0073] The processing capacity coefficient of each lawyer is compared with the preset coefficient. If it is greater than the preset coefficient, then the lawyer is the matching lawyer for the case to be handled.

[0074] For example: suppose a law firm has 4 lawyers, namely, Lawyer A, Lawyer B, Lawyer C, and Lawyer D; each lawyer handles civil cases of contract disputes with a level of complexity of level 1. Lawyer A has 10 successful cases in 2 years of work, with an average of 5 successful cases per year; Lawyer B has 60 successful cases in 4 years of work, with an average of 15 successful cases per year; Lawyer C has 100 successful cases in 5 years of work, with an average of 20 successful cases per year; Lawyer D has 180 successful cases in 6 years of work, with an average of 30 successful cases per year; a total of 105 successful cases per year;

[0075] Lawyer A's processing capacity coefficient CLA = 5 / 105 = 0.05; Lawyer B's processing capacity coefficient CLB = 15 / 105 = 0.14; Lawyer C's processing capacity coefficient CLC = 20 / 105 = 0.19; Lawyer D's processing capacity coefficient CLD = 30 / 105 = 0.29; if the preset coefficient is set to 0.1, lawyers B, C, and D are the matching lawyers for the case to be processed.

[0076] The workload of the lawyer is determined based on the number of cases that the matching lawyer currently needs to handle. The more cases there are, the greater the workload. The cases to be handled are assigned to the matching lawyer with the smallest workload.

[0077] For example: Currently, the number of cases to be handled by lawyers B, C, and D are 50, 30, and 80, respectively. Lawyer C has the smallest workload, so the pending cases will be assigned to him.

[0078] See also Figure 4 The second aspect of the present invention provides a legal business management allocation method, comprising the following steps:

[0079] Step 1: Analyze the types and complexity levels of pending cases and classify them;

[0080] Step 2: Get matching lawyers for pending cases;

[0081] Step 3: Assess the workload of currently matched lawyers and assign lawyers to pending cases.

[0082] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0083] The working principle of the present invention is as follows: according to the case information of the pending cases, the types of the pending cases are analyzed; the standard complexity attribute set of the cases is extracted to analyze the complexity level of the pending cases; the category attributes of the pending cases are determined based on the types and complexity; the processing capabilities of lawyers in processing cases with the same category attributes as the pending cases are analyzed to obtain matching lawyers for the pending cases; the workload of the current matching lawyers is evaluated to assign lawyers to the pending cases.

[0084] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A legal business management distribution system and method, characterized in that: Including case processing module and case matching module; Case classification module: used to determine the category attributes of the cases to be processed; the category attributes include type and complexity level; Analyze the types of pending cases according to the case information of the pending cases; wherein the case information includes the title and cause of the case; Extract the standard complexity attribute set of the case and analyze the complexity level of the case to be processed; Case matching module: used to analyze the ability of lawyers to handle cases of the same category and attributes as the pending case, and obtain matching lawyers for the pending case; Assess the workload of currently matched attorneys and assign attorneys to pending cases.

2. A legal business management distribution system and method according to claim 1, characterized in that: Determining the category attributes of the case to be processed includes: The cases to be processed are first classified into the primary category and then into the secondary category to determine the type of the cases to be processed; Based on the standard complexity attribute set, analyze the complexity ratio of pending cases; classify pending cases into their corresponding complexity levels; among which, the complexity attribute is an evaluation indicator of the difficulty of case handling; The pending cases are divided into corresponding case groups; the case group structure is divided into primary, secondary and tertiary categories; the tertiary category is divided according to complexity.

3. A legal business management distribution system and method according to claim 2, characterized in that: The case category group structure is divided into primary category, secondary category and tertiary category; the secondary category is a subordinate category of the primary category, and the tertiary category is a subordinate category of the secondary category; the secondary category is further subdivided into types based on the primary category, and the tertiary category is subdivided into complexity based on the secondary category.

4. A legal business management distribution system and method according to claim 2, characterized in that: The cases to be processed are first classified into the primary category and then classified into the secondary category to determine the type of the cases to be processed, including: Extract the title and cause of action of pending cases; Extracting a number of keywords from the title of the case to be processed to obtain a title keyword sequence; and extracting keywords of the first-level category to obtain a number of first-level category keyword sequences; calculating the cosine similarity between the title keyword sequence and each first-level keyword sequence to obtain a number of cosine similarities between the two, and if the cosine similarity is greater than a similarity threshold, the first-level category is matched with the first-level category; Extract the second-level categories to which the matching first-level category belongs, and perform semantic similarity analysis on the case cause of the pending case and the overview content of each second-level category to obtain the semantic similarity between the two. If the semantic similarity is greater than the semantic similarity threshold, the second-level category is the matching second-level category; The pending cases are classified into the matching secondary category under the matching primary category.

5. A legal business management distribution system and method according to claim 2, characterized in that: The complexity ratio of the cases to be processed is analyzed based on the standard complexity attribute set, including: Input the standard complexity attribute set and the case to be processed into the semantic analysis model, and the semantic analysis model outputs the complexity attributes of the case to be processed; wherein the semantic analysis model is built based on the BERT model; The number of complexity attributes of the cases to be processed is counted, and the ratio is processed with the number of elements contained in the standard complexity attribute set to obtain the complexity ratio of the cases to be processed.

6. A legal business management distribution system and method according to claim 5, characterized in that: The complexity ratio is divided into different complexity levels based on a ratio threshold range.

7. A legal business management distribution system and method according to claim 1, characterized in that: The analysis of the lawyer’s ability to handle cases of the same category and attributes as the pending case includes: Calculate the number of successful cases handled by each lawyer with the same category and attributes as the pending cases, marked as Pi, and extract the years of practice of each lawyer, marked as Si, and calculate the processing capacity coefficient of each lawyer in handling pending cases through the formula CLi = Pi / [Si×Σ(Pi / Si)]; where i is the lawyer's label, and Σ is the sum of I; The processing capacity coefficient of each lawyer is compared with the preset coefficient. If it is greater than the preset coefficient, then the lawyer is the matching lawyer for the case to be handled.

8. A legal business management distribution system and method according to claim 7, characterized in that: The assessment of the workload of currently matched lawyers and the allocation of lawyers to pending cases include: Analyze the current workload of the matched lawyers and sort the matched lawyers according to their current workload; the lawyer ranked first is the lawyer with the heaviest workload; Assign pending cases to matching attorneys with the lightest workload.

9. A legal business management distribution system and method according to claim 8, characterized in that: The current workload is determined based on the caseload that the attorney is currently required to handle.

10. A method for allocating legal business management, based on a legal business management allocation system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Analyze the types and complexity levels of pending cases and classify them; Step 2: Get matching lawyers for pending cases; Step 3: Assess the workload of currently matched lawyers and assign lawyers to pending cases.

Images