Mediation case allocation method and device based on multi-dimensional evaluation and electronic equipment

Abstract

The invention provides a mediation case allocation method and device based on multi-dimensional evaluation and electronic equipment, and relates to the technical field of case allocation, and the method comprises the steps: firstly determining the case characteristics of a current to-be-mediated case according to the pre-obtained attribute information of a to-be-allocated case, then, based on historical case records and preset attention features in the case features, determining feature indexes and corresponding scores of each dimension of the mediator, and based on the case features and the feature indexes and corresponding scores of each dimension of the mediator, determining a matching degree score of the current case to be mediated and the mediator; the matching degree score is used for generating a case allocation strategy. Through the method, the technical problems of low case allocation efficiency and inaccurate mediation demand matching in the prior art are solved, and the effects of improving the case allocation efficiency and improving the matching accuracy are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of case allocation, in particular to a mediation case allocation method and device based on multi-dimensional evaluation and electronic equipment. BACKGROUND

[0002] With the rapid development of social economy and the deepening of the rule of law, mediation, as an important part of the non-litigation dispute resolution mechanism, plays an increasingly prominent role. In the prior art, the mediator resources are mainly managed by manual allocation, and the specific process includes: after the dispute parties submit a mediation application, the mediation institution staff manually selects and matches according to the basic information such as case type and mediator professional field, and then notifies the matching result to the parties.

[0003] This traditional matching method highly depends on the experience judgment of the staff, and the matching process is time-consuming and difficult to achieve precise docking. Especially in complex scenarios such as cross-regional mediation and professional dispute mediation, due to the lack of in-depth analysis of the characteristics of the case and the specialities of the mediator, the matching degree between the case and the mediator is often not high, which affects the mediation efficiency and success rate. At the same time, with the continuous increase in the number of mediation cases and the diversification of case types, the traditional manual matching mode has been difficult to meet the efficient and precise mediation needs. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a mediation case allocation method and device based on multi-dimensional evaluation and electronic equipment, to solve the technical problems of low case allocation efficiency and inaccurate mediation demand matching in the prior art.

[0005] In a first aspect, the embodiments of the present application provide a mediation case allocation method based on multi-dimensional evaluation, which comprises: determining the case characteristics of the current mediation case according to the attribute information of the to-be-allocated case obtained in advance; determining the characteristic indicators and corresponding scores of each dimension of the mediator based on the historical case records and the pre-set attention characteristics in the above case characteristics; determining the matching degree score of the current to-be-mediated case and the mediator based on the above case characteristics and the characteristic indicators and corresponding scores of each dimension of the mediator; the matching degree score is used to generate a case allocation strategy.

[0006] In some optional implementations, the above case characteristics include case type, complexity, urgency, region and other attribute characteristics; the way of obtaining the attribute information of the to-be-allocated case includes: performing keyword extraction based on the case description text to obtain the attribute information of the current to-be-mediated case; the above attribute information includes: regional information, dispute type information, amount information, party information and historical mediation information; Determine the case characteristics of the current case to be mediated according to the attribute information of the case to be allocated obtained in advance, including: inputting the above dispute type information into a pre-trained mediation case classification model to determine the case type of the current case; determining the complexity of the current case according to the above amount of money information, the above party information and the above historical mediation information; determining the urgency of the current case according to the above case type.

[0007] In some optional implementations, the determination of the characteristic indicators and the corresponding scores of each dimension of the mediator based on the historical case records and the preset attention characteristics in the case characteristics includes: determining the attention characteristics in the case characteristics in advance; the attention characteristics are at least one of the complexity, the urgency and the region; generating a mediator capability evaluation parameter set based on the historical case records, the mediator capability evaluation parameter set including the characteristic indicators and the corresponding initial scores of each dimension of the mediator; the characteristic indicators of each dimension include: mediation efficiency, mediation success rate, professionalism of different case types, satisfaction score and current load; dynamically adjusting the evaluation weights of the characteristic indicators of each dimension based on the attention characteristics; determining the corresponding scores of the characteristic indicators of each dimension of the mediator and the comprehensive score of the mediator based on the evaluation weights and the initial scores.

[0008] In some optional implementations, the determination of the matching degree score of the current case to be mediated and the mediator based on the case characteristics and the characteristic indicators and the corresponding scores of each dimension of the mediator includes: performing feature vectorization processing on the case characteristics to generate a case feature vector; performing multi-dimensional matching calculation on the case feature vector and the corresponding scores of the characteristic indicators of each dimension of the mediator by using a preset matching algorithm; and calculating the matching degree score of the current case to be mediated and the mediator by weighting according to the multi-dimensional matching results and a preset matching weight.

[0009] In some optional implementations, the determination of the matching degree score of the current case to be mediated and the mediator based on the case characteristics and the characteristic indicators and the corresponding scores of each dimension of the mediator further includes: if the current load of the mediator reaches a preset maximum case load threshold, determining the matching degree score of the current case to be mediated and the mediator as 0; and determining the matching degree score of the current case to be mediated and the mediator as 0 according to the party information and the association relationship of the mediator.

[0010] In some optional implementations, the method further includes: determining the matching degree scores of the current case to be mediated and a plurality of mediators, and sorting the matching degree scores from high to low; and generating a case allocation strategy based on the sorting result of the matching degree scores.

[0011] In some optional implementations, the method further includes: generating mediation feedback data according to a mediation result of the current case; the mediation feedback data includes: whether mediation is successful, mediation duration, and mediation satisfaction; updating a state value function of the preset reinforcement learning model based on the mediation feedback data; and dynamically adjusting a preset matching weight according to the updated state value function; the preset matching weight is used to determine a matching degree score of the current case to be mediated and the mediator.

[0012] In a second aspect, an embodiment of the present application provides a mediation case allocation device based on multi-dimensional evaluation, which comprises: A case feature determination module is configured to determine a case feature of a current case to be mediated according to attribute information of a to-be-allocated case obtained in advance; A mediator evaluation module is configured to determine a feature index and a corresponding score of each dimension of a mediator based on a historical case record and a preset attention feature in the case feature; A matching module is configured to determine a matching degree score of the current case to be mediated and the mediator based on the case feature and the feature index and the corresponding score of each dimension of the mediator; and the matching degree score is used to generate a case allocation strategy.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program capable of running on the processor, and the processor implements the steps of the method of any one of the first aspect when executing the computer program.

[0014] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores computer executable instructions, and the computer executable instructions cause the processor to run the method of any one of the first aspect when the computer executable instructions are called and run by the processor.

[0015] The present application provides a mediation case allocation method and device based on multi-dimensional evaluation and an electronic device, the method comprising: first determining a case feature of a current case to be mediated according to attribute information of a to-be-allocated case obtained in advance, then determining a feature index and a corresponding score of each dimension of a mediator based on a historical case record and a preset attention feature in the case feature, and finally determining a matching degree score of the current case to be mediated and the mediator based on the case feature and the feature index and the corresponding score of each dimension of the mediator, which is used to generate a case allocation strategy. Through the method, the technical problems of low case allocation efficiency and insufficient precision of mediation demand matching in the prior art are solved, and the effects of improving case allocation efficiency and improving matching precision are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 A flowchart of a mediation case allocation method based on multi-dimensional evaluation provided by the embodiment of the present application is shown in the figure. Figure 2 A structural diagram of a mediation case allocation device based on multi-dimensional evaluation provided by the embodiment of the present application is shown in the figure. Figure 3 A structural diagram of a visual scheduling management system of a mediation case allocation device based on multi-dimensional evaluation provided by the embodiment of the present application is shown in the figure. Figure 4 A structural diagram of an electronic device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] The technical solutions in the embodiments of the present application will be described in combination with the drawings in the embodiments of the present application.

[0020] In order to facilitate the understanding of the present embodiment, first, a mediation case allocation method based on multi-dimensional evaluation disclosed by the embodiment of the present application will be introduced in detail, referring to Figure 1 A flowchart of a mediation case allocation method based on multi-dimensional evaluation is shown in the figure, which can be executed by an electronic device, mainly including the following steps S110 to S130: S110: determining the case characteristics of the current mediation case according to the attribute information of the to-be-allocated case obtained in advance; The attribute information of the case to be allocated can be obtained through natural language processing (NLP), for example, in an embodiment, keyword extraction can be performed based on the case description text to obtain the attribute information of the current case to be mediated; the attribute information can include regional information, dispute type information, amount information, party information, historical mediation information, etc.

[0021] The case description text can be a textual description of the case to be allocated, such as a case summary, or information extracted based on case documents.

[0022] As a specific example, the keyword extraction of the case description text to obtain the case attribute information can include: first, performing word segmentation processing on the text and identifying entities related to the case (such as region, dispute type, involved amount, party identity, number of parties, number of historical mediations, etc.); then, converting unstructured or semi-structured text information into structured data that can be used for subsequent processing (such as classification, prediction, recommendation, etc.).

[0023] In an embodiment, the case features can include case type, complexity, urgency, region, and other attribute features; the other attribute features can include involved amount features, party identity features, number of parties features, number of historical mediations features, etc. Further, in the above S110, the manner of determining the case features of the current case to be mediated can include: (S111) inputting the dispute type information into a pre-trained mediation case classification model to determine the case type of the current case; The pre-trained mediation case classification model can be a classification model trained based on historical data, such as support vector machine (SVM), bidirectional encoder model (BERT), etc. The type of the current case can be labor dispute, contract dispute, family mediation, neighbor dispute, consumer complaint, etc. (S112) determining the complexity of the current case according to the amount information, party information and historical mediation information; The purpose of this step is to quantitatively evaluate the complexity of the current case, and the model used for complexity calculation can be a rule-based weighted scoring model or a machine learning-based complexity prediction model.

[0024] As a specific example, using a rule-based weighted scoring model for complexity calculation, each attribute information (and corresponding sub-item dimension) can be assigned a certain weight, and then the weighted sum is obtained to obtain the complexity score.

[0025] For example, the amount of money information can include "the amount of money involved is greater than 1 million" (weight value is 0.2), "the amount of money is controversial" (weight value is 0.1), etc. Sub-item dimensions; Party information can include "multi-party dispute (not less than 3 parties)" (weight value is 0.15), "cross-country / cross-province" (weight value is 0.1), "minor" (weight value is 0.3), etc. Sub-item dimensions; Historical mediation information can include "the number of historical mediation times of the current case is greater than 3 times" (weight value is 0.1), etc. Sub-item dimensions. Taking the case where the total complexity score is 1 as an example, it can be divided into: low complexity (<0.4), medium complexity (0.4-0.7), and high complexity (>0.7).

[0026] As another specific example, using a machine learning-based complexity prediction model for complexity calculation, training data can be generated according to historical case data, each record containing the sub-item dimensions corresponding to the above attribute information and a pre-labeled complexity label; Then the above fields are encoded, standardized, normalized, etc. After processing, input into the complexity prediction model for prediction, output the complexity level (such as low, medium, high) or complexity score (such as 0-1) of the current case.

[0027] Among them, the complexity prediction model can choose a classification model (such as: decision tree, random forest, XGBoost, LightGBM, neural network, etc.) to predict the complexity level, or a regression model (such as: linear regression, support vector regression, neural network regression, etc.) to predict the complexity value.

[0028] (S113) Determine the urgency of the current case according to the case type.

[0029] In the process of handling the mediation process, the determination of the urgency of the case is an important link to ensure the reasonable allocation of resources, protect the rights and interests of the parties, and improve the efficiency of mediation. The purpose of this step is to automatically identify the urgency of the case based on the case type, and arrange the mediation priority accordingly. For example, disputes involving domestic violence, labor disputes, etc. Need to be handled first. In a specific implementation, the case type can be matched with the pre-set "case type-urgency" mapping rule, and according to the matching result, the corresponding urgency level (such as low, medium, high) and the corresponding urgency score (such as 0-1) are assigned to the case.

[0030] As a specific example, Table 1 shows a case type-urgency mapping rule: Table 1-Case type-urgency mapping rule;

[0031] The static rule table described above is applicable to the case where the emergency rule is relatively fixed. In some other embodiments, a dynamic strategy engine can also be used to dynamically adjust the emergency degree in combination with the case characteristics (such as whether it involves personal safety, whether it has been reported, whether it affects livelihood, etc.) through a rule engine (such as Drools) or a machine learning model.

[0032] Further, the above-mentioned multiple features can be converted into a unified dimensional numerical vector for subsequent algorithm model application. Specifically, the region and case type can be encoded respectively, and the complexity score and the emergency degree score calculated are spliced to obtain a feature vector, for example: feature vector = [region code, case type code, complexity score, emergency degree score]. In addition, the feature vector can also include the corresponding attribute feature code obtained by encoding other attribute features (such as: amount of money involved feature, party identity feature, number of parties feature, historical mediation times feature), which will not be described here. Through the above steps, the case information is parsed, the case features are extracted, and data support is provided for subsequent case allocation.

[0033] S120: determining the feature indicators and corresponding scores of each dimension of the mediator based on the historical case records and the preset attention features in the case features; The historical case records can be the records of the cases mediated by the current mediator; the preset attention features in the case features can be any one of the features pre-determined in the case feature dimensions, i.e., the direction that the case to be mediated is more concerned about, for example: the efficiency of the emergency case is more concerned about, the success rate of the complex case is more concerned about, etc.

[0034] That is, in one embodiment, the above step S120 can include: pre-determining the attention features in the case features; the attention features are at least one of the complexity, the emergency degree, and the region. The preset attention features can be used to dynamically adjust the weights of the feature indicators of each dimension of the mediator, for example, for the emergency case, the weight of the mediation efficiency of the mediator can be increased to improve the possibility of priority matching to the mediator with high efficiency, thereby helping to select a more suitable mediator and improving the matching accuracy of the case and the mediator.

[0035] In one embodiment, in the above S120, the step of determining the feature indicators and corresponding scores of each dimension of the mediator based on the historical case records and the preset attention features in the case features can further include: (S121) generating a mediator capability evaluation parameter set based on historical case records, the mediator capability evaluation parameter set can include characteristic indicators of each dimension of the mediator and corresponding initial scores; the characteristic indicators of each dimension include: mediation efficiency, mediation success rate, professionalism of different case types, satisfaction score, and current load; the purpose of this step is to build a mediator portrait, quantify the mediation capability, and generate corresponding scores of each dimension indicator as evaluation parameters for subsequent optimal matching.

[0036] As a specific example, the evaluation dimensions of the mediator and the calculation method of the corresponding indicators can be seen from Table 2: Table 2 - Calculation method of evaluation dimensions and corresponding indicators of mediators

[0037] Among them, the professional field matching degree can be a plurality of values corresponding to different case types; the calculation of historical success rate, mediation efficiency and party satisfaction can be a comprehensive calculation based on the total number of historical cases handled by the mediator, or can be calculated according to different case types.

[0038] Further, based on the calculation method of the generated numerical value of the corresponding indicators, and the initial evaluation weight of each evaluation dimension and the corresponding indicators, the initial score of each dimension characteristic indicator is generated. In some embodiments, a scoring model (such as weighted average, decision tree, regression model, etc.) can be used to generate the initial score of each dimension characteristic indicator.

[0039] (S122) dynamically adjusting the evaluation weight of each dimension characteristic indicator based on the attention feature; Preferably, the above-mentioned attention feature can be at least one of the complexity, urgency, and region pre-determined in the case feature. The determined attention feature can be used to dynamically adjust the above-mentioned initial evaluation weight, and the allocation of the evaluation weight is optimized by using a machine learning model (such as an optimized distributed gradient boosting library XGBoost).

[0040] In other embodiments, the above-mentioned attention feature can also be a case type or other attribute feature. The other attribute feature can be: an amount of money involved feature, a party identity feature, a number of disputing parties feature, a historical mediation times feature, etc.

[0041] For example, the pre-determined attention feature in the case feature is complexity. For a case with high complexity (complexity score greater than 0.7), the evaluation weight of the mediation success rate in the characteristic indicator can be appropriately increased; or, the pre-determined attention feature is the party identity feature (the party is a minor), then the evaluation weight of the mediation efficiency and / or the mediation success rate in the characteristic indicator can be appropriately increased.

[0042] (S123) Based on the evaluation weights and the initial scores, the corresponding scores of the characteristic indicators of each dimension of the mediator and the comprehensive score of the mediator are determined.

[0043] According to the dynamically adjusted evaluation weights and the initial scores of the characteristic indicators of each dimension of the mediator, the corresponding scores of the characteristic indicators of each dimension (i.e., the product of the initial score and the dynamic weight) and the comprehensive score of the mediator (the sum of the weighted scores of each dimension) are calculated.

[0044] S130: Based on the case characteristics and the characteristic indicators and corresponding scores of each dimension of the mediator, the matching degree score of the current case to be mediated and the mediator is determined; the matching degree score is used to generate a case allocation strategy.

[0045] In one embodiment, in the above S130, the step of determining the matching degree score of the current case to be mediated and the mediator based on the case characteristics and the characteristic indicators and corresponding scores of each dimension of the mediator can include: first, performing feature vectorization processing on the case characteristics to generate a case feature vector; for example, encoding the region, case type, and other attribute characteristics in the above case characteristics to generate region encoding, case type encoding, and other attribute characteristic encoding, and splicing the calculated complexity score and urgency score to generate a case feature vector. For example: feature vector = [region encoding, case type encoding, complexity score, urgency score].

[0046] Then, a preset matching algorithm is used to perform multi-dimensional matching calculation on the case feature vector and the corresponding scores of the characteristic indicators of each dimension of the mediator; and then, according to the matching results of each dimension and the preset matching weight, the matching degree score of the current case to be mediated and the mediator is calculated by weighted calculation.

[0047] The purpose of the above steps is to realize intelligent matching between the case to be mediated and the mediator by constructing a multi-dimensional feature matching model between the case and the mediator. This matching process not only considers the characteristic information of the case, but also comprehensively evaluates the professional ability, processing efficiency, and case load of the mediator in multiple dimensions, thereby improving the mediation efficiency and matching accuracy.

[0048] In some embodiments, the preset matching algorithm can be a defined matching degree function (for example, cosine similarity or Euclidean distance) for calculating the similarity or matching degree of each pair of case and mediator in each feature dimension.

[0049] In an embodiment, the preset matching weight can be a weight pre-assigned to each feature dimension according to the demand for mediation services or analysis of historical mediation cases. Further, the preset matching weight can be optimized by a machine learning model, for example, a matching weight model (such as XGBoost, LightGBM or neural network) is trained based on historical mediation data, and the trained matching weight model is used to generate the preset matching weight.

[0050] In an embodiment, the determination of the matching degree score of the current mediation case and the mediator based on the case features and the feature indicators and corresponding scores of each dimension of the mediator in S130 can further include: If the current load of the mediator reaches the preset maximum case load threshold, the matching degree score of the current mediation case and the mediator is determined to be 0. That is, in order to avoid overload allocation, a maximum case number threshold for a single mediator is set, and the current load of the mediator is judged. If the mediator is fully loaded (reaches the maximum case number), the matching degree score of the current case and the mediator is directly assigned to 0.

[0051] Preferably, in some embodiments, the preset matching weight can be adjusted according to the current load of the mediator to automatically reduce the matching priority of the mediator with a large case backlog. For example, the preset maximum case load threshold is 10, the case backlog of mediator A is 10, i.e. the current load is 100%, the preset matching weight value corresponding to the current load of mediator A can be adjusted to 0, the case backlog of mediator B is 5, i.e. the current load is 50%, the preset matching weight value corresponding to the current load of mediator B can be adjusted to 0.1, and the case backlog of mediator C is 1, i.e. the current load is 10%, the preset matching weight value corresponding to the current load of mediator C can be adjusted to 0.2. That is, as the current load of the mediator increases, the preset matching weight value corresponding to the current load of the mediator can be appropriately reduced to automatically reduce the matching priority, so as to achieve the effect of avoiding overload allocation and improving case processing efficiency.

[0052] According to the association between the party information and the mediator, the matching degree score of the current mediation case and the mediator is determined to be 0. That is, in this embodiment, the conflict of interest between the party and the mediator can also be detected, and the mediator is automatically excluded if the mediator is associated with the party.

[0053] In an embodiment, the above method can further include determining the matching degree scores of the current mediation case and a plurality of mediators, and sorting the matching degree scores from high to low; and generating a case allocation strategy based on the sorting result of the matching degree scores. The case allocation strategy is the matching mediator corresponding to the current mediation case to be allocated.

[0054] Further, when performing batch allocation for multiple cases to be allocated, the generation of the above case allocation strategy can be further optimized, for example, in some embodiments, an improved Hungarian algorithm can be used to match multiple cases and mediators to achieve batch allocation of cases. The specific steps are as follows: (1) Construct a matching matrix of cases and mediators; consider multiple cases and multiple mediators as two node sets in a bipartite graph, construct an N x M cost matrix C (where N is the number of cases and M is the number of mediators), the elements C(i,j) in the matrix represent the "mismatch degree" or cost when case i is allocated to mediator j, the smaller the numerical value, the higher the matching degree.

[0055] (2) Define multi-dimensional matching weight factors; the calculation of C(i,j) is based on the following multi-dimensional characteristic factors: (a) Case type matching degree: the similarity between the type of case i and the type of case that mediator j is good at; (b) Case complexity: the matching degree between the complexity of case i (such as the amount of money, the number of legal provisions, etc.) and the processing capacity of mediator j; (c) Current load of mediator: the number of cases or workload currently being handled by mediator j; (d) Historical matching performance: the success rate or satisfaction score of mediator j in handling similar cases in history; (e) Regional matching degree: if the case involves regional or language factors, consider whether mediator j has the corresponding conditions.

[0056] By weighting the above factors, C(i,j) is calculated as the matching cost between case i and mediator j.

[0057] (3) Improve the Hungarian algorithm process; based on the standard Hungarian algorithm, the following improvements are introduced in this embodiment: Dynamic weight adjustment mechanism: dynamically adjust the weight coefficient according to real-time data (such as the current load of the mediator); Non-square matrix processing: when the number of cases N is not equal to the number of mediators M, the matrix is expanded to a square matrix by introducing virtual nodes (such as virtual cases or virtual mediators); Parallel processing: block processing of large-scale matching matrix to improve algorithm execution efficiency; Multi-objective optimization: on the basis of minimizing the total cost, add fairness constraints (such as avoiding excessive burden on individual mediators).

[0058] (4) Output the optimal matching scheme; by executing the improved Hungarian algorithm, a set of optimal matching results is output, i.e. each case is allocated to a mediator, ensuring that the overall matching cost is minimized and meeting the fairness and efficiency constraints set by the system.

[0059] The operation process of the embodiment is described below with a specific example: assuming that there are 3 cases (A, B, C) to be allocated to 3 mediators (X, Y, Z). Each case and mediator has corresponding characteristic information, such as case type, case type to which the mediator is good at (i.e., case type in which the mediator is more professional), case complexity, mediator load, and the like. By calculating the matching cost between each case and mediator, the following cost matrix is constructed:

[0060] After applying the improved Hungarian algorithm, the optimal matching scheme is obtained as A→Z, B→Y, and C→X, and the total cost is 3+2+1=6, which is the minimum matching cost.

[0061] In an embodiment, the above method can further include a feedback learning and optimization link, specifically including: first, generating mediation feedback data according to the mediation result of the current case, which can include whether the mediation is successful, mediation duration, and mediation satisfaction, and then updating the state value function of the preset reinforcement learning model based on the mediation feedback data; that is, inputting the mediation feedback data into the preset reinforcement learning model, and preferably, the reinforcement learning model can adopt a Q-Learning algorithm. Then, dynamically adjusting the preset matching weight according to the updated state value function; the preset matching weight can be used to determine the matching degree score of the current case to be mediated and the mediator.

[0062] In some other embodiments, the above preset matching weight can also be used to: based on the dynamically adjusted preset matching weight, recalculate the matching degree score of the current case to be mediated and the mediator; and update the evaluation weight of the characteristic index of each dimension of the mediator according to the recalculated matching degree score, so as to optimize the ability evaluation of the mediator.

[0063] In summary, the mediation case allocation method based on multi-dimensional evaluation provided in the embodiments of the present application can realize scientific case allocation decision by deeply analyzing the matching relationship between case characteristics and mediator ability, establish a dynamic feedback mechanism to continuously optimize the allocation strategy, and ultimately achieve the comprehensive goal of improving mediation efficiency, guaranteeing mediation quality, and optimizing resource allocation.

[0064] Based on the same inventive concept, the embodiments of the present application also provide a mediation case allocation device based on multi-dimensional evaluation, as shown in Figure 2 The device can include: A case characteristic determination module 210 is configured to determine the case characteristics of the current case to be mediated according to the attribute information of the to-be-allocated case obtained in advance. A mediator evaluation module 220 is configured to determine the characteristic index and corresponding score of each dimension of the mediator based on the historical case records and the preset attention characteristics in the case characteristics. The matching module 230 is configured to determine a matching degree score of the current mediation case and the mediator based on the case characteristics and the characteristic indicators and corresponding scores of each dimension of the mediator; and the matching degree score is used to generate a case allocation strategy.

[0065] The mediation case allocation method based on multi-dimensional evaluation in the embodiments of the present application can be applied to an electronic device. The electronic device includes a display screen and a processor, and the display screen is configured to present a graphical user interface.

[0066] In some embodiments, when the graphical user interface is operated by the electronic device, the graphical user interface can be used to display the content local to the electronic device, or can be used to display the content of a server at the other end.

[0067] For example, the application scenario provided by the embodiments of the present application can include an electronic device and a server, and the electronic device communicates with the server through a wired network or a wireless network. The electronic device is configured to run a virtual desktop, and through the virtual desktop, the electronic device can interact with the server.

[0068] In some embodiments, the virtual desktop running on the electronic device can be used to present a mediation case allocation visual scheduling management system based on multi-dimensional evaluation.

[0069] Based on the same inventive concept, the embodiments of the present application also provide a mediation case allocation visual scheduling management system based on multi-dimensional evaluation. The system can be presented on a virtual desktop running on an electronic device, and can be used to monitor and adjust the case allocation strategy. Referring to Figure 3 As shown, the system can include: The case display unit 310 is configured to display information related to the case to be allocated, emergency cases, and the like in real time; The mediator information display unit 320 is configured to display the characteristic indicators of each dimension of the mediator, such as the current load, the field of expertise, the historical success rate, and the like; The allocation result analysis display unit 330 is configured to display the matching degree heat map, the mediation success rate trend, and other allocation results and further analysis related information; The case allocation strategy adjustment unit 340 is configured to respond to the user's operation of case allocation to support the administrator to manually adjust the case allocation result.

[0070] The mediation case allocation method based on multi-dimensional evaluation provided by the present application can adopt a modular system architecture in specific implementation, combine artificial intelligence, data processing and optimization algorithm, and realize intelligent matching between cases and mediators. The system as a whole adopts a distributed architecture, and the front end and the back end are decoupled, which facilitates high-concurrency access and continuous optimization of the model.

[0071] The backend service mainly includes an algorithm processing layer and a business logic layer. The algorithm processing layer can be implemented in Python language, built based on Django or Flask framework, responsible for core algorithm logic such as NLP analysis of case information, calculation of mediator ability evaluation model, and scheduling of matching engine. For high-concurrency request processing, the business logic layer is built in Java or Go language to improve system response speed and stability.

[0072] In terms of artificial intelligence models, case classification and feature extraction can use traditional machine learning classification models built with Scikit-learn, combined with deep learning models based on TensorFlow or PyTorch for natural language processing to improve semantic understanding of case descriptions.

[0073] System data storage can use a hierarchical structure, with structured data such as case information and mediator profiles stored in a MySQL database, and high-frequency access cache data such as current mediator load status and hot case information cached through Redis for cache optimization to improve system response speed.

[0074] The front end can use Vue.js or React framework to build a visual scheduling management platform for administrators to monitor case allocation status, mediator workload, matching effect analysis, and other information in real time, and support manual intervention and strategy adjustment.

[0075] The mediation case allocation device based on multi-dimensional evaluation provided by the embodiments of the present application can be specific hardware on a device or software or firmware installed on a device, etc. The device provided by the embodiments of the present application has the same implementation principle and technical effects as the foregoing method embodiments. For brevity and simplicity of description, the part not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can be referred to the corresponding process in the method embodiments, which will not be repeated here. The mediation case allocation device based on multi-dimensional evaluation provided by the embodiments of the present application has the same technical features as the mediation case allocation method based on multi-dimensional evaluation provided by the foregoing embodiments, so it can also solve the same technical problems and achieve the same technical effects.

[0076] The embodiments of the present application also provide an electronic device, specifically, the electronic device includes a processor and a storage device; the storage device has a computer program stored thereon, and the computer program performs the method of any one of the above embodiments when executed by the processor.

[0077] Figure 4A structural schematic diagram of an electronic device is provided in the embodiments of the present application. The electronic device 400 includes a processor 40, a memory 41, a bus 42 and a communication interface 43, and the processor 40, the communication interface 43 and the memory 41 are connected through the bus 42. The processor 40 is configured to execute an executable module stored in the memory 41, for example, a computer program.

[0078] The memory 41 can include a high-speed random access memory (RAM) and can also include a non-volatile memory, for example, at least one disk memory. The communication connection between the system network element and at least one other network element is implemented through the at least one communication interface 43 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network, etc. can be used.

[0079] The bus 42 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 4 Only one bidirectional arrow is used to represent the bus in the figure, but it does not mean that there is only one bus or only one type of bus.

[0080] The memory 41 is configured to store a program, and the processor 40 executes the program after receiving an execution instruction. The method executed by the device defined by the flow disclosed in any of the embodiments of the present application can be applied to the processor 40 or implemented by the processor 40.

[0081] The processor 40 can be an integrated circuit chip with signal processing capability. In implementation, each step of the above method can be completed by integrated logic circuit of hardware in the processor 40 or by instructions in the form of software. The processor 40 described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium in the art. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41, and combines the hardware to complete the steps of the above method.

[0082] Corresponding to the above method, the embodiment of the present application also provides a computer readable storage medium, the computer readable storage medium stores computer executable instructions, when the processor calls and runs the computer executable instructions, the computer executable instructions make the processor run the steps of the above method.

[0083] In the embodiments provided by the present application, it should be understood that the disclosed device and method can be implemented by other ways. The device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, and for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection between each other can be indirect coupling or communication connection through some communication interface, device or unit, and can be electrical, mechanical or other forms.

[0084] In addition, the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present application.

[0085] Furthermore, the functional modules in various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0086] It should be noted that if the function is realized in the form of a software function module and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and various program codes that can be stored in the medium.

[0087] In this paper, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations.

[0088] The above is only an embodiment of the present application and is not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.​

Claims

1. A method for allocating mediation cases based on multi-dimensional assessment, characterized in that, include: The characteristics of the current cases to be mediated are determined based on the attribute information of the cases to be assigned obtained in advance; Based on historical case records and the preset attention features in the case characteristics, the characteristic indicators and corresponding scores of mediators in each dimension are determined. Based on the case characteristics and the mediator's characteristic indicators and corresponding scores in each dimension, a matching score between the current case to be mediated and the mediator is determined; the matching score is used to generate a case allocation strategy.

2. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 1, characterized in that, The case characteristics include case type, complexity, urgency, geographical location, and other attribute characteristics; The methods for obtaining the attribute information of the cases to be assigned include: Keyword extraction is performed based on the case description text to obtain the attribute information of the current case to be mediated; the attribute information includes: geographical information, dispute type information, amount information, party information, and historical mediation information; The step of determining the case characteristics of the current case to be mediated based on the attribute information of the cases to be assigned in advance includes: The dispute type information is input into a pre-trained mediation case classification model to determine the case type of the current case; The complexity of the current case is determined based on the amount information, the party information, and the historical mediation information. The urgency of the current case is determined based on the case type.

3. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 2, characterized in that, The method, based on historical case records and preset attention features in the case characteristics, determines the mediator's characteristic indicators and corresponding scores for each dimension, including: Among the case characteristics, a feature of concern is pre-determined; the feature of concern is at least one of complexity, urgency, and geographic location. A set of mediator competency assessment parameters is generated based on historical case records. The set of mediator competency assessment parameters includes characteristic indicators of mediators in various dimensions and corresponding initial scores. The characteristic indicators of each dimension include: mediation efficiency, mediation success rate, professionalism in different case types, satisfaction score, and current workload. The evaluation weights of the feature indicators in each dimension are dynamically adjusted based on the aforementioned focus features. Based on the evaluation weights and the initial scores, the corresponding scores of the mediator's characteristic indicators for each dimension and the mediator's overall score are determined.

4. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 3, characterized in that, The matching score between the current case to be mediated and the mediator is determined based on the case characteristics and the mediator's characteristic indicators and corresponding scores across various dimensions, including: The case features are processed into feature vectors to generate case feature vectors; A preset matching algorithm is used to perform multi-dimensional matching calculations on the case feature vector and the corresponding scores of the mediator's feature indicators in each dimension; Based on the matching results of each dimension and the preset matching weights, the matching score between the current case to be mediated and the mediator is calculated by weighted average.

5. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 3, characterized in that, The method of determining the matching score between the current case to be mediated and the mediator based on the case characteristics and the mediator's characteristic indicators and corresponding scores in each dimension also includes: If the mediator's current workload reaches a preset maximum case workload threshold, the matching score between the current case to be mediated and the mediator is determined to be 0. Based on the information of the parties involved and the relationship between the mediator, the matching score between the current case to be mediated and the mediator is determined to be 0.

6. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 1, characterized in that, The method further includes: Determine the matching score between the current case to be mediated and the multiple mediators, and sort the matching scores from high to low; A case allocation strategy is generated based on the ranking results of the matching scores.

7. The method for allocating mediation cases based on multi-dimensional evaluation according to claim 1, characterized in that, The method further includes: Based on the mediation outcome of the current case, mediation feedback data is generated; the mediation feedback data includes: whether the mediation was successful, the mediation duration, and the mediation satisfaction level. The state value function of the preset reinforcement learning model is updated based on the mediation feedback data; The preset matching weight is dynamically adjusted based on the updated state value function; the preset matching weight is used to determine the matching score between the current case to be mediated and the mediator.

8. A mediation case allocation device based on multi-dimensional evaluation, characterized in that, include: The case feature determination module is used to determine the case features of the current case to be mediated based on the attribute information of the cases to be assigned in advance. The mediator evaluation module is used to determine the mediator's characteristic indicators and corresponding scores in various dimensions based on historical case records and preset attention characteristics in the case characteristics. The matching module is used to determine the matching score between the current case to be mediated and the mediator based on the case characteristics and the mediator's feature indicators and corresponding scores in each dimension; the matching score is used to generate a case allocation strategy.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images