Auxiliary case division method and system in combination with case quality and mediator ability
By using a neural network model to evaluate case quality and mediator capabilities, precise matching of cases and mediators is achieved, solving the problem of inaccurate allocation in existing technologies and improving mediation efficiency and success rate.
Patent Information
- Application Number
- CN202511098976.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-06
- Publication Date
- 2025-11-21
AI Technical Summary
The existing case allocation system relies on human experience and lacks in-depth analysis and intelligent matching, resulting in resource misallocation and low allocation accuracy, and is unable to adapt to dynamic changes.
By employing a neural network model that combines case feature data and mediator feature data, and through case quality scoring, case handling ability scoring, and matching degree prediction, accurate matching of cases and mediators can be achieved.
It has improved the accuracy and efficiency of case allocation, shortened the waiting time for allocation, increased the success rate of mediation and the overall efficiency of mediation, and adapted to the dynamic needs of case handling.
Smart Images
Figure CN120996450A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of computer, in particular to a case allocation method and system combining case quality and mediator capacity. BACKGROUND
[0002] The case allocation for judicial mediation and dispute handling mainly relies on artificial experience, and is allocated according to the surface characteristics of the case such as case type, simple statistical information such as the total number of cases handled by the mediator, or simple rules such as random allocation and allocation according to the idle degree of the mediator.
[0003] In the prior art, the case allocation system can only realize basic case registration and mediator information recording, lacks depth analysis and intelligent matching function, and the data interaction between modules is limited, which cannot comprehensively integrate case and mediator related information for comprehensive judgment, so as to accurately match the capacity of the mediator, cause resource mismatch, and cannot adapt to the dynamic changes of the case and the mediator. SUMMARY
[0004] The purpose of the embodiments of the present application is to provide a case allocation method and system combining case quality and mediator capacity, so as to alleviate the technical problems of low allocation accuracy, low efficiency and lack of dynamic adaptability in the prior art.
[0005] In a first aspect, the present application provides a case allocation method combining case quality and mediator capacity, comprising:
[0006] obtaining case feature data of a target case and case handling feature data of a candidate mediator;
[0007] based on the case feature data of the target case and the case handling feature data of the candidate mediator, using a case allocation model to predict the matching degree between the target case and the candidate mediator; wherein the case allocation model is to process the case feature data of the target case to obtain a case quality score of the target case; to process the case handling feature data of the candidate mediator to obtain a case handling capacity score of the candidate mediator; based on the case quality score and the case handling capacity score, to determine a neural network model of the matching degree between the target case and the candidate mediator, and the neural network model outputs at least one of the case quality score, the case handling capacity score and the matching degree between the target case and the candidate mediator.
[0008] Optionally, based on the case feature data of the target case and the case handling feature data of the candidate mediator, using the case allocation model to predict the matching degree between the target case and the candidate mediator, comprising:
[0009] based on the case feature data of the target case, determining the case quality score of the target case;
[0010] determine a case quality score of the target case based on the case type, the mediation time, and the case complexity of the target case;
[0011] predict a matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
[0012] Optionally, the case feature data comprises a case type, a mediation time, and a case complexity; and the case handling feature data comprises a mediator feature, a case feature, and an emotional analysis result.
[0013] Optionally, the case allocation model comprises a case evaluation sub-model, an ability evaluation sub-model, and a matching degree sub-model.
[0014] The case evaluation sub-model is configured to determine the case quality score of the target case based on the case type, the mediation time, and the case complexity of the target case.
[0015] The ability evaluation sub-model is configured to determine the case handling ability score of the candidate mediator based on the mediator feature, the case feature, and the emotional analysis result.
[0016] The matching degree sub-model is configured to predict the matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
[0017] Optionally, after predicting the matching degree between the target case and the candidate mediator based on the case feature data of the target case and the case handling feature data of the candidate mediator by using the case allocation model, the method further comprises:
[0018] determining the candidate mediator for handling the target case based on the matching degree between the target case and the candidate mediator and a preset threshold.
[0019] Optionally, the case allocation model is a machine learning model; and the case allocation model is determined in the following manner:
[0020] obtaining a training data set; wherein the training data set comprises a plurality of training sample data; each training sample data comprises case feature sample data, case handling feature sample data, matching degree data, a case quality score, and a case handling ability score;
[0021] performing machine learning operations on the initial case allocation model iteratively based on the training data set, until an iteration termination condition is met, and determining the case allocation model based on each model parameter of the initial case allocation model updated in the last time when the machine learning operation is performed; wherein the machine learning operation comprises:
[0022] selecting target training sample data from the training data set;
[0023] The case characteristic sample data and the case handling characteristic sample data in the target training sample data are input into the initial case distribution model, so that the initial case distribution model receives the case characteristic sample data and the case handling characteristic sample data through an input layer, processes the case characteristic sample data and the case handling characteristic sample data through a hidden layer to obtain at least one of case evaluation characteristic prediction data, ability evaluation characteristic prediction data and prediction matching degree, and outputs the at least one of the case evaluation characteristic prediction data, the ability evaluation characteristic prediction data and the prediction matching degree through an output layer;
[0024] Based on the prediction errors between the case evaluation characteristic prediction data, the case quality score in the target training sample data, the ability evaluation characteristic prediction data, the case handling ability score in the target training sample data and the prediction matching degree and the matching degree data in the target training sample data, the model parameters of the initial case distribution model are updated.
[0025] Optionally, the method further comprises:
[0026] Based on the case characteristic data of the target case, the case handling characteristic data of the candidate mediator and the matching degree between the target case and the candidate mediator, the case distribution model is optimized.
[0027] In a second aspect, the present application provides an auxiliary case distribution system combining case quality and mediator ability, comprising:
[0028] An acquisition module is configured to acquire case characteristic data of a target case and case handling characteristic data of a candidate mediator;
[0029] A case distribution module is configured to predict a matching degree between the target case and the candidate mediator based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator by using a case distribution model, wherein the case distribution model is a neural network model configured to process the case characteristic data of the target case to obtain a case quality score of the target case, process the case handling characteristic data of the candidate mediator to obtain a case handling ability score of the candidate mediator, and determine the matching degree between the target case and the candidate mediator based on the case quality score and the case handling ability score, and the neural network model outputs at least one of the case quality score, the case handling ability score and the matching degree between the target case and the candidate mediator.
[0030] In a third aspect, the present application further provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the auxiliary case distribution method combining case quality and mediator ability when executing the computer program.
[0031] In a fourth aspect, the present application further provides a computer readable storage medium, which stores machine executable instructions, and when the machine executable instructions are invoked and executed by a processor, the machine executable instructions cause the processor to execute the above-mentioned auxiliary case allocation method combining case quality and mediator capacity.
[0032] The auxiliary case allocation method combining case quality and mediator capacity provided by the present application can accurately evaluate case quality and mediator capacity by predicting case quality of a target case and capacity of a candidate mediator based on case feature data of the target case and case handling feature data of the candidate mediator through a case allocation model with characteristics of a neural network model; can accurately match cases and mediators by determining mediators needed for handling the target case based on case quality and capacity of the candidate mediator, so that mediators can effectively cope with complex situations of cases and demands of parties, thereby improving the probability of mediation success, effectively resolving disputes, reasonably allocating cases according to the capacity and specialty of mediators, giving full play to the advantages of each mediator, improving overall mediation efficiency, and realizing an automated and intelligent case allocation process, which greatly shortens the time for waiting for allocation of cases, reduces manual intervention, improves the overall speed of case handling, and adapts to the growing demand for case handling.
[0033] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS
[0034] 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 to 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.
[0035] Figure 1 A process schematic diagram of the auxiliary case allocation method combining case quality and mediator capacity provided by the embodiments of the present application;
[0036] Figure 2 A training process schematic diagram of the case evaluation sub-model provided by the embodiments of the present application;
[0037] Figure 3 A structure schematic diagram of the auxiliary case allocation system combining case quality and mediator capacity provided by the embodiments of the present application;
[0038] Figure 4A hardware structure schematic diagram of an electronic device in the present application. DETAILED DESCRIPTION
[0039] In order to make the purposes, technical solutions and beneficial effects of the present application clearer, the technical solutions of the present application will be described clearly and completely in combination with embodiments and drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.
[0040] In order to make the purposes, technical solutions and beneficial effects of the present application clearer, the technical solutions of the present application will be described clearly and completely in combination with embodiments and drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.
[0041] A case is a specific matter submitted to a judicial department or a special mediation agency for seeking solutions to various disputes and controversies between people in law. In the present application, the case is mainly a case of judicial mediation and dispute handling.
[0042] Case feature data is case feature data used to reflect the essential attributes of a case. In the present application, the case feature data includes case type, mediation time and case complexity, etc.
[0043] Case feature data is case feature data used to reflect the essential attributes of a case. In the present application, the case feature data includes case type, mediation time and case complexity, etc.
[0044] The case allocation model is a neural network model (Multilayer Perceptron-Convolutional Neural Networks, MLP-CNN) that learns the mapping relationship between case feature data and mediation feature data and the matching degree between the case and the mediator, and realizes prediction of the matching degree between the case and the mediator. In this application, the case allocation model includes a case evaluation sub-model, an ability evaluation sub-model, and a matching degree sub-model. The case evaluation sub-model is used to determine the case quality score of the case based on the case type, mediation time, and case complexity of the case. The ability evaluation sub-model is used to determine the mediation ability score of the mediator based on the mediator features, case features, and emotional analysis results. The matching degree sub-model is used to predict the matching degree between the case and the mediator based on the case quality score of the case and the mediation ability score of the mediator. In this application, the case evaluation sub-model can use random forest, linear regression model, extreme gradient boosting algorithm (XGBoost), and neural network model for prediction. The ability evaluation sub-model can use the Adam optimizer to optimize the neural network model for prediction. The matching degree sub-model can use multilayer perceptron (MLP) or convolutional neural network (CNN) to predict the matching degree between the case and the mediator.
[0045] After introducing the technical terms related to the present application, the technical solutions provided by the present application will be described in detail.
[0046] The embodiment of the present application provides an auxiliary case allocation method combining case quality and mediator ability. Referring to FIG. 1, Figure 1 The general process of the auxiliary case allocation method combining case quality and mediator ability provided by the embodiment of the present application is as follows:
[0047] In step S110, the case feature data of the target case and the mediation feature data of the candidate mediator are obtained.
[0048] In the application embodiment, when allocating a case, the case type, mediation time, and case complexity of the target case can be obtained as the case feature data of the target case, and the mediator features, case features, and emotional analysis results of the candidate mediator can be obtained as the mediation feature data.
[0049] Further, the mediator features include historical mediation success rate, number of handled cases, whether having legal education background and related field patent knowledge, and historical customer satisfaction evaluation; the case features include the difficulty of historical cases handled by the mediator, legal clause complexity, and the number of parties; and the emotional analysis results include positive and negative emotions.
[0050] The case characteristic data of the target case and the case handling characteristic data of the candidate mediator can be obtained from the target case by using regular expressions, machine learning methods, semantic analysis and the like.
[0051] In step S120, a case allocation model is used to predict the matching degree between the target case and the candidate mediator based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator. The case allocation model is a neural network model that processes the case characteristic data of the target case to obtain a case quality score of the target case, processes the case handling characteristic data of the candidate mediator to obtain a case handling ability score of the candidate mediator, and determines the matching degree between the target case and the candidate mediator based on the case quality score and the case handling ability score. The neural network model outputs at least one of the case quality score, the case handling ability score and the matching degree between the target case and the candidate mediator.
[0052] In the embodiments of the present application, the input data is the case characteristic data of the target case and the case handling characteristic data of the candidate mediator, and the output data is at least one of the case quality score, the case handling ability score and the matching degree between the target case and the candidate mediator.
[0053] Further, the case allocation model includes a case evaluation sub-model, an ability evaluation sub-model and a matching degree sub-model.
[0054] The case evaluation sub-model is used to determine the case quality score of the target case based on the case type, the mediation time and the case complexity of the target case.
[0055] The ability evaluation sub-model is used to determine the case handling ability score of the candidate mediator based on the mediator characteristics, the case characteristics and the emotional analysis result.
[0056] The matching degree sub-model is used to predict the matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
[0057] In the embodiments of the present application, the case evaluation sub-model can use a linear regression model to predict the case quality score of the target case, and the historical case data is used as a training set to train the linear regression model. The linear regression model is expressed as:
[0058] Q = w1x1 + w2x2 + w3x3 + … + wnxn + b n x n +b
[0059] In the formula, Q is the case quality score, x1, x2, x3…xn are the case characteristic data of the case, w1, w2, w3…wn are the weight values of the case characteristic data, and b is the bias value. n n
[0060] Further, the case evaluation sub-model is trained using a gradient descent method or other optimization algorithm to minimize a loss function, such as mean square error (MSE), expressed as:
[0061]
[0062] where MSE is the mean square error value, Q i is the actual case quality score, is the predicted case quality score, and m is the number of samples.
[0063] As shown in Figure 2 , in the embodiments of the present application, the case evaluation sub-model can be trained based on the case feature data of historical cases and the historical case quality scores corresponding to the historical cases, and the training process is as follows:
[0064] Step 210, obtaining a training data set; wherein the training data set includes a plurality of training sample data; each training sample data includes case feature data of a case and an actual case quality score;
[0065] Step 220, selecting target training sample data from the training data set;
[0066] Step 230, inputting the case feature data and the case quality score in the target training sample data into an initial case evaluation sub-model, so that the initial case evaluation sub-model outputs a predicted case quality score;
[0067] Step 240, updating the bias value and each weight value of the initial case evaluation sub-model based on the prediction error between the predicted case quality score and the actual case quality score;
[0068] Step 250, determining whether an iteration training termination condition is met; if yes, performing step 260; if no, returning to step 220; wherein the iteration training termination condition is that the number of iterations is not less than a number threshold, or the prediction error is not higher than an error threshold.
[0069] Step 260: obtaining the case evaluation sub-model based on the bias value and each weight value of the initial case evaluation sub-model updated when the last iteration training operation is performed.
[0070] Through the above training method, the training speed and accuracy of the case evaluation sub-model can be improved.
[0071] In the embodiments of the present application, the ability evaluation sub-model can use a neural network model to predict the case handling ability score of the candidate mediator, and train the ability evaluation sub-model based on a weighted cross-entropy loss and a loss function of mediation cost, wherein the case handling ability score is the score of successful mediation.
[0072] Further, the capability evaluation sub-model includes an input layer, a hidden layer, and an output layer. The input layer receives the case handling characteristic data of the candidate mediator, such as the mediator characteristics, the case characteristics, and the emotional analysis results, and the hidden layer processes the case handling characteristic data to obtain a case handling capability score. The output layer outputs the case handling capability score of the candidate mediator. The structure of the capability evaluation sub-model is as follows:
[0073] Input layer:
[0074] The input vector x = [x1, x2, …, x n ], where x1, x2, …, x n are the case handling characteristic data.
[0075] The hidden layer includes a first hidden layer and a second hidden layer. The first hidden layer can use a ReLU activation function to maintain nonlinearity.
[0076] h1(x) = ReLU(W1x + b1)
[0077] where W1 is the first hidden layer weight matrix, x is the first hidden layer case handling characteristic data, b1 is the first hidden layer bias term, ReLU is the activation function, and h1(x) is the first hidden layer output value.
[0078] The second hidden layer uses LeakyReLU to avoid gradient disappearance.
[0079] h2(h1) = LeakyReLU(W2h1 + b2)
[0080] where W2 is the second hidden layer weight matrix, h1 is the first hidden layer output value, b2 is the second hidden layer bias term, LeakyReLU is the activation function, and h2(h1) is the second hidden layer output value.
[0081] Output layer:
[0082] For the prediction of the case handling capability score, a Sigmoid activation function is used to output the probability p of the case handling capability score.
[0083] p = σ(W3h2 + b3)
[0084] where σ is the Sigmoid function, W3 is the weight matrix of the output layer, h2 is the second hidden layer output value, b3 is the bias term of the output layer, and is the Sigmoid function.
[0085] Further, the capability evaluation sub-model is trained based on a weighted cross-entropy loss and a mediation cost loss function, where the mediation cost loss function L is:
[0086]
[0087] wherein, N is the number of samples, c i is the mediation cost coefficient of the ith case, y i is the actual mediation result, p i is the success probability predicted by the capability evaluation submodel, wherein, c i represents the mediation cost, for the case of mediation failure, c i > 1; for the case of mediation success, c i = 1.
[0088] Further, in order to consider the influence of mediation cost on the training of the capability evaluation submodel, an adjusted loss function is used to reflect different mediation costs, so as to improve the accuracy of the capability evaluation submodel.
[0089] In order to prevent the capability evaluation submodel from overfitting in the training process, in the embodiment of the present application, an L2 regularization term is added to the loss function of the mediation cost and an Adam optimizer is used to effectively prevent the capability evaluation submodel from overfitting in the training process, so as to improve the generalization ability of the model.
[0090] wherein, the expression of the L2 regularization term L total is as follows:
[0091]
[0092] wherein, L is a weighted cross-entropy loss function, λ is a regularization strength parameter, W l is the weight matrix of the lth layer.
[0093] The expression of the Adam optimizer is as follows:
[0094]
[0095] wherein, θ is a model parameter, α is a learning rate, m t and v t are the first moment estimate and the second moment estimate of the gradient respectively, ∈ is a small constant for numerical stability.
[0096] In the embodiment of the present application, the matching degree submodel can adopt a multi-layer perceptron model or a convolutional neural network model, and the matching degree submodel is trained based on a binary cross-entropy loss.
[0097] Further, the multi-layer perceptron model is used to predict the matching degree between the target case and the candidate mediator, and the expression of the multi-layer perceptron model M thereof is as follows:
[0098] M = f (W2·σ (W1·X+b1) +b2)
[0099] In the formula, W1 is the weight matrix of the first hidden layer, X is the input feature vector, b1 is the bias vector of the first hidden layer, sigma is the activation function of the first hidden layer, W2 is the weight matrix of the second hidden layer, b2 is the bias vector of the second hidden layer, and f is the activation function of the second hidden layer.
[0100] The input feature vector X includes the case quality score and the case handling ability score; the activation function sigma of the first hidden layer is ReLU; and the activation function f of the second hidden layer is sigmoid.
[0101] Further, in order to improve the accuracy of the matching degree submodel, the binary cross entropy loss is used to train the matching degree submodel, where the binary cross entropy loss function BCE is:
[0102]
[0103] In the formula, y i is the actual matching degree, is the predicted matching degree, and m is the sample quantity.
[0104] In an optional implementation, after the matching degree between the target case and the candidate mediator is predicted by using the case allocation model based on the case feature data of the target case and the case handling feature data of the candidate mediator, the method further includes:
[0105] Based on the matching degree between the target case and the candidate mediator and a preset threshold, the candidate mediator handling the target case is determined.
[0106] For each candidate mediator, the matching degree between the candidate mediator and the target case is calculated by using the case allocation model, and then the matching degree is compared with the preset threshold. When the matching degree is greater than the preset threshold, it is considered that the candidate mediator can handle the target case; and when the matching degree is less than the preset threshold, it is considered that the candidate mediator cannot handle the target case.
[0107] The auxiliary case allocation method provided by the embodiments of the present application combines case quality and mediator ability, accurately evaluates case quality and mediator ability by using machine learning and a neural network model, and realizes efficient case allocation, thereby improving the accuracy of case allocation, making the case highly adaptive to the ability of mediators, improving the success rate and quality of mediation, improving the efficiency of case allocation, shortening the waiting time for case allocation, and quickly responding to the demand for case handling; and the dynamic adjustment of the allocation strategy is realized, and the allocation scheme is optimized according to the real-time changes of cases and mediators.
[0108] Based on the above embodiments, the embodiments of the present application provide an auxiliary case allocation system combining case quality and mediator ability, which is described with reference to Figure 3As shown, the auxiliary case division system provided by the embodiments of the present application combines case quality and mediator ability, and at least includes:
[0109] The acquisition module 310 is configured to acquire case characteristic data of a target case and case handling characteristic data of a candidate mediator;
[0110] The case division module 320 is configured to predict a matching degree between the target case and the candidate mediator based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator by using a case division model; wherein the case division model is a neural network model that receives the case characteristic data of the target case and the case handling characteristic data of the candidate mediator through an input layer, processes the case characteristic data of the target case and the case handling characteristic data of the candidate mediator through a hidden layer to obtain at least one of a case quality score of the target case, a case handling ability score of the candidate mediator and a matching degree between the target case and the candidate mediator, and outputs at least one of the case quality score of the target case, the case handling ability score of the candidate mediator and the matching degree between the target case and the candidate mediator through an output layer.
[0111] In an optional embodiment, the case division module 320 is further configured to:
[0112] determine the case quality score of the target case based on the case characteristic data of the target case;
[0113] determine the case handling ability score of the candidate mediator based on the case handling characteristic data of the candidate mediator;
[0114] predict the matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
[0115] In an optional embodiment, the case characteristic data includes a case type, a mediation time and a case complexity; and the case handling characteristic data includes a mediator characteristic, a case characteristic and an emotional analysis result.
[0116] In an optional embodiment, the case division model includes a case evaluation sub-model, an ability evaluation sub-model and a matching degree sub-model.
[0117] The case evaluation sub-model is configured to determine the case quality score of the target case based on the case type, the mediation time and the case complexity of the target case;
[0118] The ability evaluation sub-model is configured to determine the case handling ability score of the candidate mediator based on the mediator characteristic, the case characteristic and the emotional analysis result.
[0119] The matching degree sub-model is configured to predict the matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
[0120] In an optional embodiment, the case distribution module 320 is further configured to:
[0121] The candidate mediator handling the target case is determined based on the matching degree between the target case and the candidate mediator and a preset threshold.
[0122] In an optional embodiment, the case distribution module 320 is further configured to:
[0123] The training data set is obtained, wherein the training data set includes a plurality of training sample data, each training sample data includes case feature sample data, case handling feature sample data, matching degree data, case quality score and case handling ability score.
[0124] The machine learning operation is iteratively performed on the initial case distribution model based on the training data set, and when it is determined that the iteration termination condition is met, the case distribution model is determined based on the model parameters of the initial case distribution model updated in the last time of performing the machine learning operation, wherein the machine learning operation includes:
[0125] The target training sample data is selected from the training data set.
[0126] The case feature sample data and the case handling feature sample data in the target training sample data are input into the initial case distribution model, so that the initial case distribution model receives the case feature sample data and the case handling feature sample data through the input layer, processes the case feature sample data and the case handling feature sample data through the hidden layer to obtain at least one of the case evaluation feature prediction data, the ability evaluation feature prediction data and the predicted matching degree, and outputs at least one of the case evaluation feature prediction data, the ability evaluation feature prediction data and the predicted matching degree through the output layer.
[0127] The model parameters of the initial case distribution model are updated based on the prediction errors between each case evaluation feature prediction data and the case quality score in the target training sample data, between the ability evaluation feature prediction data and the case handling ability score in the target training sample data, and between the predicted matching degree and the matching degree data in the target training sample data.
[0128] In an optional embodiment, the case distribution module 320 is further configured to:
[0129] The case distribution model is optimized based on the case feature data of the target case, the case handling feature data of the candidate mediator and the matching degree between the target case and the candidate mediator.
[0130] It should be noted that the principle of solving the technical problem of the auxiliary case division system provided by the embodiments of the present application in combination with the case quality and the ability of mediators is similar to the auxiliary case division method provided by the embodiments of the present application in combination with the case quality and the ability of mediators. Therefore, the implementation of the auxiliary case division system provided by the embodiments of the present application in combination with the case quality and the ability of mediators can refer to the implementation of the auxiliary case division method provided by the embodiments of the present application in combination with the case quality and the ability of mediators, and the repeated parts will not be described here.
[0131] After introducing the auxiliary case division method and system provided by the embodiments of the present application in combination with the case quality and the ability of mediators, next, the electronic device provided by the embodiments of the present application is simply introduced.
[0132] Referring to Figure 4 The electronic device 500 provided by the embodiments of the present application at least includes a processor 501, a memory 502, and a computer program stored in the memory 502 and executable on the processor 501, and the processor 501 implements the auxiliary case division method provided by the embodiments of the present application when executing the computer program.
[0133] The electronic device 500 provided by the embodiments of the present application can also include a bus 503 connected to different components (including the processor 501 and the memory 502). Among them, the bus 503 represents one or more of several types of bus structures, including a memory bus, a peripheral bus, a local bus, etc.
[0134] The memory 502 can include a readable storage medium in the form of a volatile memory, such as a random access memory (RAM) 5021 and / or a cache memory 5022, and can further include a read-only memory (ROM) 5023. The memory 502 can also include a program tool 5025 having a set of (at least one) program modules 5024, including but not limited to an operating subsystem, one or more application programs, other program modules, and program data, each of these examples or some combination thereof can include the implementation of a network environment.
[0135] The processor 501 can be one processing element or a collective term for multiple processing elements. For example, the processor 501 can be a central processing unit (CPU), or one or more integrated circuits configured to implement the method of assisting case quality and mediator ability for case division provided by the embodiments of the present application. Specifically, the processor 501 can be a general-purpose processor, including but not limited to a CPU, 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, etc.
[0136] The electronic device 500 can communicate with one or more external devices 504 (such as a keyboard, a remote control, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 500 (such as a phone, a computer, etc.), and / or with one or more devices that enable the electronic device 500 to communicate with one or more other electronic devices 500 (such as a router, a modem, etc.). Such communication can occur via the input / output (I / O) interface 505. Also, the electronic device 500 can communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet) via the network adapter 506. As Figure 4 illustrated, the network adapter 506 communicates with the other components of the electronic device 500 via the bus 503. It should be understood that although not shown, other hardware and / or software components could be used in conjunction with the electronic device 500. These components, as is well known in the art, include, inter alia, microcode, device drivers, redundant processors, external disk drive arrays, RAID (Redundant Arrays of Independent Disks) subsystems, tape drives, and data archival storage subsystems, etc. Figure 4
[0137] It should be noted that the electronic device 500 as shown is only one example of an electronic device and should not be construed to limit the functionality and use of embodiments of the present application. Figure 4 It should be noted that the electronic device 500 as shown is only one example of an electronic device and should not be construed to limit the functionality and use of embodiments of the present application.
[0138] The computer readable storage medium provided by the embodiments of the present application is described below. The computer readable storage medium provided by the embodiments of the present application stores computer instructions, and the computer instructions are executed by a processor to implement the case quality and mediator capacity combined auxiliary case division method provided by the embodiments of the present application. Specifically, the computer instructions can be built-in or installed in the processor, so that the processor can implement the case quality and mediator capacity combined auxiliary case division method provided by the embodiments of the present application by executing the built-in or installed computer instructions.
[0139] In addition, the case quality and mediator capacity combined auxiliary case division method provided by the embodiments of the present application can also be implemented as a computer program product, which includes program codes that implement the case quality and mediator capacity combined auxiliary case division method provided by the embodiments of the present application when running on a processor.
[0140] The computer program product provided by the embodiments of the present application can adopt one or more computer readable storage media, and the computer readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any suitable combination of the above. Specifically, more specific examples (non-exhaustive list) of the computer readable storage medium include an electrical connection with one or more wires, a portable disk, a hard disk, RAM, ROM, Erasable Programmable Read Only Memory (EPROM), an optical fiber, a portable Compact Disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0141] The computer program product provided by the embodiments of the present application can adopt a CD-ROM and include program codes, and can also run on an electronic device such as a computer. However, the computer program product provided by the embodiments of the present application is not limited to this. In the embodiments of the present application, the computer readable storage medium can be any tangible medium containing or storing program codes, which can be used or combined with an instruction execution system, device or component.
[0142] It should be noted that although several units or sub-units of the apparatus are mentioned in the above detailed description, such division is only exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided into multiple units.
[0143] Furthermore, although the operations of the method(s) herein can be described in a particular, sequential order, this order is not meant to be a limitation and one or more of the operations described can be performed in parallel, or in a different order, including simultaneous execution of one or more operations, unless otherwise specified herein. In addition, the description sometimes uses terms like "operate" or "execute" to describe the operation of certain steps, which should not be construed to refer to a specific computer or memory location. Rather, such terms are used to generally describe the operation of the steps.
[0144] While the preferred embodiments of the application have been described above, it will be recognized and understood that various modifications and changes can be made to the embodiments of the application without departing from the spirit and scope of the application. It is intended that the appended claims be construed to include all such modifications and changes insofar as they fall within the scope of the present application.
[0145] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.
Claims
1. An assisted case assignment method that combines case quality and mediator capacity, characterized in that, The method comprises the following steps: obtaining case characteristic data of a target case and case handling characteristic data of a candidate mediator; based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator, using a case allocation model to predict the matching degree between the target case and the candidate mediator; wherein the case allocation model is that the case characteristic data of the target case is processed to obtain a case quality score of the target case; the case handling characteristic data of the candidate mediator is processed to obtain a case handling ability score of the candidate mediator; based on the case quality score and the case handling ability score, a neural network model is used to determine the matching degree between the target case and the candidate mediator, and the neural network model outputs at least one of the case quality score, the case handling ability score and the matching degree between the target case and the candidate mediator.
2. The method of claim 1, wherein, Based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator, using a case allocation model to predict the matching degree between the target case and the candidate mediator, comprising: based on the case characteristic data of the target case, determining the case quality score of the target case; based on the case handling characteristic data of the candidate mediator, determining the case handling ability score of the candidate mediator; based on the case quality score of the target case and the case handling ability score of the candidate mediator, predicting the matching degree between the target case and the candidate mediator.
3. The method of claim 1, wherein, The case characteristic data comprises case type, mediation time and case complexity; the case handling characteristic data comprises mediator characteristics, case characteristics and emotional analysis results.
4. The method of claim 3, wherein, The case allocation model comprises a case evaluation sub-model, an ability evaluation sub-model and a matching degree sub-model; The case evaluation sub-model is used to determine the case quality score of the target case based on the case type, the mediation time and the case complexity of the target case; The ability evaluation sub-model is used to determine the case handling ability score of the candidate mediator based on the mediator characteristics, the case characteristics and the emotional analysis results; The matching degree sub-model is used to predict the matching degree between the target case and the candidate mediator based on the case quality score of the target case and the case handling ability score of the candidate mediator.
5. The method of claim 1, wherein, Based on the case characteristic data of the target case and the case handling characteristic data of the candidate mediator, using a case allocation model to predict the matching degree between the target case and the candidate mediator, further comprising: based on the matching degree between the target case and the candidate mediator and a preset threshold, determining the candidate mediator to handle the target case.
6. The assisted singulation method of any one of claims 1-5, wherein, The case allocation model is a machine learning model; the case allocation model is determined in the following way: obtaining a training data set; wherein the training data set comprises a plurality of training sample data; each training sample data comprises case characteristic sample data, case handling characteristic sample data, matching degree data, case quality score and case handling ability score; The machine learning operation is iteratively performed on the initial allocation model based on the training data set until it is determined that an iteration termination condition is met, and the allocation model is determined based on each model parameter of the initial allocation model updated when the machine learning operation is last performed; wherein the machine learning operation comprises: selecting target training sample data from the training data set; inputting case feature sample data and case handling feature sample data in the target training sample data into the initial allocation model, so that the initial allocation model receives the case feature sample data and the case handling feature sample data through an input layer, and processes the case feature sample data and the case handling feature sample data through a hidden layer to obtain at least one of case evaluation feature prediction data, ability evaluation feature prediction data and prediction matching degree, and then outputs at least one of the case evaluation feature prediction data, the ability evaluation feature prediction data and the prediction matching degree through an output layer; updating each model parameter of the initial allocation model based on the prediction error between each case evaluation feature prediction data and the case quality score in the target training sample data, the ability evaluation feature prediction data and the case handling ability score in the target training sample data, and the prediction matching degree and the matching degree data in the target training sample data.
7. The method of claim 1, wherein, Further comprising: optimizing the allocation model based on the case feature data of the target case, the case handling feature data of the candidate mediator and the matching degree between the target case and the candidate mediator.
8. An auxiliary case assignment system that combines case quality and mediator capacity, characterized in that, Comprise: an acquisition module, configured to acquire case feature data of a target case and case handling feature data of a candidate mediator; an allocation module, configured to predict a matching degree between the target case and the candidate mediator based on the case feature data of the target case and the case handling feature data of the candidate mediator by using an allocation model; wherein the allocation model is a neural network model that receives the case feature data of the target case and the case handling feature data of the candidate mediator through an input layer, processes the case feature data of the target case and the case handling feature data of the candidate mediator through a hidden layer to obtain at least one of a case quality score of the target case, a case handling ability score of the candidate mediator and a matching degree between the target case and the candidate mediator, and then outputs at least one of the case quality score of the target case, the case handling ability score of the candidate mediator and the matching degree between the target case and the candidate mediator through an output layer.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method of any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores machine executable instructions, and when the machine executable instructions are called and executed by the processor, the machine executable instructions cause the processor to execute the method of any one of claims 1-7.