Task allocation processing method and apparatus

Abstract

Embodiments of the present application provide a task allocation processing method and device. The method comprises: obtaining a target task to be allocated, and determining a target customer; obtaining customer characteristic information of the target customer and personnel characteristic information of each worker, performing normalization processing on each numerical characteristic information of the target customer and each worker, and inputting each category characteristic information of the target customer and each worker into an embedding layer of a task completion rate prediction model to obtain a category characteristic vector; inputting the numerical characteristic vector and the category characteristic vector of the target customer and each worker into the task completion rate prediction model to output a predicted completion rate of the target task corresponding to each worker; and sending the target task to a terminal corresponding to a target worker with the highest predicted completion rate. Based on the method provided by the present application, it can be ensured that the target task can be completed with the highest probability, and the purpose of improving the task completion rate is achieved.

Description

Technical Field

[0001] This application relates to the field of data processing, and more particularly to a task allocation processing method and apparatus. Background Technology

[0002] The insurance sales management system can design specific sales processes based on sales expectations for insurance products, and assign corresponding sales tasks to staff by specifying sales targets at different points in the sales process, so that staff can complete their sales targets and ultimately complete the sales of insurance products.

[0003] Currently, insurance sales management systems allocate sales tasks in several ways, including by the time the task is generated, by average workload, and by the previous responsible person.

[0004] However, the current insurance sales management system's method of allocating sales tasks can only meet the normal needs of staff to execute sales tasks, but the completion rate of insurance product sales tasks is generally low. Summary of the Invention

[0005] This application provides a task allocation processing method and apparatus to improve the completion rate of insurance product sales tasks.

[0006] In a first aspect, embodiments of this application provide a task allocation processing method, applied to a server, the method comprising:

[0007] Obtain the target tasks to be assigned and determine the target customers corresponding to the target tasks;

[0008] Obtain customer characteristic information of the target customer, wherein the customer characteristic information of the target customer includes each numerical characteristic information and each category characteristic information of the target customer;

[0009] Obtain the personnel characteristic information of each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member;

[0010] Each numerical feature of the target customer is normalized to obtain the numerical feature vector of the target customer.

[0011] The target customer's category feature information is input into the task completion rate prediction model, which includes an embedding layer, a connection layer, a first module, a second module, and an output layer. The first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer. The task completion rate prediction model then performs the following steps:

[0012] Each category feature is input into the embedding layer to output the category feature vector of the target customer;

[0013] The numerical feature vector and category feature vector of the target customer are input into the connection layer to output the complete feature vector corresponding to the target customer;

[0014] For each staff member, each numerical feature information of the staff member is normalized to obtain the numerical feature vector of the staff member;

[0015] For each worker, the category feature information of each worker is input into the embedding layer to output the category feature vector of the worker;

[0016] For each worker, the numerical feature vector and category feature vector of the worker are input into the connection layer to output the complete feature vector corresponding to the worker;

[0017] For each staff member, the complete feature vector corresponding to the staff member and the complete feature vector corresponding to the target customer are input into the first module for fusion and activation processing, so as to output the first fused feature corresponding to the staff member;

[0018] Select reference staff from among the staff, and determine the complete feature vectors corresponding to the reference staff as reference complete feature vectors;

[0019] For each staff member, the complete feature vector corresponding to the staff member and the reference complete feature vector are input into the second module for linear transformation and dimensionality reduction processing to output the second fused feature corresponding to the staff member;

[0020] For each worker, the first fusion feature and the second fusion feature corresponding to the worker are input into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker;

[0021] Compare the predicted completion rates of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member;

[0022] The target task is sent to the terminal corresponding to the target staff member.

[0023] In one possible implementation, the normalization process for each numerical feature information of the target customer to obtain the numerical feature vector of the target customer includes:

[0024] Obtain the preset upper limit and preset lower limit values ​​corresponding to each numerical feature information of the target customer;

[0025] The absolute value of each numerical feature is compared with the difference between the corresponding preset upper limit and preset lower limit to obtain the ratio result for each numerical feature.

[0026] The ratio results corresponding to each numerical feature information of the target customer are concatenated to obtain the numerical feature vector of the target customer;

[0027] The step of normalizing each numerical feature information of the staff member to obtain the numerical feature vector of the staff member includes:

[0028] Obtain the preset upper limit and preset lower limit values ​​corresponding to each numerical feature information of the staff member;

[0029] The absolute value of each numerical feature is compared with the difference between the corresponding preset upper limit and preset lower limit to obtain the ratio result for each numerical feature.

[0030] The ratio results corresponding to each numerical feature information of the staff member are concatenated to obtain the numerical feature vector of the staff member.

[0031] In one possible implementation, the selection of reference staff from among the staff includes:

[0032] Obtain task completion records, wherein the task completion records include each completed historical task and the staff member who completed each historical task;

[0033] The staff member who completed the most historical tasks was identified as the reference staff member.

[0034] In one possible implementation, the embedding layer includes an autoencoder fully connected neural network model;

[0035] Accordingly, the step of inputting each category feature information into the embedding layer to output the category feature vector of the target customer includes:

[0036] Each category feature information is input into the autoencoder fully connected neural network model to output the category feature vector of the target customer;

[0037] Accordingly, the step of inputting each category feature information of the staff member into the embedding layer to output the category feature vector of the staff member includes:

[0038] Each category feature information of the staff member is input into the autoencoder fully connected neural network model to output the category feature vector of the staff member.

[0039] In one possible implementation, the training process of the autoencoder fully connected neural network model includes:

[0040] Acquire first training data, which includes training category feature information;

[0041] The training category feature information is input into the autoencoder fully connected neural network model to output the iteration result;

[0042] The iterative results output by the autoencoder fully connected neural network model are input into the autoencoder fully connected neural network model to output new iterative results;

[0043] The process of repeatedly executing the step of inputting the iterative results output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output new iterative results;

[0044] The loss function is used to calculate the loss value between each input and the corresponding output of the autoencoder fully connected neural network model. If the loss value exceeds a first preset value, the parameters of the autoencoder fully connected neural network model are updated until the loss value is less than the first preset value, at which point the training of the autoencoder fully connected neural network model is stopped.

[0045] In one possible implementation, the training process of the task completion rate prediction model includes:

[0046] Obtain multiple interaction records, wherein the interaction records include customer information, staff personnel information, and interaction content between customers and staff;

[0047] Based on the interaction content between the customer and the staff in each interaction record, determine the actual completion rate corresponding to each interaction record;

[0048] Acquire second training data, wherein the second training data includes customer feature information of the customer corresponding to each interaction record and personnel feature information of the staff, wherein the customer feature information includes each numerical feature information and each category feature information of the customer, and the personnel feature information of the staff includes each numerical feature information and each category feature information of the staff.

[0049] Initialize the model parameters of the task completion rate prediction model;

[0050] Each numerical feature and each category feature of the customer, along with each numerical feature and each category feature of each employee, are input into the task completion rate prediction model to output the predicted completion rate for each employee.

[0051] The loss value between the predicted completion rate and the actual completion rate for each worker is calculated using a loss function.

[0052] If the loss value exceeds the second preset value, the model parameters of the task completion rate prediction model are updated until the second training data completes the training of the task completion rate prediction model.

[0053] Secondly, embodiments of this application provide a task allocation processing apparatus, the apparatus comprising:

[0054] The acquisition module is used to acquire target tasks to be assigned and determine the target customers corresponding to the target tasks.

[0055] The acquisition module is also used to acquire customer feature information of the target customer, wherein the customer feature information of the target customer includes each numerical feature information and each category feature information of the target customer;

[0056] The acquisition module is also used to acquire personnel characteristic information for each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member;

[0057] The processing module is used to normalize each numerical feature information of the target customer to obtain the numerical feature vector of the target customer.

[0058] The input module is used to input the category feature information of the target customer into the task completion rate prediction model. The task completion rate prediction model includes an embedding layer, a connection layer, a first module, a second module, and an output layer. The first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer. The task completion rate prediction model then performs the following steps:

[0059] The input module is also used to input each category feature information into the embedding layer to output the category feature vector of the target customer;

[0060] The input module is also used to input the numerical feature vector and category feature vector of the target customer into the connection layer to output the complete feature vector corresponding to the target customer;

[0061] The processing module is also used to normalize each numerical feature information of each worker to obtain the numerical feature vector of the worker.

[0062] The input module is also used to input each category feature information of each worker into the embedding layer for each worker, so as to output the category feature vector of the worker;

[0063] The input module is also used to input the numerical feature vector and category feature vector of each worker into the connection layer to output the complete feature vector corresponding to the worker.

[0064] The input module is also used to input the complete feature vector corresponding to each staff member and the complete feature vector corresponding to the target customer into the first module for fusion and activation processing, so as to output the first fused feature corresponding to the staff member;

[0065] The determination module is used to select reference staff from among the staff members and determine the complete feature vector corresponding to the reference staff member as the reference complete feature vector.

[0066] The input module is also used to input the complete feature vector corresponding to each staff member and the reference complete feature vector into the second module for linear transformation and dimensionality reduction processing, so as to output the second fused feature corresponding to the staff member;

[0067] The input module is also used to input the first fusion feature and the second fusion feature corresponding to each worker into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker;

[0068] The determination module is also used to compare the predicted completion rate of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member;

[0069] The sending module is used to send the target task to the terminal corresponding to the target worker.

[0070] Thirdly, embodiments of this application provide a computer device, including: a memory and a processor;

[0071] The memory stores the instructions that the computer executes;

[0072] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0073] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0074] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0075] The task allocation processing method and apparatus provided in this application normalize each numerical feature information of the target customer and each staff member to obtain numerical feature vectors of the target customer and each staff member. Each category feature information of the target customer and each staff member is input into the embedding layer of a task completion rate prediction model to output the category feature vectors of the target customer and each staff member. The numerical feature vectors and category feature vectors of the target customer and each staff member are input into the connection layer of the task completion rate prediction model to output the complete feature vectors corresponding to the target customer and each staff member. The complete feature vectors corresponding to each staff member and the target customer are input into the first module of the task completion rate prediction model. The process involves fusion and activation to output a first fusion feature for each worker. Reference workers are selected from among the workers, and their corresponding complete feature vectors are determined as reference complete feature vectors. Each worker and the reference complete feature vectors are input into the second module of the task completion rate prediction model for linear transformation and dimensionality reduction to output a second fusion feature for each worker. The first and second fusion features for each worker are then input into the output layer of the task completion rate prediction model for prediction to output the predicted completion rate of the target task for that worker. The target task is then sent to the terminal corresponding to the target worker with the highest predicted completion rate. This application, by predicting the completion rate of each worker for the target task and then assigning the target task to the worker with the highest predicted completion rate, ensures that the target task is completed with the highest probability, thereby improving the task completion rate. Attached Figure Description

[0076] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0077] Figure 1 This is a schematic diagram illustrating a scenario used in this application;

[0078] Figure 2 Flowchart of the task allocation processing method provided in this application Figure 1 ;

[0079] Figure 3 This is a schematic diagram of the structure of a task completion rate prediction model as an example;

[0080] Figure 4 A schematic diagram of the cross-product transform layer as an example;

[0081] Figure 5 Flowchart of the task allocation processing method provided in this application Figure 2 ;

[0082] Figure 6 Flowchart of the task allocation processing method provided in this application Figure 3 ;

[0083] Figure 7 Flowchart of the task allocation processing method provided in this application Figure 4 ;

[0084] Figure 8 Flowchart of the task allocation processing method provided in this application Figure 5 ;

[0085] Figure 9 A schematic diagram of the structure of an autoencoder fully connected neural network model as an example;

[0086] Figure 10 Flowchart of the task allocation processing method provided in this application Figure 6 ;

[0087] Figure 11 Flowchart of the task allocation processing method provided in this application Figure 7 ;

[0088] Figure 12 A schematic diagram of the task allocation processing device provided in this application;

[0089] Figure 13 A schematic diagram of the structure of the computer device provided in this application.

[0090] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0091] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0092] Figure 1 This is a schematic diagram illustrating a scenario from this application, such as... Figure 1As shown, the specific application scenarios of this application mainly include the server-side ① of the insurance sales management system and the client-side ② of the staff. The server-side ① can design a specific sales process based on the sales expectations of the insurance products, and send the sales tasks to the staff's client-side ② by specifying the sales tasks at different time points in the sales process, so as to assign the corresponding sales tasks to the staff, enabling the staff to complete the sales tasks and ultimately complete the sales of the insurance products.

[0093] Based on the above scenarios, it can be seen that existing insurance sales management systems allocate sales tasks in various ways, such as by task creation time, average workload, and previous supervisor. However, current methods of allocating sales tasks in insurance sales management systems only meet the basic needs of staff to perform their tasks, resulting in generally low completion rates for insurance product sales targets.

[0094] The task allocation processing method provided in this application first normalizes each numerical feature information of the target customer and each staff member to obtain numerical feature vectors for the target customer and each staff member. Then, it inputs each category feature information of the target customer and each staff member into the embedding layer of the task completion rate prediction model to output the category feature vectors of the target customer and each staff member. Next, it inputs the numerical feature vectors and category feature vectors of the target customer and each staff member into the connection layer of the task completion rate prediction model to output the complete feature vectors corresponding to the target customer and each staff member. Finally, it inputs the complete feature vectors corresponding to each staff member and the target customer into the first module of the task completion rate prediction model for fusion. The process involves several steps: first, fusion and activation processing to output a first fusion feature for each worker; then, selecting reference workers from among them and determining their complete feature vectors as reference complete feature vectors; next, inputting each worker and the reference complete feature vectors into the second module of the task completion rate prediction model for linear transformation and dimensionality reduction to output a second fusion feature for each worker; then, inputting the first and second fusion features of each worker into the output layer of the task completion rate prediction model for prediction processing to output the predicted completion rate of the target task for each worker; finally, sending the target task to the terminal corresponding to the target worker with the highest predicted completion rate. This application, by predicting the completion rate of each worker for the target task and then assigning the target task to the worker with the highest predicted completion rate, ensures that the target task is completed with the highest probability, thereby improving the task completion rate.

[0095] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of related data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0096] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0097] Figure 2 Flowchart of the task allocation processing method provided in this application Figure 1 ,like Figure 2 As shown, the method includes:

[0098] S201. Obtain the target task to be assigned and determine the target customer corresponding to the target task.

[0099] The execution subject of this embodiment is Figure 1 The server-side component.

[0100] With the example scenario, a dedicated sales task can be created for each customer to be sold to. For instance, if there are multiple customers to be sold to, such as user A, user B, and user C, a separate sales task can be created for each of them. The target task is the next sales task to be executed, and the customer corresponding to the target task is the target customer. For example, if the sales task for user A is the target task, then user A is the target customer.

[0101] S202. Obtain the customer feature information of the target customer, wherein the customer feature information of the target customer includes each numerical feature information and each category feature information of the target customer.

[0102] Based on scenario examples, numerical feature information includes, but is not limited to: age, family income, number of properties, number of vehicles, assets, liabilities, weight, height, current premiums paid, and the corresponding sum insured. Categorical feature information includes, but is not limited to: occupation, nationality, employer, residential address, education level, dietary habits, exercise habits, gender (male = 1, female = 0), family role - breadwinner (yes = 1, no = 0), family role - needing to support parents (yes = 1, no = 0), property type - owner-occupied home (yes = 1, no = 0), vehicle type - private car (yes = 1, no = 0), whether suffering from an illness (yes = 1, no = 0), whether having a surgical history (yes = 1, no = 0), whether smoking (yes = 1, no = 0), current insurance type - life insurance (yes = 1, no = 0), and current insurance type - medical insurance (yes = 1, no = 0), etc.

[0103] S203. Obtain the personnel characteristic information of each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member.

[0104] Based on the scenario example, the numerical feature information of each staff member is of the same type as the numerical feature information of the target customer, and the category feature information of each staff member is of the same type as the category feature information of the target customer.

[0105] S204. Normalize each numerical feature information of the target customer to obtain the numerical feature vector of the target customer.

[0106] Based on scenario examples, normalization is performed on each numerical feature of the target customer to avoid excessive attention to a particular numerical feature due to its large absolute value. Then, the results of normalization of each numerical feature are concatenated in a preset order to obtain the numerical feature vector of the target customer.

[0107] S205. Input the category feature information of the target customer into the task completion rate prediction model, wherein the task completion rate prediction model includes an embedding layer, a connection layer, a first module, a second module, and an output layer, wherein the first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer, so as to perform the following steps through the task completion rate prediction model:

[0108] Combined with scenario examples, Figure 3 A schematic diagram of the structure of a task completion rate prediction model is shown below. Figure 3As shown, the task completion rate prediction model includes an embedding layer, a connection layer, a first module, a second module, and an output layer. The first module comprises three fully connected layers, each a deep neural network with ReLU activation function, and dimensions of 1024, 512, and 256, respectively. The second module includes a cross-product transformation layer and a weight layer, and the output layer uses a Sigmoid activation function.

[0109] S206. Input each category feature information into the embedding layer to output the category feature vector of the target customer.

[0110] Based on the scenario example, the target customer's category features are divided into binary features and non-binary features. Binary features include gender (male = 1, female = 0), family role - breadwinner (yes = 1, no = 0), family role - needing to support parents (yes = 1, no = 0), property type - owner-occupied house (yes = 1, no = 0), vehicle type - private car (yes = 1, no = 0), whether they have a disease (yes = 1, no = 0), whether they have a surgical history (yes = 1, no = 0), whether they smoke (yes = 1, no = 0), current insurance type - life insurance (yes = 1, no = 0), and current insurance type - medical insurance (yes = 1, no = 0). These features are already numerical and only have two values: 0 and 1. For binary features, since they are already numerical, no further embedding layer encoding is needed. For non-binary features, such as occupation, nationality, workplace, residential address, education level, dietary habits, and exercise habits, all non-binary features can be concatenated into a temporary vector in a preset order and then input into the embedding layer. The embedding layer encodes these features and outputs a 20-dimensional encoded vector. In this example, the 20-dimensional encoded vector output by the embedding layer is used as the target user's category feature vector. The fully connected layer of the encoder can encode each non-binary feature using one-hot encoding. Taking occupation as an example, in the embedding layer, there can be 10 occupations: Occupation 1, Occupation 2... Occupation 10. If the target customer's occupation is Occupation 1, then the target customer's occupation, after one-hot encoding by the embedding layer, will be (1, 0, 0, 0, 0, 0, 0, 0, 0). Similarly, for residential addresses, the embedding layer can have N possible values: Residential Address 1, Residential Address 2, ..., Residential Address N. If the target customer's residential address is Residential Address 1, then after one-hot encoding in the embedding layer, the target customer's residential address will be (0, 1, 0, 0, 0...). In this way, the temporary vector is encoded through the embedding layer. The encoded result of the temporary vector is then concatenated with each binary feature to obtain the target customer's category feature vector.

[0111] S207. Input the numerical feature vector and category feature vector of the target customer into the connection layer to output the complete feature vector corresponding to the target customer.

[0112] Combined with scenario examples, Figure 3 The numerical feature vector and categorical feature vector of the target customer are input into the connection layer and concatenated to obtain the complete feature vector of the target customer.

[0113] S208. For each staff member, normalize each numerical feature information of the staff member to obtain the numerical feature vector of the staff member.

[0114] In the example scenario, the processing method for the numerical features of each staff member is consistent with the processing method for the numerical features of the target customer. Specifically, for each staff member, each of their corresponding numerical feature information is normalized to avoid excessive attention to any single numerical feature information with an excessively large absolute value. Then, the results of the normalized numerical feature information are concatenated in a preset order to obtain the numerical feature vector corresponding to the staff member.

[0115] S209. For each staff member, input each category feature information of the staff member into the embedding layer to output the category feature vector of the staff member.

[0116] In the example scenario, the processing method for the category feature information of each employee is consistent with the processing method for the category feature information of the target customer. Specifically, for each employee, the employee's category features are divided into binary features and non-binary features. Each non-binary feature of the employee is first concatenated into a temporary vector, and then the temporary vector is input into the embedding layer to complete the encoding of the temporary vector. The encoding result of the temporary vector is concatenated with each binary feature to obtain the employee's category feature vector.

[0117] S2010. For each staff member, the numerical feature vector and category feature vector of the staff member are input into the connection layer to output the complete feature vector corresponding to the staff member.

[0118] Combined with scenario examples, Figure 3 For each staff member, the numerical feature vector and the categorical feature vector of the staff member are input into the connection layer and concatenated to obtain the complete feature vector of the staff member.

[0119] S2011. For each staff member, the complete feature vector corresponding to the staff member and the complete feature vector corresponding to the target customer are input into the first module for fusion and activation processing, so as to output the first fused feature corresponding to the staff member.

[0120] In the example scenario, the complete feature vectors of the target customer and the complete feature vectors of the staff are connected in the connection layer and then input into the first module. The first module consists of three deep neural networks with ReLU as the activation function, with dimensions of 1024, 512 and 256 respectively. Therefore, after processing by the first module, the first fused feature with a dimension of 256 corresponding to the staff can be obtained.

[0121] S2012. Select reference staff from among the staff members, and determine the complete feature vector corresponding to the reference staff member as the reference complete feature vector.

[0122] In a scenario example, the reference worker can be the worker who has completed the most sales tasks. For example, if worker A is the worker who has completed the most sales tasks, then worker A can be used as the reference worker, and worker A's complete feature vector can be used as the reference complete feature vector.

[0123] S2013. For each staff member, the complete feature vector corresponding to the staff member and the reference complete feature vector are input into the second module for linear transformation and dimensionality reduction processing to output the second fused feature corresponding to the staff member.

[0124] Combined with scenario examples, Figure 3 The second module includes a cross-product transform layer and a weight layer. For each worker, the obtained complete feature vector and a reference complete feature vector are input into the second module, and the cross-product transform layer performs a linear transformation on the worker's complete feature vector and the reference complete feature vector. Figure 4 A schematic diagram of the cross-product transform layer is shown below. Figure 4 As shown, the cross-product transform layer can perform multiple cross-product transforms. Specifically, the output vector of each layer is calculated using the following steps.

[0125] x1=x0x T w0+b0+x

[0126]

[0127]

[0128] Where x0 is the reference complete feature vector, x is the complete feature vector of the staff, and x l For each layer's output vector, w0, w l These are the weight vectors for the first layer and the corresponding layer, b0, b l These are the bias vectors for the first layer and the corresponding layer, respectively.

[0129] Then, the last layer, which is xl The output vector of the layer is linearly transformed to obtain the result of the linear transformation. Specifically, x... l The layer's output vector is multiplied by a preset weight matrix and then added to a preset bias term to obtain x. l The result of the linear transformation corresponding to the output vector of the layer.

[0130] The result of the linear transformation is then adjusted in dimensionality by a weighting layer. Specifically, the result of the linear transformation is adjusted to 256 dimensions by the weighting layer to obtain the second fusion feature of the staff.

[0131] S2014. For each worker, the first fusion feature and the second fusion feature corresponding to the worker are input into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker.

[0132] Based on the scenario example, the first and second fusion features of each worker are added together in the Sigmoid output layer and then passed through another layer to produce a one-dimensional output. After passing through the Sigmoid activation function, the final output is the prediction completion rate of each worker for the target task, which is in the range of 0 to 1.

[0133] S2015. Compare the predicted completion rate of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member.

[0134] Based on the scenario example, the higher the prediction completion rate of a staff member, the greater the probability that the staff member will complete the target task. Therefore, the staff member with the highest prediction completion rate is selected as the target staff member.

[0135] S2016. Send the target task to the terminal corresponding to the target staff member.

[0136] Based on the scenario example, each staff member has their own corresponding terminal, which can be a server. The target task should be assigned to the target staff member, so the target task can be sent to the target staff member's terminal so that the target staff member can receive the target task through the corresponding terminal.

[0137] This example demonstrates how to predict each employee's completion rate for a target task and then assign the task to the employee with the highest predicted completion rate. This ensures that the target task is completed with the highest probability, thereby increasing the task completion rate.

[0138] Optional, Figure 5 Flowchart of the task allocation processing method provided in this application Figure 2 ,like Figure 5 As shown, S204 includes:

[0139] S501. Obtain the preset upper limit value and preset lower limit value corresponding to each numerical feature information of the target customer.

[0140] Based on scenario examples, for the target customer's age, family income, number of properties, number of vehicles, assets, liabilities, weight, height, current premiums paid and insurance coverage, obtain the preset upper limit and preset lower limit values ​​for each numerical feature.

[0141] S502. Compare the absolute value of each numerical feature with the difference between the corresponding preset upper limit value and preset lower limit value to obtain the ratio result corresponding to each numerical feature.

[0142] In conjunction with a scenario example, as shown in the following formula, when normalizing each numerical feature to obtain the normalized value, the absolute value of the target customer's numerical feature is compared with the difference between the preset upper limit value and the preset lower limit value.

[0143]

[0144] For example, taking the age of the target customer as an example, if the target customer is 40 years old, and the preset upper limit and preset lower limit of age are 60 years old and 18 years old respectively, then the normalized value of the target customer's age after normalization is 95%.

[0145] S503. The ratio results corresponding to each numerical feature information of the target customer are concatenated to obtain the numerical feature vector of the target customer.

[0146] Based on the scenario example, by using the above formula, after normalizing each numerical feature information of the target customer, the normalized value corresponding to each numerical feature information is obtained. The normalized values ​​corresponding to each numerical feature information are then concatenated in a preset order to obtain the numerical feature vector corresponding to the target customer.

[0147] Optional, Figure 6 Flowchart of the task allocation processing method provided in this application Figure 3 ,like Figure 6 As shown, S208 includes:

[0148] S601. Obtain the preset upper limit value and preset lower limit value corresponding to each numerical feature information of the staff member.

[0149] Based on the scenario example, the numerical feature information of each staff member is normalized sequentially according to the method used to process the numerical feature information of the target customer. Specifically, for each staff member, the preset upper limit value and preset lower limit value for each numerical feature corresponding to the staff member are obtained.

[0150] S602. Compare the absolute value of each numerical feature with the difference between the corresponding preset upper limit value and preset lower limit value to obtain the ratio result corresponding to each numerical feature.

[0151] Using the above formula and combining it with the scenario example, we can obtain the normalized value corresponding to each numerical feature information of the staff.

[0152] S603. The ratio results corresponding to each numerical feature information of the staff are concatenated to obtain the numerical feature vector of the staff.

[0153] Based on the scenario example, the normalized values ​​corresponding to each numerical feature information of each staff member are concatenated in a preset order to obtain the numerical feature vector of each staff member.

[0154] Based on the method provided in this example, normalizing each numerical feature information corresponding to the target customer and each staff member can avoid misleading the task completion rate prediction model by focusing too much on some numerical features due to excessively large absolute values.

[0155] Optional, Figure 7 Flowchart of the task allocation processing method provided in this application Figure 4 ,like Figure 7 As shown, in step S2012, the selection of reference staff from among the staff includes:

[0156] S701. Obtain task completion records, wherein the task completion records include each completed historical task and the staff member who completed each historical task.

[0157] Based on the scenario example, the task completion record can be in the form of a table. The historical tasks represent the sales tasks that have been assigned to the corresponding staff members, and the task completion record is mainly responsible for recording each completed sales task and the corresponding staff member.

[0158] S702. The staff member who has completed the most historical tasks is identified as the reference staff member.

[0159] Based on the scenario example, the task completion record can be in the form of a table. For example, there is one task completion record for each month. The staff member with the most recorded historical tasks can be determined as the reference staff member based on the historical tasks recorded in the task completion record of the current month.

[0160] Based on the method provided in this example, reference workers can be identified based on task completion records.

[0161] Optionally, the embedding layer includes an autoencoder fully connected neural network model;

[0162] Accordingly, Figure 8 Flowchart of the task allocation processing method provided in this application Figure 5 ,like Figure 8 As shown, S206 includes:

[0163] S801. Input each category feature information into the autoencoder fully connected neural network model to output the category feature vector of the target customer;

[0164] Based on the scenario example, the embedding layer can be implemented using an autoencoder fully connected neural network model. Figure 9 A schematic diagram of the structure of an autoencoder fully connected neural network model is shown below. Figure 9 As shown, the autoencoder fully connected neural network model includes a fully connected layer for the encoder and a fully connected layer for the decoder. After inputting the category feature information of each target customer into the autoencoder fully connected neural network model, the fully connected layer of the encoder encodes it to output a 20-dimensional encoded vector, which serves as the category feature vector of the target customer. This 20-dimensional encoded vector can then pass through the fully connected layer of the decoder, producing an output vector with the same dimension as the temporary vector.

[0165] Accordingly, S209 includes:

[0166] S802. The category feature information of each staff member is input into the autoencoder fully connected neural network model to output the category feature vector of the staff member.

[0167] Based on the scenario example, consistent with the processing method of the category feature information of the target customer, for each staff member, after inputting each category feature information of the staff member into the autoencoder fully connected neural network model, the fully connected layer of the encoder encodes it to output a 20-dimensional encoded vector, which is used as the category feature vector of the staff member.

[0168] Based on the method provided in this example, the feature information of the target customer and each staff member can be encoded using an autoencoder fully connected neural network model to obtain the corresponding category feature vector.

[0169] Optional, Figure 10 Flowchart of the task allocation processing method provided in this application Figure 6 ,like Figure 10 As shown, the training process of the autoencoder fully connected neural network model includes:

[0170] S1001. Obtain first training data, the first training data including training category feature information.

[0171] Based on the scenario example, the first training data is used to train the autoencoder fully connected neural network model. The autoencoder fully connected neural network model is used to encode category feature information, so the first training data may include multiple preset training category feature information.

[0172] S1002. Input the training category feature information into the autoencoder fully connected neural network model to output the iteration result.

[0173] Using a scenario example, the feature information of each training category included in the first training data is concatenated in a preset order to obtain a temporary training vector. This temporary training vector is then input into an autoencoder fully connected neural network model. The encoder in the autoencoder fully connected neural network model encodes the temporary training vector to obtain the first iteration result, and then outputs the iteration result.

[0174] S1003. Input the iterative result output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output a new iterative result.

[0175] In a scenario example, the result of the first iteration is used as input to the autoencoder fully connected neural network model. The encoder in the autoencoder fully connected neural network model encodes the output of the first iteration result to obtain a new iteration result, which is the second iteration result, and then outputs the second iteration result.

[0176] S1004. The process of repeatedly executing the step of inputting the iterative results output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output new iterative results.

[0177] Using a scenario example, the second iteration result output by the autoencoder fully connected neural network model is again taken as input and fed into the autoencoder fully connected neural network model. The encoder in the autoencoder fully connected neural network model encodes the second iteration result and obtains a new iteration result, which is the third iteration result, and outputs the third iteration result. This process of continuously executing the output iteration result as input to the autoencoder fully connected neural network model to output a new iteration result is repeated.

[0178] S1005. Calculate the loss value between the input and the corresponding output of the autoencoder fully connected neural network model for each input using a loss function. If the loss value exceeds a first preset value, update the parameters of the autoencoder fully connected neural network model until the loss value is less than the first preset value, then stop training the autoencoder fully connected neural network model.

[0179] Based on the scenario example, mean squared error can be used as the loss function to calculate the loss value between each input and output during the training of the autoencoder fully connected neural network model. The first preset value can be determined according to the actual situation; in this example, the first preset value can be set to 2%. If the loss value between each input and output exceeds 2%, the gradient of the loss function relative to the model parameters is calculated using backpropagation of the calculated loss value to update it, and the loop process of S1005 continues. If the loss value between each input and output is less than or equal to 2%, the loop process of S1005 can be stopped to obtain the trained autoencoder fully connected neural network model. Based on the method provided in this example, the training of the autoencoder fully connected neural network model can be completed.

[0180] Optional, Figure 11 Flowchart of the task allocation processing method provided in this application Figure 7 ,like Figure 11 As shown, the training process of the task completion rate prediction model includes:

[0181] S1101. Obtain multiple interaction records, wherein the interaction records include customer information, staff personnel information, and interaction content between customers and staff.

[0182] Based on the scenario example, the interaction record is the interaction information between the staff and the customer when the staff completes the sales task. It includes the staff information and customer information corresponding to the interaction record, as well as the interaction content between the staff and the customer. The interaction content represents the communication information between the staff and the customer, which can be text information or voice information.

[0183] S1102. Based on the interaction content between the customer and the staff in each interaction record, determine the actual completion rate corresponding to each interaction record.

[0184] In this example, the actual completion rate mentioned is a preset probability and does not represent the customer's actual consumption of the product. In other words, the customer's intention to purchase the product can be determined based on the interaction between the customer and staff, and a corresponding actual completion rate can be preset based on this intention. The actual completion rate is between 0 and 1. If the customer shows a strong intention to purchase based on the interaction, but the actual purchase does not occur, the actual completion rate is not 0, but rather a value predicted based on the customer's purchase intention, such as 90%.

[0185] S1103. Obtain second training data, wherein the second training data includes customer feature information of the customer corresponding to each interaction record and personnel feature information of the staff, wherein the customer feature information includes each numerical feature information and each category feature information of the customer, and the personnel feature information of the staff includes each numerical feature information and each category feature information of the staff.

[0186] Based on the scenario example, the second training data includes multiple interaction records. According to the interaction content between customers and staff included in each interaction record, the actual completion rate corresponding to each interaction record is preset.

[0187] S1104. Initialize the model parameters of the task completion rate prediction model.

[0188] Using a scenario example, before training the task completion rate prediction model, we first initialize each parameter in the model to obtain the initial value of each parameter.

[0189] S1105. Input each numerical feature information and each category feature information of the customer, and each numerical feature information and each category feature information of the staff member, into the task completion rate prediction model to output the predicted completion rate for each staff member.

[0190] Following the scenario example and the process described in the above embodiment, the processing of each interaction record in the second training data is as follows: Normalize each numerical feature information of the customer to obtain the customer's numerical feature vector; encode each category feature information of the customer through an embedding layer to obtain the customer's category feature vector; similarly, obtain the numerical feature vector and category feature vector of the staff. After concatenating the customer's numerical feature vector and category feature vector, and the staff's numerical feature vector and category feature vector, in a connection layer, process them through the first module to output a first fused feature. Process the staff's numerical feature vector and category feature vector, as well as a reference staff's numerical feature vector and category feature vector, through the second module to obtain a second fused feature. Process the first fused feature and the second fused feature through a Sigmoid output layer to obtain the corresponding prediction completion rate. In this way, the prediction completion rate corresponding to the staff in each interaction record in each of the second training data is obtained.

[0191] S1106. Calculate the loss value between the predicted completion rate and the actual completion rate for each worker using the loss function.

[0192] Using a scenario example, the loss value between the predicted completion rate and the actual completion rate is calculated using the mean squared error loss function.

[0193] S1107. If the loss value exceeds the second preset value, the model parameters of the task completion rate prediction model are updated until the second training data completes the training of the task completion rate prediction model.

[0194] Based on the scenario example, the second training data can be divided into multiple batches. If the second training data contains 1000 interaction records, these 1000 records can be divided into 100 batches. When training the task completion rate prediction model, after initializing the parameters of the model, one batch is randomly selected from the 100 batches to begin iterative training. If the loss value between the predicted completion rate and the actual completion rate for each interaction record exceeds a second preset value, the parameters in the task completion rate prediction model are updated through backpropagation. After all 1000 interaction records have completed training, the training of the task completion rate prediction model ends, resulting in a trained task completion rate prediction model. Based on the method provided in this example, the training of a task completion rate prediction model can be completed.

[0195] Figure 12 A schematic diagram of the task allocation processing device provided in this application is shown below. Figure 12 As shown, the task allocation processing device includes:

[0196] The acquisition module 121 is used to acquire the target task to be assigned and determine the target customer corresponding to the target task;

[0197] The acquisition module 121 is further configured to acquire customer feature information of the target customer, wherein the customer feature information of the target customer includes each numerical feature information and each category feature information of the target customer;

[0198] The acquisition module 121 is also used to acquire personnel characteristic information of each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member;

[0199] The processing module 122 is used to normalize each numerical feature information of the target customer to obtain the numerical feature vector of the target customer.

[0200] Input module 123 is used to input the category feature information of the target customer into the task completion rate prediction model, wherein the task completion rate prediction model includes an embedding layer, a connection layer, a first module, a second module, and an output layer, wherein the first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer, so as to perform the following steps through the task completion rate prediction model:

[0201] The input module 123 is also used to input each category feature information into the embedding layer to output the category feature vector of the target customer;

[0202] The input module 123 is also used to input the numerical feature vector and category feature vector of the target customer into the connection layer to output the complete feature vector corresponding to the target customer;

[0203] The processing module 122 is also used to normalize each numerical feature information of each worker to obtain the numerical feature vector of the worker.

[0204] The input module 123 is also configured to input each category feature information of each worker into the embedding layer for each worker, so as to output the category feature vector of the worker;

[0205] The input module 123 is also used to input the numerical feature vector and category feature vector of each worker into the connection layer to output the complete feature vector corresponding to the worker;

[0206] The input module 123 is also used to input the complete feature vector corresponding to the staff member and the complete feature vector corresponding to the target customer into the first module for fusion and activation processing for each staff member, so as to output the first fused feature corresponding to the staff member;

[0207] The determination module 124 is used to select reference staff from among the staff members and determine the complete feature vector corresponding to the reference staff member as the reference complete feature vector.

[0208] The input module 123 is also used to input the complete feature vector corresponding to the staff member and the reference complete feature vector into the second module for linear transformation and dimensionality reduction processing for each staff member, so as to output the second fused feature corresponding to the staff member;

[0209] The input module 123 is also used to input the first fusion feature and the second fusion feature corresponding to each worker into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker;

[0210] The determining module 124 is also used to compare the predicted completion rate of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member;

[0211] The sending module 125 is used to send the target task to the terminal corresponding to the target worker.

[0212] Optionally, the processing module 122 is specifically used to obtain the preset upper limit value and preset lower limit value corresponding to each numerical feature information of the target customer;

[0213] The processing module 122 is further used to compare the absolute value of each numerical feature information with the difference between the corresponding preset upper limit value and preset lower limit value, so as to obtain the ratio result corresponding to each numerical feature information.

[0214] The processing module 122 is further configured to concatenate the ratio results corresponding to each numerical feature information of the target customer to obtain the numerical feature vector of the target customer.

[0215] The processing module 122 is further configured to normalize each numerical feature information of the staff member to obtain the numerical feature vector of the staff member, including:

[0216] The processing module 122 is further configured to obtain the preset upper limit value and preset lower limit value corresponding to each numerical feature information of the staff member;

[0217] The processing module 122 is further used to compare the absolute value of each numerical feature information with the difference between the corresponding preset upper limit value and preset lower limit value, so as to obtain the ratio result corresponding to each numerical feature information.

[0218] The processing module 122 is further configured to concatenate the ratio results corresponding to each numerical feature information of the staff member to obtain the numerical feature vector of the staff member.

[0219] Optionally, module 124 is specifically used to obtain task completion records, wherein the task completion records include each completed historical task and the staff member who completed each historical task;

[0220] The determination module 124 is further used to determine the staff member who has completed the most historical tasks as the reference staff member.

[0221] Optionally, the embedding layer includes an autoencoder fully connected neural network model;

[0222] Optionally, the input module 123 is specifically used to input each category feature information into the autoencoder fully connected neural network model to output the category feature vector of the target customer;

[0223] The input module 123 is further used to input each category feature information of the staff into the autoencoder fully connected neural network model to output the category feature vector of the staff.

[0224] Optionally, the processing module 122 is further configured to acquire first training data, which includes training category feature information;

[0225] The processing module 122 is further configured to input the training category feature information into the autoencoder fully connected neural network model to output the iteration result;

[0226] The processing module 122 is further configured to input the iterative result output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output a new iterative result;

[0227] The processing module 122 is further configured to repeatedly execute the process of inputting the iterative results output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output new iterative results;

[0228] The processing module 122 is further configured to calculate the loss value between the input and the corresponding output of the autoencoder fully connected neural network model each time using a loss function. If the loss value exceeds a first preset value, the parameters of the autoencoder fully connected neural network model are updated until the loss value is less than the first preset value, at which point the training of the autoencoder fully connected neural network model is stopped.

[0229] Optionally, the processing module 122 is further configured to acquire multiple interaction records, wherein the interaction records include customer information, staff personnel information, and interaction content between the customer and the staff.

[0230] The processing module 122 is also used to determine the actual completion rate of each interaction record based on the interaction content between the customer and the staff in each interaction record;

[0231] The processing module 122 is further configured to acquire second training data, wherein the second training data includes customer feature information of the customer corresponding to each interaction record and personnel feature information of the staff, wherein the customer feature information includes each numerical feature information and each category feature information of the customer, and the personnel feature information of the staff includes each numerical feature information and each category feature information of the staff.

[0232] The processing module 122 is further used to initialize the model parameters of the task completion rate prediction model;

[0233] The processing module 122 is further configured to input each numerical feature information and each category feature information of the customer and each numerical feature information and each category feature information of the staff member into the task completion rate prediction model, so as to output the predicted completion rate for each staff member;

[0234] The processing module 122 is also specifically used to calculate the loss value between the predicted completion rate and the actual completion rate for each worker through a loss function;

[0235] The processing module 122 is further configured to update the model parameters of the task completion rate prediction model if the loss value exceeds a second preset value, until the second training data completes the training of the task completion rate prediction model.

[0236] The task allocation processing device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0237] Figure 13 A schematic diagram of the structure of the computer device provided in this application. Figure 13As shown, the computer device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0238] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0239] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0240] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0241] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0242] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0243] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0244] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0245] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0246] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0247] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0248] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0249] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0250] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0251] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0252] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A task allocation and processing method, characterized in that, The method is applied to the server side and includes: Obtain the target tasks to be assigned and determine the target customers corresponding to the target tasks; Obtain customer characteristic information of the target customer, wherein the customer characteristic information of the target customer includes each numerical characteristic information and each category characteristic information of the target customer; Obtain the personnel characteristic information of each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member; Each numerical feature of the target customer is normalized to obtain the numerical feature vector of the target customer. The target customer's category feature information is input into the task completion rate prediction model, which includes an embedding layer, a connection layer, a first module, a second module, and an output layer. The first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer. The task completion rate prediction model then performs the following steps: Each category feature is input into the embedding layer to output the category feature vector of the target customer; The numerical feature vector and category feature vector of the target customer are input into the connection layer to output the complete feature vector corresponding to the target customer; For each staff member, each numerical feature information of the staff member is normalized to obtain the numerical feature vector of the staff member; For each worker, the category feature information of each worker is input into the embedding layer to output the category feature vector of the worker; For each worker, the numerical feature vector and category feature vector of the worker are input into the connection layer to output the complete feature vector corresponding to the worker; For each staff member, the complete feature vector corresponding to the staff member and the complete feature vector corresponding to the target customer are input into the first module for fusion and activation processing, so as to output the first fused feature corresponding to the staff member; Select reference staff from among the staff, and determine the complete feature vectors corresponding to the reference staff as reference complete feature vectors; For each staff member, the complete feature vector corresponding to the staff member and the reference complete feature vector are input into the second module for linear transformation and dimensionality reduction processing to output the second fused feature corresponding to the staff member; For each worker, the first fusion feature and the second fusion feature corresponding to the worker are input into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker; Compare the predicted completion rates of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member; The target task is sent to the terminal corresponding to the target staff member.

2. The method according to claim 1, characterized in that, The step of normalizing each numerical feature information of the target customer to obtain the numerical feature vector of the target customer includes: Obtain the preset upper limit and preset lower limit values ​​corresponding to each numerical feature information of the target customer; The absolute value of each numerical feature is compared with the difference between the corresponding preset upper limit and preset lower limit to obtain the ratio result for each numerical feature. The ratio results corresponding to each numerical feature information of the target customer are concatenated to obtain the numerical feature vector of the target customer; The step of normalizing each numerical feature information of the staff member to obtain the numerical feature vector of the staff member includes: Obtain the preset upper limit and preset lower limit values ​​corresponding to each numerical feature information of the staff member; The absolute value of each numerical feature is compared with the difference between the corresponding preset upper limit and preset lower limit to obtain the ratio result for each numerical feature. The ratio results corresponding to each numerical feature information of the staff member are concatenated to obtain the numerical feature vector of the staff member.

3. The method according to claim 1, characterized in that, The selection of reference staff from among the staff includes: Obtain task completion records, wherein the task completion records include each completed historical task and the staff member who completed each historical task; The staff member who completed the most historical tasks was identified as the reference staff member.

4. The method according to claim 1, characterized in that, The embedding layer includes an autoencoder fully connected neural network model; Accordingly, the step of inputting each category feature information into the embedding layer to output the category feature vector of the target customer includes: Each category feature information is input into the autoencoder fully connected neural network model to output the category feature vector of the target customer; Accordingly, the step of inputting each category feature information of the staff member into the embedding layer to output the category feature vector of the staff member includes: Each category feature information of the staff member is input into the autoencoder fully connected neural network model to output the category feature vector of the staff member.

5. The method according to claim 4, characterized in that, The training process of the autoencoder fully connected neural network model includes: Acquire first training data, which includes training category feature information; The training category feature information is input into the autoencoder fully connected neural network model to output the iteration result; The iterative results output by the autoencoder fully connected neural network model are input into the autoencoder fully connected neural network model to output new iterative results; The process of repeatedly executing the step of inputting the iterative results output by the autoencoder fully connected neural network model into the autoencoder fully connected neural network model to output new iterative results; The loss function is used to calculate the loss value between each input and the corresponding output of the autoencoder fully connected neural network model. If the loss value exceeds a first preset value, the parameters of the autoencoder fully connected neural network model are updated until the loss value is less than the first preset value, at which point the training of the autoencoder fully connected neural network model is stopped.

6. The method according to any one of claims 1-5, characterized in that, The training process of the task completion rate prediction model includes: Obtain multiple interaction records, wherein the interaction records include customer information, staff personnel information, and interaction content between customers and staff; Based on the interaction content between the customer and the staff in each interaction record, determine the actual completion rate corresponding to each interaction record; Acquire second training data, wherein the second training data includes customer feature information of the customer corresponding to each interaction record and personnel feature information of the staff, wherein the customer feature information includes each numerical feature information and each category feature information of the customer, and the personnel feature information of the staff includes each numerical feature information and each category feature information of the staff. Initialize the model parameters of the task completion rate prediction model; Each numerical feature and each category feature of the customer, along with each numerical feature and each category feature of each employee, are input into the task completion rate prediction model to output the predicted completion rate for each employee. The loss value between the predicted completion rate and the actual completion rate for each worker is calculated using a loss function. If the loss value exceeds the second preset value, the model parameters of the task completion rate prediction model are updated until the second training data completes the training of the task completion rate prediction model.

7. A task allocation and processing device, characterized in that, The device includes: The acquisition module is used to acquire target tasks to be assigned and determine the target customers corresponding to the target tasks. The acquisition module is also used to acquire customer feature information of the target customer, wherein the customer feature information of the target customer includes each numerical feature information and each category feature information of the target customer; The acquisition module is also used to acquire personnel characteristic information for each staff member, wherein the personnel characteristic information of the staff member includes each numerical characteristic information and each category characteristic information of the staff member; The processing module is used to normalize each numerical feature information of the target customer to obtain the numerical feature vector of the target customer. The input module is used to input the category feature information of the target customer into the task completion rate prediction model. The task completion rate prediction model includes an embedding layer, a connection layer, a first module, a second module, and an output layer. The first module includes multiple fully connected layers, and the second module includes a cross-product transform layer and a weight layer. The task completion rate prediction model then performs the following steps: The input module is also used to input each category feature information into the embedding layer to output the category feature vector of the target customer; The input module is also used to input the numerical feature vector and category feature vector of the target customer into the connection layer to output the complete feature vector corresponding to the target customer; The processing module is also used to normalize each numerical feature information of each worker to obtain the numerical feature vector of the worker. The input module is also used to input each category feature information of each worker into the embedding layer for each worker, so as to output the category feature vector of the worker; The input module is also used to input the numerical feature vector and category feature vector of each worker into the connection layer to output the complete feature vector corresponding to the worker. The input module is also used to input the complete feature vector corresponding to each staff member and the complete feature vector corresponding to the target customer into the first module for fusion and activation processing, so as to output the first fused feature corresponding to the staff member; The determination module is used to select reference staff from among the staff members and determine the complete feature vector corresponding to the reference staff member as the reference complete feature vector. The input module is also used to input the complete feature vector corresponding to each staff member and the reference complete feature vector into the second module for linear transformation and dimensionality reduction processing, so as to output the second fused feature corresponding to the staff member; The input module is also used to input the first fusion feature and the second fusion feature corresponding to each worker into the output layer for prediction processing, so as to output the prediction completion rate of the target task corresponding to the worker; The determination module is also used to compare the predicted completion rate of each staff member for the target task, and determine the staff member with the highest predicted completion rate as the target staff member; The sending module is used to send the target task to the terminal corresponding to the target worker.

8. A computer device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

Images