Artificial intelligence-based task allocation method and device, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence and the field of financial science and technology, and relates to a task allocation method and device based on artificial intelligence, computer equipment and a storage medium. Calling a capability prediction model to predict the historical performance data of the online seat and screening a target prediction result of the online seat corresponding to the target business field; generating a domain score of the online seat based on the target prediction result; screening a first seat of which the domain score is greater than an skilled score threshold from the online seats; generating a comprehensive processing score based on the task load data of the first seat and the target domain score; and obtaining the target comprehensive processing score with the maximum value in the comprehensive processing scores. And if the target comprehensive processing score is greater than a score threshold, screening a target seat from the first seats based on the target comprehensive processing score, and allocating the task to the target seat. According to the invention, the accuracy and intelligence of task allocation are improved, and the task processing efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and the field of financial technology, and in particular to an artificial intelligence-based task allocation method, device, computer equipment and storage medium. Background Art

[0002] Call centers based on telephone communication are an important way for insurance companies to communicate with their customers. When customers encounter problems in financial insurance scenarios and need consultation, they trigger a call task to the call center. Properly handling the customer's call task is an important responsibility of the call center.

[0003] However, at present, for call tasks triggered by customers, the task processing method adopted by insurance companies is usually to randomly assign seats to customers to handle the customer's call tasks. Such task processing method often cannot quickly solve customer needs, and cannot accurately assign an adapted seat to the customer to perform service processing for the customer's call tasks, resulting in low intelligence and accuracy in the allocation of call tasks, low efficiency in the processing of call tasks, and the service quality of the service processing of call tasks cannot be guaranteed, resulting in low customer satisfaction. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a task allocation method, device, computer equipment and storage medium based on artificial intelligence to solve the technical problem that the existing task processing method cannot accurately allocate an adapted seat to the customer to perform service processing for the customer's call task, resulting in low intelligence and accuracy in the allocation of call tasks, low efficiency in the processing of call tasks, and the service quality of the service processing of call tasks cannot be guaranteed, resulting in low customer satisfaction.

[0005] In order to solve the above technical problems, the embodiment of the present application provides a task allocation method based on artificial intelligence, which adopts the following technical solutions:

[0006] Upon receiving a task to be assigned, obtaining a target business domain of the task;

[0007] Calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filtering out the target prediction results of each online agent corresponding to the target business field from the prediction results;

[0008] Generating a domain score for each online agent corresponding to the target business domain based on the target prediction result;

[0009] Filter out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents;

[0010] generating a comprehensive processing score for the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain;

[0011] Obtaining a target comprehensive processing score with the largest value among the comprehensive processing scores, and determining whether the target comprehensive processing score is greater than a preset score threshold;

[0012] If yes, a target seat is selected from the first seats based on the target comprehensive processing score, and the task is assigned to the target seat.

[0013] Furthermore, the step of selecting a target seat from the first seats based on the target comprehensive processing score specifically includes:

[0014] Selecting a second seat from the first seat that matches the target comprehensive processing score;

[0015] Determine whether the number of the second seat is 1;

[0016] If the number of the second seat is 1, the second seat is used as the target seat;

[0017] If the number of second seats is greater than 1, obtaining the idle time of each of the second seats;

[0018] Select the third seat with the longest idle time from all the second seats;

[0019] The third seat is used as the target seat.

[0020] Furthermore, the step of generating a comprehensive processing score of the first agent based on the task load data of the first agent and the target domain score of the first agent corresponding to the target business domain specifically includes:

[0021] Acquire task load data of the first seat, and determine a first preset weight corresponding to the task load data;

[0022] Obtaining a target domain score of the first seat corresponding to the target business domain, and determining a second preset weight corresponding to the target domain score;

[0023] Get the preset score calculation formula;

[0024] Based on the score calculation formula, the task load data, the first preset weight, the target domain score and the second preset weight are calculated to obtain a corresponding calculation result;

[0025] The calculation result is used as the comprehensive processing score of the first seat.

[0026] Furthermore, the step of generating a domain score for each online agent corresponding to the target business domain based on the target prediction result specifically includes:

[0027] Obtaining a specified prediction result corresponding to a specified seat; wherein the specified seat is any one of all the online seats;

[0028] Get the preset conversion factor;

[0029] Calculating the product between the specified prediction result and the conversion factor;

[0030] The product is used as the designated field score of the designated seat corresponding to the target business field.

[0031] Furthermore, before the step of calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, the method further includes:

[0032] Get agent performance data within a preset time period;

[0033] Performing sample construction processing on the agent performance data to obtain corresponding sample data;

[0034] Dividing the sample data into a training set and a test set;

[0035] Training a preset classification model based on the training set to obtain a corresponding specified classification model;

[0036] Performing model evaluation on the specified classification model based on the test set;

[0037] If the designated classification model passes the model evaluation, the designated classification model is used as the capability prediction model.

[0038] Furthermore, the step of performing model evaluation on the specified classification model based on the test set specifically includes:

[0039] Input the test set into the specified classification model for performance testing, and obtain a performance indicator value of the specified classification model; wherein the performance indicator value is a numerical value corresponding to a preset performance indicator, and the number of the performance indicators includes multiple;

[0040] Determine whether all the performance indicator values ​​are greater than the corresponding performance indicator thresholds;

[0041] If so, determining that the specified classification model passes the model evaluation;

[0042] If not, it is determined that the specified classification model fails the model evaluation.

[0043] Furthermore, after the step of determining whether the target comprehensive processing score is greater than a preset score threshold, the method further includes:

[0044] If the target comprehensive processing score is less than the score threshold, calling a preset backup processing strategy;

[0045] Determining a processing personnel corresponding to the backup processing strategy;

[0046] The task is assigned to the processing personnel.

[0047] In order to solve the above technical problems, the embodiment of the present application further provides a task allocation device based on artificial intelligence, which adopts the following technical solution:

[0048] A first acquisition module is used to acquire a target business field of a task when receiving a task to be assigned;

[0049] The first processing module is used to call the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filter out the target prediction results of each online agent corresponding to the target business field from the prediction results;

[0050] A first generating module, configured to generate a domain score for each online agent corresponding to the target business domain based on the target prediction result;

[0051] A screening module, used to screen out a first seat whose domain score is greater than a preset proficiency score threshold from all the online seats;

[0052] a second generating module, configured to generate a comprehensive processing score of the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain;

[0053] A judgment module, used to obtain a target comprehensive processing score with the largest value among the comprehensive processing scores, and to judge whether the target comprehensive processing score is greater than a preset score threshold;

[0054] The second processing module is configured to: if yes, select a target seat from the first seats based on the target comprehensive processing score, and assign the task to the target seat.

[0055] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0056] Upon receiving a task to be assigned, obtaining a target business domain of the task;

[0057] Calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filtering out the target prediction results of each online agent corresponding to the target business field from the prediction results;

[0058] Generating a domain score for each online agent corresponding to the target business domain based on the target prediction result;

[0059] Filter out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents;

[0060] generating a comprehensive processing score for the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain;

[0061] Obtaining a target comprehensive processing score with the largest value among the comprehensive processing scores, and determining whether the target comprehensive processing score is greater than a preset score threshold;

[0062] If yes, a target seat is selected from the first seats based on the target comprehensive processing score, and the task is assigned to the target seat.

[0063] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0064] Upon receiving a task to be assigned, obtaining a target business domain of the task;

[0065] Calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filtering out the target prediction results of each online agent corresponding to the target business field from the prediction results;

[0066] Generating a domain score for each online agent corresponding to the target business domain based on the target prediction result;

[0067] Filter out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents;

[0068] generating a comprehensive processing score for the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain;

[0069] Obtaining a target comprehensive processing score with the largest value among the comprehensive processing scores, and determining whether the target comprehensive processing score is greater than a preset score threshold;

[0070] If yes, a target seat is selected from the first seats based on the target comprehensive processing score, and the task is assigned to the target seat.

[0071] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0072] The present application first obtains the target business domain of the task upon receiving the task to be assigned; then calls the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and screens out the target prediction results corresponding to the target business domain of each online agent from the prediction results; then generates a domain score corresponding to the target business domain for each online agent based on the target prediction results; and screens out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents; subsequently, generates a comprehensive processing score for the first agent based on the task load data of the first agent and the target domain score of the first agent corresponding to the target business domain; finally, obtains the target comprehensive processing score with the largest value in the comprehensive processing scores, and determines whether the target comprehensive processing score is greater than the preset score threshold; if so, screens out the target agent from the first agent based on the target comprehensive processing score, and assigns the task to the target agent. When receiving a task to be assigned, the present application predicts and processes the historical performance data of multiple online agents by using a capability prediction model, and selects the target prediction results corresponding to the target business field of each online agent from the obtained prediction results, and generates the field score corresponding to the target business field of each online agent based on the target prediction results. Then, the first agent whose field score is greater than the proficiency score threshold is selected from all the online agents, and the comprehensive processing score of the first agent is generated based on the task load data of the first agent and the target field score of the first agent corresponding to the target business field. Subsequently, based on the target comprehensive processing score with the largest value in the obtained comprehensive processing score, the agent with the best processing capability corresponding to the task load data and the target field score corresponding to the target business field in the above-mentioned first agent is used as the agent for the final service processing of the above-mentioned task, which effectively realizes the selection of an agent that is proficient in the target business field and has the best processing capability that matches the target business field of the task to process the task, effectively improving the accuracy and intelligence of task allocation, which is conducive to improving the processing efficiency of tasks, and helps to improve customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0074] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0075] Figure 2 A flowchart of an embodiment of the artificial intelligence-based task allocation method according to the present application;

[0076] Figure 3 is a structural schematic diagram of an embodiment of an artificial intelligence-based task allocation device according to the present application;

[0077] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0078] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0079] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0080] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0081] like Figure 1As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0082] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0083] The terminal devices 101, 102, and 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0084] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0085] It should be noted that the artificial intelligence-based task allocation method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based task allocation device is generally set in the server / terminal device.

[0086] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0087] Continue to refer Figure 2, showing a flowchart of an embodiment of an artificial intelligence-based task allocation method according to the present application. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted. The artificial intelligence-based task allocation method provided in the embodiment of the present application can be applied to any scenario where task allocation is required, and the artificial intelligence-based task allocation method can be applied to products in these scenarios, for example, the allocation and processing of insurance business tasks in the financial insurance field. The artificial intelligence-based task allocation method comprises the following steps:

[0088] Step S201, when receiving a task to be assigned, obtaining a target business domain of the task.

[0089] In this embodiment, the task allocation method based on artificial intelligence is executed on the electronic device (e.g. Figure 1 The server / terminal device shown in the figure) can obtain tasks and target business areas of tasks through wired connection or wireless connection. It should be pointed out that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, Wi MAX connection, Zigbee connection, UWB (Ultra Wi DeBand) connection, and other wireless connection methods currently known or to be developed in the future. The present application can be applied to the task distribution scenario of the call center of an insurance company. The above-mentioned tasks to be assigned may be call tasks triggered when customers encounter problems in financial insurance scenarios and need consultation. In financial insurance scenarios, the above-mentioned call tasks may include insurance cost consultation tasks, insurance business operation consultation tasks, insurance claims tasks, and the like. Correspondingly, the business areas of tasks may include expense areas, business operation areas, claims areas, and the like.

[0090] Step S202 , calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filtering out the target prediction results of each online agent corresponding to the target business field from the prediction results.

[0091] In this embodiment, the online agent refers to a person who works in the call center of an insurance company and handles services, sales, data collection, information investigation, business return visits and other matters related to insurance business by answering calls or making outbound calls. The historical performance data may at least include data such as the resolution rate, response time, and customer satisfaction score of the online agent. The historical performance data of the online agent is input into a trained capability prediction model, so that the historical performance data of the online agent is predicted by the capability prediction model, and the prediction probability of the online agent's proficiency in each business field is output, that is, the above-mentioned prediction result. Among them, the above-mentioned target business field refers to the business field to which the above-mentioned task to be assigned belongs. After obtaining the prediction probability of the online agent's proficiency in each business field output by the capability prediction model, the target prediction probability of each online agent corresponding to the target business field can be further screened from the above-mentioned prediction results, that is, the above-mentioned target prediction result.

[0092] Step S203: generating a domain score for each online agent corresponding to the target business domain based on the target prediction result.

[0093] In this embodiment, the specific implementation process of generating the field score of each online agent corresponding to the target business field based on the target prediction result will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0094] Step S204: Filter out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents.

[0095] In this embodiment, according to business requirements, a score threshold is preset to classify the domain scores into different proficiency levels (such as proficient, average, not proficient, etc.). The proficiency score threshold is a score threshold corresponding to the proficiency level.

[0096] Step S205 : generating a comprehensive processing score of the first agent based on the task load data of the first agent and the target domain score of the first agent corresponding to the target business domain.

[0097] In this embodiment, the specific implementation process of generating the comprehensive processing score of the first agent based on the task load data of the first agent and the target field score of the first agent corresponding to the target business field will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0098] Step S206, obtaining the target comprehensive processing score with the largest value among the comprehensive processing scores, and determining whether the target comprehensive processing score is greater than a preset score threshold.

[0099] In this embodiment, the comprehensive processing scores of all the first seats can be numerically compared to screen out the target comprehensive processing score with the largest value. There is no specific limitation on the selection of the value of the above-mentioned score threshold, which can be set according to the actual business needs of the field of expertise assessment. If the comprehensive processing score of the agent is less than the score threshold, it indicates that the comprehensive processing score of the agent is low, for example, there is a situation where the agents are very busy or the agents are not good at the target business field.

[0100] Step S207: If yes, select a target seat from the first seats based on the target comprehensive processing score, and assign the task to the target seat.

[0101] In this embodiment, the specific implementation process of selecting the target seat from the first seat based on the target comprehensive processing score will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0102] When receiving a task to be assigned, the present application predicts and processes the historical performance data of multiple online agents by using a capability prediction model, and selects the target prediction results corresponding to the target business field of each online agent from the obtained prediction results, and generates the field score corresponding to the target business field of each online agent based on the target prediction results. Then, the first agent whose field score is greater than the proficiency score threshold is selected from all the online agents, and the comprehensive processing score of the first agent is generated based on the task load data of the first agent and the target field score of the first agent corresponding to the target business field. Subsequently, based on the target comprehensive processing score with the largest value in the obtained comprehensive processing score, the agent with the best processing capability corresponding to the task load data and the target field score corresponding to the target business field in the above-mentioned first agent is used as the agent for the final service processing of the above-mentioned task, which effectively realizes the selection of an agent that is proficient in the target business field and has the best processing capability that matches the target business field of the task to process the task, effectively improving the accuracy and intelligence of task allocation, which is conducive to improving the processing efficiency of tasks, and helps to improve customer satisfaction.

[0103] In some optional implementations, the step of selecting a target seat from the first seats based on the target comprehensive processing score in step S207 includes the following steps:

[0104] A second seat matching the target comprehensive processing score is selected from the first seats.

[0105] In this embodiment, the second seat refers to a seat whose comprehensive processing score is equal to the target comprehensive processing score.

[0106] Determine whether the number of the second seat is 1.

[0107] In this embodiment, the number of the second seats may be 1 or greater than 1.

[0108] If the number of the second seat is 1, the second seat is used as the target seat.

[0109] In this embodiment, if the number of the second seat is 1, the second seat is directly used as the target seat. By using the second seat with the highest comprehensive processing score, that is, the seat with the best processing capability corresponding to the task load data and the target field score corresponding to the target business field as the seat for the final service processing of the above task, it is effectively achieved to select the seat that is good at the target business field and has the best processing capability to process the task, which effectively improves the accuracy and intelligence of task allocation, is conducive to improving the processing efficiency of tasks, and helps to improve customer satisfaction.

[0110] If the number of second seats is greater than 1, the idle time of each second seat is obtained.

[0111] In this embodiment, the agent processing data of each online agent is collected and maintained in advance, and the agent processing data may at least include data such as the agent's task load, idle time, historical performance, etc. The agent processing data of the online agent can be obtained, and then the idle time of each second agent can be queried from the agent processing data.

[0112] Select the third seat with the longest idle time from all the second seats;

[0113] In this embodiment, the idle time of all the second seats may be compared to select the third seat with the longest idle time from all the second seats.

[0114] The third seat is used as the target seat.

[0115] In this embodiment, when there are multiple second seats that match the target comprehensive processing score, the third seat with the longest idle time is screened out from all the second seats. Since the third seat has the most abundant idle time, the third seat is subsequently used as the seat that finally serves to process the above task, thereby improving the intelligence of task allocation.

[0116] This application selects the second seat that matches the target comprehensive processing score from the first seats; then determines whether the number of the second seats is 1; if the number of the second seats is 1, the second seat is used as the target seat; and if the number of the second seats is greater than 1, the idle time of each second seat is obtained; then the third seat with the longest idle time is selected from all the second seats; and the third seat is subsequently used as the target seat. When the application detects that the number of the second seats is 1, the second seat with the highest comprehensive processing score, that is, the second seat with the best processing capability corresponding to the task load data and the target domain score corresponding to the target business domain, is directly used as the seat for the final service processing of the above-mentioned task, thereby effectively selecting the seat that matches the target business domain of the task, is good at the target business domain and has the best processing capability to process the task, effectively improving the accuracy and intelligence of task allocation, which is conducive to improving task processing efficiency and helping to improve customer satisfaction. In addition, when there are multiple second seats that match the target comprehensive processing score, the third seat with the longest idle time is selected from all the second seats. Since the third seat has the most abundant idle time, the third seat is subsequently used as the final service seat for processing the above task. While selecting an agent that matches the target business field of the task, is good at the target business field and has the best processing capability to process the task, the intelligence of task allocation is further improved.

[0117] In some optional implementations of this embodiment, step S205 includes the following steps:

[0118] The task load data of the first agent is obtained, and a first preset weight corresponding to the task load data is determined.

[0119] In this embodiment, the agent processing data of the online agent can be obtained, and then the task load data of the first agent can be queried from the agent processing data. The task load data may include the number of currently unprocessed tasks. There is no specific limitation on the selection of the value of the first preset weight, which can be set according to actual business needs.

[0120] A target domain score of the first seat corresponding to the target business domain is obtained, and a second preset weight corresponding to the target domain score is determined.

[0121] In this embodiment, the target domain score of the first seat corresponding to the target business domain can be screened out from the generated domain scores of each of the online seats corresponding to the target business domain. There is no specific limitation on the selection of the value of the second preset weight, which can be set according to actual business needs.

[0122] Get the preset score calculation formula.

[0123] In this embodiment, the score calculation formula specifically includes: S=a / X+b*Y, where S is the comprehensive processing score, X is the task load data, a is the first preset weight, Y is the target field score, and b is the second preset weight.

[0124] The task load data, the first preset weight, the target domain score and the second preset weight are calculated and processed based on the score calculation formula to obtain a corresponding calculation result.

[0125] In this embodiment, the task load data, the first preset weight, the target domain score and the second preset weight can be substituted into the corresponding positions in the above score calculation formula to perform calculations to obtain corresponding calculation results.

[0126] The calculation result is used as the comprehensive processing score of the first agent.

[0127] The present application obtains the task load data of the first seat and determines the first preset weight corresponding to the task load data; obtains the target field score of the first seat corresponding to the target business field and determines the second preset weight corresponding to the target field score; then obtains the preset score calculation formula; subsequently, based on the score calculation formula, the task load data, the first preset weight, the target field score and the second preset weight are calculated and processed to obtain the corresponding calculation result, and the calculation result is used as the comprehensive processing score of the first seat. The present application uses the first preset weight corresponding to the task load data and the second preset weight corresponding to the target field score to calculate and process the obtained task load data of the first seat and the target field score of the first seat corresponding to the target business field, and calls the score calculation formula to calculate and process the obtained task load data of the first seat. This can achieve fast and accurate calculation of the comprehensive processing score of the first seat, improve the calculation efficiency of the comprehensive processing score, and ensure the data accuracy of the obtained comprehensive processing score.

[0128] In some optional implementations, step S203 includes the following steps:

[0129] Get the specified prediction result corresponding to the specified seat.

[0130] In this embodiment, the predetermined seat is any one of all the online seats. The numerical form of the specified prediction result is a probability value.

[0131] Gets the preset conversion factor.

[0132] In this embodiment, the conversion factor is a pre-constructed factor for converting the prediction result output by the capability prediction model in the form of a probability value into a score. There is no specific limitation on the value of the conversion factor, which can be set according to actual business needs, for example, 100.

[0133] The product between the specified prediction result and the conversion factor is calculated.

[0134] In this embodiment, the designated prediction result is multiplied by the conversion factor to obtain a corresponding product.

[0135] The product is used as the designated field score of the designated seat corresponding to the target business field.

[0136] The present application obtains the designated prediction result corresponding to the designated seat; then obtains the preset conversion factor; then calculates the product between the designated prediction result and the conversion factor; and subsequently uses the product as the designated field score of the designated seat corresponding to the target business field. The present application obtains the preset conversion factor, and then calculates the product between the designated prediction result corresponding to the designated seat and the conversion factor, so as to quickly and accurately complete the conversion processing of the designated prediction result and generate the designated field score of the designated seat corresponding to the target business field, effectively improving the processing efficiency and processing intelligence of the conversion processing of the prediction result.

[0137] In some optional implementations, before step S202, the electronic device may further perform the following steps:

[0138] Get agent performance data within a preset time period.

[0139] In this embodiment, the time selection of the above-mentioned preset time period is not specifically limited and can be set according to actual business needs. For example, the time period can be selected as one year before the current time, or half a year before the current time, or three months before the current time, etc. The above-mentioned agent performance data may at least include identification information of the agent (such as work number), business field identification (such as business A, business B), and performance characteristics reflecting the agent in the business field (such as resolution rate, response time, customer satisfaction score, etc.), and the performance characteristics may also include professional knowledge, service capabilities, conversion rate and other data.

[0140] The agent performance data is processed by sample construction to obtain corresponding sample data.

[0141] In this embodiment, the above-mentioned sample construction process may include data cleaning, feature extraction and label generation processing. Among them, data cleaning includes: processing missing values, outliers, duplicate data and other processing. Feature extraction includes: based on business understanding and data analysis, selecting features that have a significant impact on predicting whether the agent is proficient in a certain business field. These features may include: resolution rate, response time, customer satisfaction score, knowledge mastery of specific products (assessed through tests or question-and-answer sessions) in a specific business field, quality of communication with customers (such as emotional stability, communication clarity), etc. Label generation: If the agent's proficiency is not directly given in the agent performance data, it is necessary to generate a label based on the performance indicator (such as good, average, not good, etc.).

[0142] The sample data is divided into a training set and a test set.

[0143] In this embodiment, the obtained sample data is divided into a training set and a test set according to a division ratio. There is no specific limitation on the numerical value selection of the division ratio, for example, a ratio of 70%:30% can be used.

[0144] The preset classification model is trained based on the training set to obtain the corresponding specified classification model.

[0145] In this embodiment, there is no specific limitation on the selection of the above-mentioned classification model. For example, any one of linear regression, logistic regression, decision tree, random forest, gradient boosting tree (such as XGBoost, LightGBM) or neural network can be used as the above-mentioned classification model. Among them, the process of training the preset classification model based on the training set includes: according to the selected classification model, configuring its parameters, such as learning rate, number of trees, depth of trees, etc. Then, the classification model is trained using the training set data, and the model parameters are adjusted through an optimization algorithm (such as gradient descent) to minimize the loss function, thereby obtaining a specified classification model with the function of predicting the performance data of the agent. And output the model function of predicting the probability of the agent's proficiency in various business fields. Among them, the above-mentioned optimization algorithm can specifically adopt the gradient descent algorithm, and the above-mentioned loss function can adopt mean square error, cross entropy, etc.

[0146] The specified classification model is evaluated based on the test set.

[0147] In this embodiment, the specific implementation process of the model evaluation of the specified classification model based on the test set will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated in detail here.

[0148] If the designated classification model passes the model evaluation, the designated classification model is used as the capability prediction model.

[0149] In this embodiment, if the specified classification model passes the model evaluation, it indicates that the model effect of the generated specified classification model has achieved the expected performance effect, and the specified classification model will be directly used as the capability prediction model. After the capability prediction model is constructed, the capability prediction model can be further stored to ensure the security of the capability prediction model. There is no specific limitation on the storage method of the above-mentioned capability prediction model, for example, blockchain storage, local database storage, cloud server storage and other storage methods can be used.

[0150] The present application obtains the agent performance data within a preset time period by performing sample construction processing on the agent performance data to obtain corresponding sample data; then the sample data is divided into a training set and a test set; subsequently, a preset classification model is trained based on the training set to obtain a corresponding designated classification model; further, a model evaluation is performed on the designated classification model based on the test set; if the designated classification model passes the model evaluation, the designated classification model is used as the capability prediction model. The present application obtains sample data by performing sample construction processing on the agent performance data within a preset time period, and then trains and tests the preset classification model based on the sample data, thereby realizing the rapid and accurate construction of a capability prediction model that meets the model construction requirements, improving the efficiency of constructing the capability prediction model, and ensuring the model effect of the obtained capability prediction model. It is beneficial to use the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents in the subsequent process to accurately output the corresponding prediction results.

[0151] In some optional implementations of this embodiment, the performing model evaluation on the specified classification model based on the test set includes the following steps:

[0152] The test set is input into the specified classification model for performance testing, and a performance indicator value of the specified classification model is obtained; wherein the performance indicator value is a numerical value corresponding to a preset performance indicator, and the number of the performance indicators includes multiple ones.

[0153] In this embodiment, the performance test is performed by inputting the test set into the specified classification model, and after the performance test of the specified classification model is completed, the performance indicator value corresponding to the preset performance indicator is recorded. Among them, the above-mentioned performance indicators specifically include accuracy, precision, recall rate, and F1 score. Correspondingly, the above-mentioned performance indicator values ​​include accuracy value, precision value, recall rate value, and F1 score value.

[0154] Determine whether all of the performance indicator values ​​are greater than the corresponding performance indicator thresholds.

[0155] In this embodiment, according to the actual model performance test requirements, a performance indicator threshold value corresponding to the normal performance expectation of the model is pre-constructed for each performance indicator. Specifically, an accuracy indicator threshold value matching the accuracy indicator, an accuracy indicator threshold value matching the precision indicator, a recall indicator threshold value matching the recall indicator, and an F1 score indicator threshold value matching the F1 score indicator are pre-constructed.

[0156] If so, it is determined that the specified classification model passes the model evaluation.

[0157] In this embodiment, if all the performance indicator values ​​are greater than the corresponding performance indicator thresholds, it indicates that the model effect (accuracy, precision, recall rate, F1 score) of the generated specified classification model has achieved the expected performance effect, and the specified classification model will be determined to have passed the model evaluation.

[0158] If not, it is determined that the specified classification model fails the model evaluation.

[0159] In this embodiment, if at least one performance indicator value among all the performance indicator values ​​is smaller than the corresponding performance indicator threshold, it is determined that the specified classification model fails the model evaluation.

[0160] The present application performs a performance test by inputting the test set into the specified classification model and obtains the performance index value of the specified classification model; then determines whether all the performance index values ​​are greater than the corresponding performance index threshold; if so, determines that the specified classification model has passed the model evaluation, otherwise determines that the specified classification model has not passed the model evaluation. The present application performs a model evaluation on the performance index value of the specified classification model corresponding to the performance index obtained based on the use of the performance index threshold, and then can accurately generate a model evaluation result corresponding to the specified classification model, thereby ensuring the accuracy of the obtained model evaluation result.

[0161] In some optional implementations of this embodiment, after step S206, the electronic device may further perform the following steps:

[0162] If the target comprehensive processing score is less than the score threshold, a preset backup processing strategy is called.

[0163] In this embodiment, if the target comprehensive processing score is less than the score threshold, that is, the comprehensive scores of all first seats are very low, for example, all first seats are very busy or there is no seat that is good at the target business field. Then, the backup processing strategy will be considered. The strategy content of the backup processing strategy includes allocating to a backup seat or notifying management personnel to intervene. Or the strategy content of the backup processing strategy may also include putting the task into a waiting queue.

[0164] A processing personnel corresponding to the backup processing strategy is determined.

[0165] In this embodiment, the backup processing strategy can be analyzed to find out the processing personnel included in the backup processing strategy, wherein the processing personnel may include backup seats or management actors.

[0166] The task is assigned to the processing personnel.

[0167] In this embodiment, the communication information of the processing personnel can be obtained, and then the task can be assigned to the processing personnel according to the communication information. The communication information can include an email address or a mobile phone number.

[0168] When the present application detects that the target comprehensive processing score is less than the score threshold, the preset backup processing strategy is called; then the processing personnel corresponding to the backup processing strategy are determined; and the task is subsequently assigned to the processing personnel. When the present application detects that the target comprehensive processing score is less than the score threshold, the processing personnel corresponding to the backup processing strategy are intelligently determined, and then the task is assigned to the processing personnel, so as to realize the intelligent distribution and processing of the task, which is conducive to ensuring the smooth execution of the task.

[0169] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0170] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned target comprehensive processing score, the above-mentioned target comprehensive processing score can also be stored in a node of a blockchain.

[0171] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0172] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0173] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0174] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0175] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0176] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a task allocation device based on artificial intelligence. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0177] like Figure 3 As shown, the artificial intelligence-based task allocation device 300 described in this embodiment includes: a first acquisition module 301, a first processing module 302, a first generation module 303, a screening module 304, a second generation module 305, a judgment module 306 and a second processing module 307. Among them:

[0178] A first acquisition module 301 is used to acquire a target business domain of a task when receiving a task to be assigned;

[0179] The first processing module 302 is used to call the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filter out the target prediction results corresponding to the target business field of each online agent from the prediction results;

[0180] A first generating module 303 is used to generate a domain score corresponding to the target business domain for each online seat based on the target prediction result;

[0181] A screening module 304 is used to screen out a first seat whose domain score is greater than a preset proficiency score threshold from all the online seats;

[0182] A second generating module 305, configured to generate a comprehensive processing score for the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain;

[0183] The judgment module 306 is used to obtain the target comprehensive processing score with the largest value among the comprehensive processing scores, and to judge whether the target comprehensive processing score is greater than a preset score threshold;

[0184] The second processing module 307 is configured to select a target seat from the first seats based on the target comprehensive processing score and assign the task to the target seat.

[0185] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based task allocation method in the aforementioned implementation mode, and will not be repeated here.

[0186] In some optional implementations of this embodiment, the second processing module 307 includes:

[0187] A first screening submodule, configured to screen out a second seat matching the target comprehensive processing score from the first seat;

[0188] A first judgment submodule, used to judge whether the number of the second seat is 1;

[0189] a first determining submodule, configured to use the second seat as the target seat if the number of the second seat is 1;

[0190] A first acquisition submodule, configured to acquire the idle time of each second seat if the number of the second seats is greater than 1;

[0191] A second screening submodule, used for screening out the third seat with the longest idle time from all the second seats;

[0192] The second determining submodule is configured to set the third seat as the target seat.

[0193] In some optional implementations of this embodiment, the second generation module 305 includes:

[0194] A second acquisition submodule, configured to acquire task load data of the first seat and determine a first preset weight corresponding to the task load data;

[0195] A third acquisition submodule is used to acquire a target domain score of the first seat corresponding to the target business domain, and determine a second preset weight corresponding to the target domain score;

[0196] The fourth acquisition submodule is used to obtain a preset score calculation formula;

[0197] A first calculation submodule, configured to calculate the task load data, the first preset weight, the target domain score, and the second preset weight based on the score calculation formula to obtain a corresponding calculation result;

[0198] The third determining submodule is configured to use the calculation result as the comprehensive processing score of the first agent.

[0199] In some optional implementations of this embodiment, the first generating module 303 includes:

[0200] A fifth acquisition submodule is used to acquire a designated prediction result corresponding to a designated seat; wherein the designated seat is any one of all the online seats;

[0201] A sixth acquisition submodule, used to acquire a preset conversion factor;

[0202] A second calculation submodule, used for calculating the product between the specified prediction result and the conversion factor;

[0203] The fourth determining submodule is configured to use the product as a designated field score of the designated seat corresponding to the target business field.

[0204] In some optional implementations of this embodiment, the task allocation device based on artificial intelligence further includes:

[0205] The second acquisition module is used to acquire the agent performance data within a preset time period;

[0206] A construction module, used to perform sample construction processing on the agent performance data to obtain corresponding sample data;

[0207] A partitioning module, used for partitioning the sample data into a training set and a test set;

[0208] A training module, used to train a preset classification model based on the training set to obtain a corresponding specified classification model;

[0209] An evaluation module, used for performing model evaluation on the specified classification model based on the test set;

[0210] The first determination module is used to use the designated classification model as the capability prediction model if the designated classification model passes the model evaluation.

[0211] In some optional implementations of this embodiment, the evaluation module includes:

[0212] A test submodule, used to input the test set into the specified classification model for performance testing, and obtain a performance index value of the specified classification model; wherein the performance index value is a numerical value corresponding to a preset performance index, and the number of the performance indexes includes a plurality;

[0213] A second judgment submodule is used to judge whether all the performance indicator values ​​are greater than the corresponding performance indicator threshold;

[0214] A first determination submodule, configured to determine that if yes, the specified classification model passes the model evaluation;

[0215] The second determination submodule is used to determine if the specified classification model fails the model evaluation.

[0216] In some optional implementations of this embodiment, the task allocation device based on artificial intelligence further includes:

[0217] A calling module, configured to call a preset backup processing strategy if the target comprehensive processing score is less than the score threshold;

[0218] A second determination module is used to determine a processing personnel corresponding to the backup processing strategy;

[0219] The allocation module is used to allocate the task to the processing personnel.

[0220] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0221] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital processors (DSPs), embedded devices, etc.

[0222] The computer device may be a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with the user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0223] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of a task allocation method based on artificial intelligence, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0224] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions stored in the memory 41 or process data, such as computer-readable instructions for running the artificial intelligence-based task allocation method.

[0225] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0226] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned artificial intelligence-based task allocation method.

[0227] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0228] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A task allocation method based on artificial intelligence, characterized in that: The steps include: Upon receiving a task to be assigned, obtaining a target business domain of the task; Calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filtering out the target prediction results of each online agent corresponding to the target business field from the prediction results; Generating a domain score for each online agent corresponding to the target business domain based on the target prediction result; Filter out the first agent whose domain score is greater than a preset proficiency score threshold from all the online agents; generating a comprehensive processing score for the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain; Obtaining a target comprehensive processing score with the largest value among the comprehensive processing scores, and determining whether the target comprehensive processing score is greater than a preset score threshold; If yes, a target seat is selected from the first seats based on the target comprehensive processing score, and the task is assigned to the target seat.

2. The task allocation method based on artificial intelligence according to claim 1 is characterized in that: The step of selecting a target seat from the first seat based on the target comprehensive processing score specifically includes: Selecting a second seat from the first seat that matches the target comprehensive processing score; Determine whether the number of the second seat is 1; If the number of the second seat is 1, the second seat is used as the target seat; If the number of second seats is greater than 1, obtaining the idle time of each of the second seats; Select the third seat with the longest idle time from all the second seats; The third seat is used as the target seat.

3. The task allocation method based on artificial intelligence according to claim 1 is characterized in that: The step of generating a comprehensive processing score for the first agent based on the task load data of the first agent and the target domain score of the first agent corresponding to the target business domain specifically includes: Acquire task load data of the first seat, and determine a first preset weight corresponding to the task load data; Obtaining a target domain score of the first seat corresponding to the target business domain, and determining a second preset weight corresponding to the target domain score; Get the preset score calculation formula; Based on the score calculation formula, the task load data, the first preset weight, the target domain score and the second preset weight are calculated to obtain a corresponding calculation result; The calculation result is used as the comprehensive processing score of the first seat.

4. The task allocation method based on artificial intelligence according to claim 1, characterized in that: The step of generating a domain score for each online agent corresponding to the target business domain based on the target prediction result specifically includes: Obtaining a specified prediction result corresponding to a specified seat; wherein the specified seat is any one of all the online seats; Get the preset conversion factor; Calculating the product between the specified prediction result and the conversion factor; The product is used as the designated field score of the designated seat corresponding to the target business field.

5. The task allocation method based on artificial intelligence according to claim 1 is characterized in that: Before the step of calling the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, the method further includes: Get agent performance data within a preset time period; Performing sample construction processing on the agent performance data to obtain corresponding sample data; Dividing the sample data into a training set and a test set; Training a preset classification model based on the training set to obtain a corresponding specified classification model; Performing model evaluation on the specified classification model based on the test set; If the designated classification model passes the model evaluation, the designated classification model is used as the capability prediction model.

6. The task allocation method based on artificial intelligence according to claim 5 is characterized in that: The step of performing model evaluation on the specified classification model based on the test set specifically includes: Input the test set into the specified classification model for performance testing, and obtain a performance indicator value of the specified classification model; wherein the performance indicator value is a numerical value corresponding to a preset performance indicator, and the number of the performance indicators includes multiple; Determine whether all the performance indicator values ​​are greater than the corresponding performance indicator thresholds; If so, determining that the specified classification model passes the model evaluation; If not, it is determined that the specified classification model fails the model evaluation.

7. The task allocation method based on artificial intelligence according to claim 1 is characterized in that: After the step of determining whether the target comprehensive processing score is greater than a preset score threshold, the method further includes: If the target comprehensive processing score is less than the score threshold, calling a preset backup processing strategy; Determining a processing personnel corresponding to the backup processing strategy; The task is assigned to the processing personnel.

8. A task allocation device based on artificial intelligence, characterized in that: include: A first acquisition module is used to acquire a target business field of a task when receiving a task to be assigned; The first processing module is used to call the trained capability prediction model to perform prediction processing on the historical performance data of multiple online agents to obtain prediction results, and filter out the target prediction results of each online agent corresponding to the target business field from the prediction results; A first generating module, configured to generate a domain score for each online agent corresponding to the target business domain based on the target prediction result; A screening module, used to screen out a first seat whose domain score is greater than a preset proficiency score threshold from all the online seats; a second generating module, configured to generate a comprehensive processing score of the first agent based on the task load data of the first agent and a target domain score of the first agent corresponding to the target business domain; A judgment module, used to obtain a target comprehensive processing score with the largest value among the comprehensive processing scores, and to judge whether the target comprehensive processing score is greater than a preset score threshold; The second processing module is configured to: if yes, select a target seat from the first seats based on the target comprehensive processing score, and assign the task to the target seat.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the artificial intelligence-based task allocation method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the artificial intelligence-based task allocation method according to any one of claims 1 to 7.

Images