Customer service work order allocation method and device

Abstract

This application discloses a customer service work order allocation method and apparatus. The method includes: determining a first profile score for the target object and a second profile score for the customer service agent based on target object feature data and customer service agent feature data, wherein the customer service work order corresponds to the customer service request of the target object; determining a target score based on the first and second profile scores; statistically analyzing the target scores to obtain a score set for all customer service work orders corresponding to each customer service agent to be assigned; identifying target customer service work orders with scores higher than a set threshold from the score set, and assigning the target customer service work orders to the target customer service agent. This application solves the technical problem of low customer service work order processing efficiency caused by low accuracy in matching customer service work orders with customer service agents.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and more specifically, to a customer service work order allocation method and apparatus. Background Technology

[0002] When maintaining existing customer relationships, a common approach is to generate outbound service tickets, match service plans, and have customer service representatives make the calls. When multiple outbound service tickets are pending, it's crucial to allocate agents efficiently to complete the service quickly and accurately. The main technologies employed are as follows: one method calculates the "user-side resource conversion rate" based on user profiles and distributes or randomly matches outbound calls based on agent availability; another method concentrates tickets with high estimated success rates for allocation to high-performing agents; and a third method matches tickets with agent profiles and allocates them based on estimated completion probabilities. However, these methods fail to comprehensively consider user characteristics, service plan characteristics, and agent characteristics, potentially leading to poor communication due to inappropriate matching and underutilization of agent skills. Furthermore, they lack the ability to dynamically adjust ticket allocation, resulting in suboptimal overall efficiency.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a customer service work order allocation method and apparatus to at least solve the technical problem of low customer service work order processing efficiency caused by the low accuracy of matching customer service work orders with customer service agents.

[0005] According to one aspect of the embodiments of this application, a customer service work order allocation method is provided, comprising: determining a first profile score of the target object and a second profile score of the customer service agent based on target object feature data and customer service agent feature data, wherein the customer service work order corresponds to the customer service request of the target object; determining a target score based on the first profile score and the second profile score; statistically analyzing the target scores to obtain a score set of all customer service work orders corresponding to each customer service agent to be allocated; determining target customer service work orders with scores higher than a set threshold from the score set, and allocating the target customer service work orders to the target customer service agent.

[0006] Optionally, after determining the target score based on the first profile score and the second profile score, the method further includes: determining the first weight and the second weight of the target score within a preset time period; determining the first target score based on the first weight and the second weight, and statistically analyzing the first target score to obtain a first score set.

[0007] Optionally, determining the first weight of the target score within a preset time period includes: extracting the first temporal emotion feature from the customer service agent feature data; combining the first temporal emotion feature into a higher-order feature, the higher-order feature representing the change curve of the customer service agent's emotional capacity within the preset time period; and determining the higher-order feature as the first weight of the target score matrix.

[0008] Optionally, determining a second weight for the target score within a preset time period includes: determining the status characteristics of the previous customer service work order that the customer service agent to be assigned has completed, the status characteristics of the customer service work order indicating whether the customer service work order has been successfully processed; and determining the status characteristics of the previous customer service work order as the second weight for the target score.

[0009] Optionally, the target customer service work order with a score higher than a set threshold is determined from the score set, and the target customer service work order is assigned to the target customer service agent. This includes: if the target customer service agent has not yet processed the customer service work order, determining the scores of all customer service work orders corresponding to the target customer service agent from the score set, and determining the customer service work order with the highest score among all pending customer service work orders as the target customer service work order; if the target customer service agent has already started processing the customer service work order, determining the customer service work order with a score higher than the set threshold among all customer service work orders from the first score set as the target customer service work order.

[0010] Optionally, based on the target object feature data and customer service agent feature data, the first profile score of the target object and the second profile score of the customer service agent are determined respectively, including: acquiring the target object feature data and the customer service agent feature data; concatenating the target object feature data and the customer service agent feature data into a sparse long vector, and encoding the sparse long vector into a dense vector of the same length; inputting the dense vector into a preset model to obtain the first profile score of the target object and the second profile score of the customer service agent.

[0011] Optionally, after obtaining the target object feature data and customer service agent feature data, the method further includes cleaning the abnormal data in the target object feature data and customer service agent feature data.

[0012] According to another aspect of the embodiments of this application, a customer service work order allocation method is also provided, including: receiving a customer service request from a target object; determining a first profile score corresponding to the target object and obtaining a second profile score of all customer service agents in the customer service system; determining a target score based on the first profile score and the second profile score, and statistically analyzing the target scores to obtain a score set of all customer service agents to be allocated corresponding to the customer service request; determining a target customer service agent whose score is higher than a set threshold from the score set, and allocating the customer service request to the target customer service agent.

[0013] According to another aspect of the embodiments of this application, a customer service work order allocation device is also provided, comprising: a first determining module, configured to determine a first profile score of the target object and a second profile score of the customer service agent based on target object feature data and customer service agent feature data, wherein the customer service work order corresponds to a customer service request of the target object; a second determining module, configured to determine a target score based on the first profile score and the second profile score; a statistics module, configured to statistically analyze the target scores to obtain a score set of all customer service work orders corresponding to each customer service agent to be allocated; and an allocation module, configured to determine target customer service work orders with scores higher than a set threshold from the score set and allocate the target customer service work orders to the target customer service agent.

[0014] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored program, wherein, when the program is running, it controls the device where the non-volatile storage medium is located to execute the above-mentioned customer service work order allocation method.

[0015] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the processor is used to run a program, wherein the program executes the above-described customer service work order allocation method when it runs.

[0016] In this embodiment, a first profile score for the target object and a second profile score for the customer service agent are determined based on target object feature data and customer service agent feature data, respectively. A customer service work order corresponds to a customer service request from the target object. A target score is determined based on the first and second profile scores. The target scores are statistically analyzed to obtain a score set for all customer service work orders corresponding to each customer service agent to be assigned. Target customer service work orders with scores higher than a set threshold are identified from the score set and assigned to the target customer service agent. By acquiring target object feature data and customer service agent feature data, determining the first and second profile scores, and combining the first and second profile scores to obtain the target score, and determining the target work order matching the customer service agent based on the target score, the method achieves the goal of comprehensively considering the relationship between the target object and the customer service agent. This improves the accuracy of matching customer service work orders with customer service agents, thereby solving the technical problem of low customer service work order processing efficiency caused by low accuracy in matching customer service work orders with customer service agents. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a hardware structure block diagram of a computer terminal (or mobile device) for a customer service work order allocation method according to an embodiment of this application;

[0019] Figure 2 This is a schematic diagram of a customer service work order allocation method according to this application;

[0020] Figure 3 A schematic diagram of another optional customer service work order allocation method according to an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of an optional customer service work order allocation device according to an embodiment of this application. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] According to an embodiment of this application, an embodiment of a customer service work order allocation method is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] The methods and embodiments provided in this application can be executed on mobile terminals, computer terminals, cloud servers, or similar computing devices. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing a customer service ticket allocation method is shown. Figure 1As shown, the computer terminal 10 (or mobile device 10) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0026] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0027] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the customer service work order allocation method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned customer service work order allocation method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0028] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0029] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0030] According to an embodiment of this application, an embodiment of a customer service work order allocation method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] Figure 2 This is a flowchart of a customer service work order allocation method according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0032] Step S202: Based on the target object feature data and customer service agent feature data, determine the first profile score of the target object and the second profile score of the customer service agent respectively, and the customer service work order corresponds to the customer service request of the target object.

[0033] Step S204: Determine the target score based on the first portrait score and the second portrait score;

[0034] Step S206: Calculate the target scores to obtain the score set of all customer service work orders corresponding to each customer service representative to be assigned;

[0035] Step S208: Determine the target customer service work order with a score higher than the set threshold from the score set, and assign the target customer service work order to the target customer service agent.

[0036] Through the above steps, it is possible to obtain the target object feature data and customer service agent feature data, determine the first profile score and the second profile score, and obtain the target score by combining the first profile score and the second profile score. Based on the target score, the target work order matching the customer service agent can be determined. This achieves the goal of comprehensively considering the relationship between the target object and the customer service agent, thereby improving the technical effect of matching customer service work orders with customer service agents. This solves the technical problem of low customer service work order processing efficiency caused by the low accuracy of matching customer service work orders with customer service agents.

[0037] In step S202, the target object feature data includes at least: the target object's attribute features, the target object's historical consumption information, the target object's historical complaint information, and the target object's customer service telephone voice data. The target object's attribute features include, but are not limited to, the target object's age, gender, and occupation. The customer service agent feature data includes at least: the customer service agent's customer service dialogue voice data, the customer service agent's attribute features, the customer service agent's work performance information, and the customer service agent's historical order information. The customer service agent's attribute features include, but are not limited to, the customer service agent's age and gender.

[0038] In step S204, the first profile score can be a profile score matrix for the target object, where the elements of the matrix can be the target object's scores for different products, reflecting the target object's acceptance of different products; the second profile score can be a profile score matrix for customer service agents, where the elements of the matrix can be the customer service agents' scores for different products, reflecting the customer service agents' recommendation inertia for different products. Based on the first and second profile scores, the target score is determined by multiplying the target object's profile score matrix and the customer service agent's profile score matrix by transpose, resulting in the target score matrix. The elements of this matrix can be the scores of different customer service agents corresponding to each pending customer service work order.

[0039] Step S204 not only considers the one-sided understanding between the subject (target object and customer service agent) and product features, but also considers the cross-feature understanding between the target object and customer service agent by multiplying the matrix transpose. The weights of the neural network are updated using the stochastic gradient descent algorithm. The model parameters are selected and optimized by calculating the cross-entropy value. The first profile rating matrix corresponding to the target object and the second profile rating matrix corresponding to the customer service agent are output. The target rating matrix reflecting the relationship between the customer service agent and the target object is obtained by transposing and multiplying the matrix. It can be understood that the target rating matrix reflects the rating of each customer service work order to be processed corresponding to different customer service agents.

[0040] In step S208, the target customer service work orders with scores higher than the set threshold are selected from the score set as target customer service work orders. One option is to determine a target customer service agent, select all customer service work orders with scores higher than the set threshold from the score set as target customer service work orders for the target customer service agent, and assign the target customer service work orders to the target customer service agent.

[0041] The following detailed embodiments illustrate steps S202 to S208.

[0042] It should be noted that, in related technologies, the allocation scheme is no longer updated for assigned work orders. However, this application, for the assigned scheme, dynamically calculates the "service scheme resource conversion rate". When the status of each work order changes and there is at least one available agent, the assigned work order is reassigned. This maximizes the probability of resource conversion through dynamic allocation management by customer service, thereby achieving the goal of improving customer perception.

[0043] Specifically, after determining the target score in step S206, a first weight and a second weight for the target score within a preset time period are determined; a first target score is determined based on the first weight and the second weight, and the first target score is statistically analyzed to obtain a first score set. After determining the target score using the first profile score and the second profile score, the target score is dynamically updated, thereby maximizing the rationality of customer service work order allocation.

[0044] In some embodiments of this application, a method for determining the first weight is proposed as follows: extracting the first temporal emotion feature from the customer service agent feature data; combining the first temporal emotion feature into a higher-order feature, the higher-order feature representing the change curve of the customer service agent's emotional capacity within a preset time period; and determining the higher-order feature as the first weight of the target scoring matrix.

[0045] It should be noted that the emotional resilience curves include, but are not limited to, stress resilience curves, fatigue resilience curves, and negative energy resilience curves. The first weight value corresponding to the emotional resilience curve changes continuously over time, and the target score recalculated by incorporating the first weight value is also dynamically updated over time.

[0046] Furthermore, an optional method for determining the second weight is as follows: determine the status characteristics of the previous customer service work order that the customer service agent to be assigned has completed, the status characteristics of the customer service work order indicating whether the customer service work order has been successfully processed; determine the status characteristics of the previous customer service work order as the second weight of the target score.

[0047] In some embodiments of this application, the status characteristics of customer service tickets include, but are not limited to: processed (successful), processed (failed), and processed (rollback). In practical application scenarios, processed (successful) indicates that the customer service ticket has been successfully processed, for example: recommending a new product to a user, and the customer has accepted it. Processed (failed) indicates that the customer service ticket has been processed, but the user is not satisfied, for example: recommending a new product to a user, and the customer does not accept it. Processed (rollback) indicates, for example: recommending a new product to a user, but the user's willingness to accept it is weak or the user does not answer the call.

[0048] Since the status characteristics of customer service tickets also reflect whether the ticket can be successfully processed in a second communication, it is understandable that if the first processing of a customer service ticket fails, the success rate of processing the ticket again is also low. By adding the status characteristics of customer service tickets as the second weight in the calculation of the target score, the impact of the status characteristics of customer service tickets on the success rate of ticket processing is fully considered, making the allocation of customer service tickets more accurate and reasonable.

[0049] In step S206, the target customer service work order with a score higher than a set threshold is determined from the score set, and the target customer service work order is assigned to the target customer service agent. This includes: if the target customer service agent has not yet processed the customer service work order, determining the scores of all customer service work orders corresponding to the target customer service agent from the score set, and determining the customer service work order with the highest score from all pending customer service work orders as the target customer service work order; if the target customer service agent has already started processing the customer service work order, determining the customer service work order with a score higher than the set threshold from all customer service work orders in the first score set as the target customer service work order.

[0050] Specifically, before a customer service agent starts processing a customer service ticket, the impact of the first and second weights on the target score is temporarily disregarded because no tickets have been processed yet. The customer service ticket with the highest score corresponding to each agent is directly selected from the score set as the first customer service ticket matched for each agent. After the agent processes the first customer service ticket, the first and second weights are added to update the target score. For each agent, the target score can be updated when the status characteristic of the customer service ticket changes. For example, when the status characteristic of the customer service ticket changes from pending to processed (successful), the first and second weights at the corresponding moment are collected, the first target score for all agents to be assigned to the customer service ticket is calculated, and the original target score of the customer service ticket is replaced with the first target score to complete the update of the original target score.

[0051] Understandably, in this application, after recalculating the target score of the customer service ticket, the customer service ticket that has already been processed can be reassigned to one of the customer service agents to be assigned, instead of being fixed to be handled by that agent after being assigned to a customer service agent. This further improves the accuracy of customer service ticket assignment and thus improves the processing efficiency of customer service tickets.

[0052] In step S202, based on the target object feature data and customer service agent feature data, the first profile score of the target object and the second profile score of the customer service agent are determined respectively. First, the target object feature data and the customer service agent feature data are obtained; then, the target object feature data and the customer service agent feature data are concatenated into a sparse long vector, and the sparse long vector is encoded into a dense vector of the same length; finally, the dense vector is input into a preset model to obtain the first profile score of the target object and the second profile score of the customer service agent.

[0053] It should be noted that the preset model can be a recommendation algorithm model, such as the DeepFM model. This model considers the one-sided understanding between the subject and product features, the cross-feature understanding between subjects, and the selection of scenarios by mining the explicit and implicit interaction effects between users and products, customer service representatives and products, and users and customer service representatives. It uses a stochastic gradient descent algorithm to update the weights of the neural network, and uses a random sample set for each training iteration to ensure that each scenario and each user feature is sufficiently trained. Model parameters are selected and optimized by calculating the cross-entropy value. First, a decay factor is added to adaptive learning rate methods like Adam to accelerate model optimization efficiency. Second, cross-entropy is calculated to positively guide model performance improvement. After obtaining the target object feature data and customer service representative feature data, it is also necessary to clean the abnormal data in these two datasets.

[0054] Since the target object feature data and customer service agent feature data obtained include data, text, and voice, they need to be processed separately.

[0055] For linear data, discrete variables are encoded and transformed into sparse matrices as low-order input data. For continuous variables, maximum and minimum values ​​are calculated based on data characteristics and dimensions, and then standardized or normalized. For time-series variables, the average value and fluctuation range are calculated before processing as continuous variables, and the trend is calculated before processing as discrete variables. For the cross-information between the target audience and customer service agents, keywords in the text data are analyzed to serve as the primary basis for predicting user service solution categories. The frequency of effective words in the text data is statistically analyzed to mark user pain points. Time-series feedback content in the text data is given special attention, and time-series features are added based on the feedback timeframe and processing status. For the target audience's voice data, the user emotion recognition results corresponding to the dialogue voice data can reflect the characteristics of the user's expression. Users are tagged using voice data features, which serves as an important basis for matching subsequent service solutions. The tagged information also needs to be correlated with the user emotion recognition scenario and the agent's response; therefore, correlation tagging is required, service scenarios must be distinguished, and other variables in the service must be excluded. For customer service agent voice data, the emotional recognition results, silence duration, and tone of voice of the corresponding customer service representatives can be extracted and analyzed as features of service effectiveness to obtain higher-order features. The higher-order features can be expanded to include pressure bearing curve, fatigue bearing curve, negative energy bearing curve, etc., which can be used as dimensions and weights for customer service representative profiling and dynamic calculation.

[0056] This application also provides a customer service work order allocation method, such as... Figure 3 As shown, the process includes: step S302, receiving a customer service request from a target object; step S304, determining the first profile score corresponding to the target object and obtaining the second profile scores of all customer service agents in the customer service system; step S306, determining the target score based on the first profile score and the second profile score, and statistically analyzing the target scores to obtain a set of scores for all customer service agents to be assigned to the customer service request; and step S308, determining the target customer service agent whose score is higher than a set threshold from the set of scores and assigning the customer service request to the target customer service agent.

[0057] In practical applications, by acquiring target object feature data and customer service agent feature data, a first profile score and a second profile score are determined. The target score is obtained by combining the first profile score and the second profile score. Based on the target score, a target work order matching with a customer service agent is determined. This achieves the goal of comprehensively considering the relationship between the target object and the customer service agent, thereby improving the technical effect of matching customer service work orders with customer service agents. This solves the technical problem of low customer service work order processing efficiency caused by the low accuracy of matching customer service work orders with customer service agents.

[0058] The customer service work order allocation method provided in this application embodiment can also be applied to a customer service work order allocation device provided in this application embodiment, such as... Figure 4 As shown, it includes: a first determining module 40, used to determine a first profile score of the target object and a second profile score of the customer service agent based on the target object feature data and the customer service agent feature data, respectively, and the customer service work order corresponds to the customer service request of the target object; a second determining module 42, used to determine the target score based on the first profile score and the second profile score; a statistics module 44, used to statistically analyze the target scores to obtain a score set of all customer service work orders corresponding to each customer service agent to be assigned; and an allocation module 46, used to determine the target customer service work orders with scores higher than a set threshold from the score set, and allocate the target customer service work orders to the target customer service agents.

[0059] The second determining module 42 includes: an updating submodule, which is used to determine the first weight and the second weight of the target score within a preset time period; determine the first target score based on the first weight and the second weight, and statistically analyze the first target score to obtain a first score set.

[0060] The update submodule includes: a first weight unit, a second weight unit, and an allocation unit;

[0061] The first weighting unit is used to extract the first temporal emotion feature from the customer service agent feature data; the first temporal emotion feature is combined into a higher-order feature, which represents the change curve of the customer service agent's emotional capacity within a preset time period; the higher-order feature is determined as the first weight of the target scoring matrix.

[0062] The second weighting unit is used to determine the status characteristics of the previous customer service work order that the customer service agent to be assigned has completed. The status characteristics of the customer service work order indicate whether the customer service work order has been successfully processed; the status characteristics of the previous customer service work order are determined as the second weight of the target score.

[0063] The allocation unit is used to determine the scores of all customer service work orders corresponding to the target customer service agent from the score set when the target customer service agent has not yet processed the customer service work order, and to determine the customer service work order with the highest score from all pending customer service work orders as the target customer service work order; when the target customer service agent has started processing the customer service work order, it determines the customer service work order with a score higher than a set threshold from all customer service work orders in the first score set as the target customer service work order.

[0064] The first determining module 40 includes a training submodule, which is used to acquire target object feature data and customer service agent feature data; concatenate the target object feature data and customer service agent feature data into a sparse long vector, and encode the sparse long vector into a dense vector of the same length; input the dense vector into a preset model to obtain the first profile score of the target object and the second profile score of the customer service agent.

[0065] The training submodule also includes a cleaning unit, which is used to clean abnormal data in the target object feature data and customer service agent feature data.

[0066] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, including a stored program, wherein, when the program is running, it controls the device where the non-volatile storage medium is located to execute the above-described customer service work order allocation method.

[0067] According to another aspect of the embodiments of this application, a computer device is also provided, including a memory and a processor, wherein the processor is used to run a program, wherein the program executes the above-described customer service work order allocation method when it runs.

[0068] The aforementioned computer equipment executes the aforementioned customer service work order allocation method. By acquiring target object feature data and customer service agent feature data, it determines the first profile score and the second profile score, and combines the first profile score and the second profile score to obtain the target score. Based on the target score, it determines the target work order that matches the customer service agent. This achieves the goal of comprehensively considering the relationship between the target object and the customer service agent, thereby improving the technical effect of matching customer service work orders with customer service agents. This solves the technical problem of low customer service work order processing efficiency caused by the low accuracy of matching customer service work orders with customer service agents.

[0069] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0070] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0071] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and 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 displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0072] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0073] Furthermore, the functional units in the various embodiments of this application 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0074] If the integrated unit 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 application, in essence, or the part that contributes to the prior art, or all or 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 application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0075] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of customer care ticket assignment, the method comprising: The method comprises the following steps: Based on the target object feature data and the customer service representative feature data, a first portrait score of the target object and a second portrait score of the customer service representative are determined respectively, and the customer service order corresponds to a customer service request of the target object; According to the first portrait score and the second portrait score, a target score is determined; Statistics of the target score obtains a score set corresponding to all customer service orders of each to-be-assigned customer service; From the score set, a target customer service order with a score higher than a set threshold is determined, and the target customer service order is assigned to a target customer service representative; after determining the target score, a first weight and a second weight of the target score in a preset time period are determined; based on the first weight and the second weight, a first target score is determined, and a first score set is obtained by statistics of the first target score, wherein determining the first weight of the target score in the preset time period comprises: extracting a first time sequence emotion feature in the customer service representative feature data; the first time sequence emotion feature is combined into a high-order feature, and the high-order feature represents an emotion carrying capacity change curve of the customer service representative in the preset time period; the high-order feature is determined as the first weight of the target score matrix, and the second weight is determined by the following way: determining a state feature of a previous customer service order completed by the to-be-assigned customer service representative, and the state feature of the customer service order represents whether the customer service order is successfully processed; the state feature of the previous customer service order is determined as the second weight of the target score, the target object feature data comprises: attribute features of the target object, historical consumption information of the target object, historical complaint information of the target object and customer service telephone voice data of the target object, and the customer service representative feature data comprises: customer service dialogue voice data of the customer service representative, attribute features of the customer service representative, work performance information of the customer service representative and historical order information of the customer service representative.

2. The method of claim 1, wherein, From the score set, a target customer service order with a score higher than a set threshold is determined, and the target customer service order is assigned to a target customer service representative, comprising: In the case that the target customer service representative has not processed the customer service order, the scores of all customer service orders corresponding to the target customer service representative are determined from the score set, and the customer service order with the highest score among all to-be-assigned customer service orders is determined as the target customer service order; In the case that the target customer service representative has started processing the customer service order, the customer service order with a score higher than the set threshold among all customer service orders is determined as the target customer service order from the first score set.

3. The method of claim 1, wherein, Based on the target object feature data and the customer service representative feature data, a first portrait score of the target object and a second portrait score of the customer service representative are determined respectively, comprising: Obtaining the target object feature data and the customer service representative feature data; Splicing the target object feature data and the customer service representative feature data into a sparse long vector, and encoding the sparse long vector into a dense vector with the same length; Inputting the dense vector into a preset model to obtain the first portrait score of the target object and the second portrait score of the customer service representative.

4. The method of claim 3, wherein, After the target object feature data and the customer service agent feature data are acquired, the method further includes: Abnormal data in the target object feature data and the customer service agent feature data are cleaned.

5. A method of customer care ticket assignment, the method comprising: Comprise: Receiving a customer service request of a target object; Determine the first portrait score corresponding to the target object, and acquire the second portrait score of all customer service agents in the customer service system; According to the first portrait score and the second portrait score, determine the target score, and count the score set of all to-be-assigned customer service agents corresponding to the customer service request according to the target score; Determine the target customer service agent with a score higher than a set threshold from the score set, and assign the customer service request to the target customer service agent; after determining the target score, determine a first weight and a second weight of the target score in a preset time period; determine a first target score based on the first weight and the second weight, and count a first score set according to the first target score, wherein determining the first weight of the target score in the preset time period comprises: extracting a first time sequence emotion feature in the customer service agent feature data; combining the first time sequence emotion feature into a high-order feature, the high-order feature representing an emotion carrying capacity change curve of the customer service agent in the preset time period; determining the high-order feature as the first weight of the target score matrix, and determining the second weight by the following way: determining a state feature of a previous customer service order completed by the to-be-assigned customer service agent, the state feature of the customer service order indicating whether the customer service order is successfully processed; determining the state feature of the previous customer service order as the second weight of the target score, the customer service agent feature data comprising: customer service conversation voice data of the customer service agent, customer service agent attribute features, work performance information of the customer service agent, and historical order information of the customer service agent.

6. A customer care ticket assignment apparatus, characterized by, Comprise: The first determination module is used for determining the first portrait score of the target object and the second portrait score of the customer service agent based on the target object feature data and the customer service agent feature data, the customer service order corresponding to the customer service request of the target object; The second determination module is used for determining the target score according to the first portrait score and the second portrait score; The statistical module is used for counting the score set of all customer service orders corresponding to each to-be-assigned customer service agent according to the target score. The distribution module is configured to determine a target customer service order with a score higher than a set threshold from the score set and distribute the target customer service order to a target customer service agent; after determining the target score, determine a first weight and a second weight of the target score within a preset time period; determine a first target score based on the first weight and the second weight, and count a first score set obtained by the first target score, wherein the first weight of the target score within the preset time period is determined by extracting a first time sequence emotion feature in the customer service agent feature data; the first time sequence emotion feature is combined into a high-order feature, the high-order feature representing an emotion bearing capacity change curve of the customer service agent within the preset time period; the high-order feature is determined as the first weight of the target score matrix, and the second weight is determined by determining a state feature of a previous customer service order completed by the to-be-distributed customer service agent, the state feature of the customer service order indicating whether the customer service order is successfully processed; the state feature of the previous customer service order is determined as the second weight of the target score, the target object feature data includes attribute features of a target object, historical consumption information of the target object, historical complaint information of the target object and customer service telephone voice data of the target object, and the customer service agent feature data includes customer service conversation voice data of the customer service agent, attribute features of the customer service agent, work performance information of the customer service agent and historical order information of the customer service agent.

7. A non-volatile storage medium, characterized by The non-volatile storage medium includes a stored program, wherein when the program runs, the device in which the non-volatile storage medium is located executes the customer service order distribution method in any one of claims 1 to 5.

8. A computer device, comprising: The device includes a memory and a processor, and the processor is configured to run a program, wherein when the program runs, the device executes the customer service order distribution method in any one of claims 1 to 5.

Images