Customer service allocation method and computer readable storage medium

Abstract

The invention discloses a customer service allocation method and a computer readable storage medium. The method comprises the steps of obtaining a question tag currently selected by a user; under the condition that customer service allocation conditions are met at present, candidate customer services matched with the user at the current moment are screened out from a customer service resource library according to the question labels; determining the matching degree between each candidate customer service and the user according to a matching degree evaluation method corresponding to the question label under the condition that the number of the candidate customer service is multiple; and determining a target customer service of the user according to the matching degree and the candidate customer service. According to the scheme, based on the question label of the user, a multi-level matching method is adopted, and the target customer service really meeting the user requirement can be intelligently, rapidly and accurately matched. The customer service resource utilization efficiency, the customer service accuracy and the service efficiency can be remarkably improved, and the user satisfaction can be further improved.

Description

Technical Field

[0001] This application belongs to the field of customer service, and particularly relates to a customer service allocation method and a computer-readable storage medium. Background Art

[0002] With the rapid development of Internet technology, online customer service systems have become an important channel for enterprises to communicate with users. Online customer service systems can provide users with real-time consultation services, help users solve problems, and improve the user experience. However, with the increase in the number of users and the diversification of consultation needs, how to efficiently allocate appropriate customer service personnel to users has become an urgent problem to be solved. Summary of the Invention

[0003] Embodiments of this application provide a customer service allocation method, a customer service allocation device, a customer service system, a terminal device, a computer-readable storage medium, and a computer program product, which can quickly and accurately match a target customer service that truly meets the user's needs based on the user's problem tags, and improve the quality of customer service.

[0004] The first aspect of the embodiments of this application provides a customer service allocation method, including: Obtain the problem tags currently selected by the user; When the current customer service allocation conditions are met, screen out candidate customer services that match the user at the current moment from the customer service resource library according to the problem tags; When the number of candidate customer services is multiple, determine the matching degrees between each candidate customer service and the user respectively according to the matching degree evaluation method corresponding to the problem tags; Determine the target customer service of the user according to the matching degree and the candidate customer services.

[0005] The second aspect of the embodiments of this application provides a customer service allocation device, including: An obtaining module, configured to obtain the problem tags currently selected by the user; A screening module, configured to screen out candidate customer services that match the user at the current moment from the customer service resource library according to the problem tags when the current customer service allocation conditions are met; A matching module, configured to determine the matching degrees between each candidate customer service and the user respectively according to the matching degree evaluation method corresponding to the problem tags when the number of candidate customer services is multiple; A determining module, configured to determine the target customer service of the user according to the matching degree and the candidate customer services.

[0006] A third aspect of the embodiments of the present application provides a customer service system, including: a service front end and a service back end, where the service front end is used to receive operation information of a user selecting a problem tag, and the service back end is used to implement the steps of the above customer service allocation method in response to the user's operation information, so that a target customer service provides services to the user.

[0007] A fourth aspect of the embodiments of the present application provides a terminal device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the above customer service allocation method when executing the computer program.

[0008] A fifth aspect of the embodiments of the present application provides a computer-readable storage medium storing a computer program, where the computer program implements the steps of the above customer service allocation method when executed by a processor.

[0009] A sixth aspect of the embodiments of the present application provides a computer program product, which enables a terminal device to implement the steps in the above customer service allocation method when the computer program product runs on the terminal device.

[0010] For the customer service allocation method provided in the first aspect of the embodiments of the present application, on the basis of obtaining the problem tags currently selected by the user, first, when it is determined that the current customer service allocation conditions are met, according to the problem tags, candidate customer services that match the user at the current moment are screened out from the customer service resource library. Then, when the number of candidate customer services is multiple, further according to the matching degree evaluation method corresponding to the problem tags, the matching degrees between each candidate customer service and the user are determined. Finally, according to the matching degree and the candidate customer services, the target customer service of the user is determined. This solution can adopt a multi-level matching method based on the user's problem tags to intelligently, quickly, and accurately match the target customer service that truly meets the user's needs. It can significantly improve the utilization efficiency of customer service resources, the accuracy and service efficiency of customer service, and can further improve user satisfaction.

[0011] It can be understood that the beneficial effects of the second to sixth aspects above can refer to the relevant descriptions in the first aspect above, and will not be repeated here. Description of the Drawings

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0013] Figure 1 It is a flowchart of the customer service allocation method provided by an embodiment of the present application; Figure 2 It is a schematic diagram of the overall business process of the customer service system provided by an embodiment of the present application; Figure 3a It is a partial process schematic diagram of the customer service allocation method provided by another embodiment of the present application; Figure 3b It is a partial process schematic diagram of the customer service allocation method provided by yet another embodiment of the present application; Figure 4 It is a structural schematic diagram of the customer service allocation device provided by an embodiment of the present application; Figure 5 It is a structural schematic diagram of the customer service system provided by an embodiment of the present application; Figure 6 It is a structural schematic diagram of the terminal device provided by an embodiment of the present application. Detailed implementation manners

[0014] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0015] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0016] It should also be understood that the term " / and" as used in the specification and the appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0017] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for differentiating descriptions and cannot be understood as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that specific features, structures, or characteristics described in connection with that embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but rather mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0019] As mentioned above, with the increase in the number of users and the diversification of consultation needs, how to efficiently allocate suitable customer service staff to users has become an urgent problem to be solved. In the prior art, the allocation method of human customer service usually uses a simple single-level allocation mechanism to allocate customer service to users, resulting in the allocated customer service not being able to truly meet the needs of users, and thus unable to solve the problems encountered by current users in a timely and efficient manner.

[0020] Through a large amount of research, it is found that the customer service allocation methods in the prior art have the following problems: 1) Lack of a precise matching mechanism based on problem tags: The existing methods often fail to make full use of the problem tag information selected by users and cannot accurately match user problems with the most professional customer service.

[0021] 2) The classification and grading of customer service are not detailed enough: The prior art has not established a perfect customer service grouping and grading mechanism and cannot make reasonable allocations according to the professional fields and role attributes of customer service.

[0022] 3) Insufficient utilization of historical service data: The existing methods have not effectively utilized the historical service data of customer service to evaluate the matching degree between customer service and specific problem types, resulting in the lack of pertinence of allocation results.

[0023] 4) Lack of a flexible matching degree evaluation method: The prior art usually adopts a single or fixed evaluation criterion and cannot dynamically adjust the evaluation weights according to the characteristics of different problem tags.

[0024] 5) The strategy for dealing with the situation of insufficient customer service resources is not perfect: When the most suitable customer service is unavailable, the existing methods lack effective alternative solutions and flexible queuing strategies.

[0025] To at least partially solve the above technical problems, an embodiment of the present application provides a customer service allocation method. The customer service allocation method provided in this embodiment can be applied to customer service systems of various types and industries, including but not limited to online customer service systems, telephone customer service systems, or other forms of customer service systems. Specifically, the customer service system may include a service front end and a service back end. Among them, the service front end can be used at least for interacting with users, that is, the service front end can include a user side. The service back end can be used to process business logics (such as routing allocation, work order management), data storage (such as session records, user portraits), and security control (such as identity authentication, data encryption, etc.). The customer service system also includes a server used by the customer service, that is, a customer service side. The customer service side can be classified as a service front end different from the user side or can be classified as a service back end.

[0026] In a specific example, the customer service system may include a user side, a customer service side, and a server. Among them, the server is respectively connected to each user side and each customer service side, and is used to process business logics, data storage, security control, execute the steps of the customer service allocation method provided in this embodiment, and implement message routing between the user side and the customer service side used by the target customer service, so as to achieve precise service of the target customer service to the user.

[0027] As Figure 1 shown, the customer service allocation method provided in the embodiment of the present application includes the following steps: Step S110, obtain the problem tags currently selected by the user.

[0028] Specifically, the customer service system in the embodiment of the present application includes a customer service resource library. Multiple pieces of information of customer services can be stored in the customer service resource library, including but not limited to basic information of the customer service, service area, professional skills, service type, online status, etc. Exemplarily, the internationalized customer service resource library may include customer services corresponding to different service areas.

[0029] In the embodiment of the present application, various suitable methods can be used to obtain the problem tags currently selected by the user. In one example, when the user accesses the customer service system through the user side application or web page, the system can provide a problem classification list, and the user can select one or more tags related to their problem from it. For example, the problem tags may include "account problem", "payment problem", "product consultation", "after-sales service", etc. In another example, the user can directly input the problem or search for problem keywords, and the system can also match the corresponding problem tags according to the keywords. The system can record the problem tags selected by the user as an important basis for subsequent customer service allocation, so that the target customer service assigned to the user can fully meet the user's needs.

[0030] Step S120, when the current customer service allocation conditions are met, according to the problem tags, screen out the candidate customer services that match the user at the current moment from the customer service resource library.

[0031] In the embodiments of the present application, various appropriate judgment methods can be used to judge whether the current customer service allocation conditions are met. In one implementation, the system can determine that the current customer service allocation conditions are met according to at least one of the following: The current customer service resource library includes a preset number of customer service resources; The user requests to connect to a human customer service; The intelligent customer service fails to solve the user's problem; The user is not in the blacklist; The current time is within the customer service time range, etc.

[0032] In the embodiments of the present application, when it is determined that the current customer service allocation conditions are met, the system can screen out the candidate customer services that match the user at the current moment from the customer service resource library according to the problem tags selected by the user. As mentioned before, since the customer service resource library can be a database storing all customer service information, including basic information, service area, professional skills, service type, online status, service experience, etc. of the customer service. Exemplarily but not restrictively, the system can initially screen out the customer services with the professional knowledge and skills required to solve the corresponding type of problems from the customer service resource library according to the type of the problem tags and the basic information of each customer service in the customer service resource library, as candidate customer services. For example, when the user selects the "account problem" tag, the system can screen out the online customer services that can handle account-related problems from the resource library as candidate customer services. Thus, the first-level matching between the customer service and the user can be achieved.

[0033] Step S130, when the number of candidate customer services is multiple, according to the matching degree evaluation method corresponding to the problem tags, determine the matching degree between each candidate customer service and the user respectively.

[0034] Step S140, determine the target customer service of the user according to the matching degree and the candidate customer services.

[0035] In the embodiments of the present application, when the number of the screened candidate customer services is multiple, the system can further evaluate the matching degree between each candidate customer service and the user, so as to select the most suitable customer service to provide services to the user. Specifically, the system can calculate the matching degree between each candidate customer service and the user according to the matching degree evaluation method corresponding to the problem tags.

[0036] In the embodiments of the present application, the matching degree evaluation method is an evaluation algorithm designed based on the characteristics and importance of problem tags. Different problem tags may correspond to different matching degree evaluation methods. For example, for problem tags with strong technicality, the matching degree evaluation pays more attention to the professional knowledge and problem-solving ability of the customer service; for problem tags related to after-sales service, the matching degree evaluation pays more attention to the patience, communication ability, response speed, etc. of the customer service. That is to say, the reference factors, the weights of the reference factors, and the evaluation algorithms of the matching degree evaluation methods corresponding to different problem tags can be different.

[0037] In the embodiments of the present application, after calculating the matching degree between each candidate customer service and the user, the system can screen out the target customer service of the user from the candidate customer services according to the level of the matching degree between each candidate customer service and the user. For example, the system can select the candidate customer service with the highest matching degree as the target customer service and assign the user's problem to this customer service for processing.

[0038] After the target customer service is determined, the system can establish a communication connection between the user and the target customer service, enabling the user to communicate with the target customer service in real time to solve the user's problem. At the same time, the system can also record the relevant information of this customer service assignment, including user information, problem tags, target customer service information, service time, etc., for subsequent service quality evaluation and performance appraisal of the customer service.

[0039] In a specific example, in the scenario of matching a target customer service for a user, the method of priority queue can be used to dynamically sort and dynamically allocate the candidate customer services.

[0040] First, a priority calculation formula can be predefined. Specifically, matching degree scoring functions corresponding to different problem tags can be designed. The key indicators in this scoring function can include the skill matching degree of the candidate customer service, the current load rate, the service score, the average response time, and the weight coefficients of each indicator. The skill matching degree can represent the matching degree between the customer's problem and the professional skills of the customer service. A preset skill matching evaluation method can be used to calculate the skill matching score between the candidate customer service and the current problem tag. The higher the skill matching score, the higher the priority. The current load rate can refer to the ratio of the current number of sessions of the customer service to the maximum number of sessions it can bear. The lower the load, the higher the priority. The service score can be the evaluation score given by historical users to the candidate customer service. The higher the score, the higher the priority. The average response time can be the average response time for the customer service to handle problems. The shorter the time, the higher the priority. The weight coefficients can be dynamically adjusted according to business needs (such as emphasizing response speed in peak seasons and service quality in off-seasons).

[0041] For example, the following formula can be used to calculate the priority score of the candidate customer service: Priority score = (skill matching score × α) + (1 - current load rate) × β + (service score × γ) + (1 / average response time) × δ. Wherein, α, β, γ, and δ all represent weight coefficients.

[0042] Exemplarily, the structure of the priority queue can adopt a Min-Heap structure, and the top element of the heap corresponds to the current optimal customer service (target customer service). And a dynamic update mechanism can be adopted to update the priority queue. Specifically, the update trigger conditions can include at least one of the following: change in customer service status (new session access / end); regular heartbeat detection (such as refreshing the load rate every 10 seconds); the customer service receives a new evaluation. Exemplarily, the update operations can include at least one of the following: removing the old record from the heap; recalculating the priority score; inserting a new record (the heap automatically adjusts the structure).

[0043] Exemplarily, the allocation strategy can at least include a competitive allocation strategy, and the candidate customer service with the highest priority score is allocated first. Exemplarily, the allocation strategy can also include: a downgrading strategy when the queue is empty (such as starting a backup customer service group), etc.

[0044] Exemplarily, when the scores of multiple candidate customer services are the same, the following method can be used to sort these customer services with the same score. Exemplarily, the sorting variables can be, in order of sorting priority: idle duration, service continuity, skill depth, etc. That is, the idle duration can be compared first, and the customer service that has not received an order for the longest time is preferentially allocated; secondly, the service continuity can be compared, and the customer service that is handling other sessions of the same customer is preferred; finally, the skill depth can be compared, and the customer service with more relevant certifications is preferred.

[0045] Using the above method of the priority queue to dynamically sort and dynamically allocate candidate customer services can significantly improve the allocation efficiency and allocation accuracy, and can also improve the service quality of customer services and the throughput of the system.

[0046] The customer service allocation method provided in the first aspect of the embodiments of the present application, on the basis of obtaining the problem tags currently selected by the user, first, when it is determined that the current customer service allocation conditions are met, according to the problem tags, candidate customer services that match the user at the current moment are screened out from the customer service resource library. Then, when the number of candidate customer services is multiple, further according to the matching degree evaluation method corresponding to the problem tags, the matching degree between each candidate customer service and the user is determined. Finally, according to the matching degree and the candidate customer services, the target customer service for the user is determined. This solution can intelligently, quickly, and accurately match the target customer service that truly meets the user's needs by adopting a multi-level matching method based on the user's problem tags. It can significantly improve the utilization efficiency of customer service resources, the accuracy and service efficiency of customer service, and further improve user satisfaction.

[0047] In one embodiment, step S120 screens out candidate customer service representatives that match the user at the current moment from the customer service resource library, including the following steps: Step S121, when the user currently meets the conditions for matching a customer service group, determine the customer service group that matches the user at the current moment according to the problem tags as the candidate customer service group to be selected; Step S122, screen out the customer service representatives that match the user at the current moment from the candidate customer service group to be selected as candidate customer service representatives.

[0048] In the embodiments of the present application, a customer service group may be a set of customer service representatives with similar professional skills, responsible for a similar business scope, or responsible for the same region. For example, each customer service representative in the customer service resource library may be divided into an account customer service group, a payment customer service group, a product customer service group, an after-sales customer service group, etc. according to the business type. Alternatively, they may also be divided into a general user customer service group, a VIP user customer service group, etc. according to the user level of service. Another example is that they may also be pre-divided into different regional groups according to the regions served by the customer service representatives. Exemplarily, the same customer service representative may be in different customer service groups. For example, a customer service representative with comprehensive business skills may be assigned to multiple business groups. Another example is that a customer service representative with multi-language skills may be assigned to different regional groups.

[0049] In the embodiments of the present application, various appropriate methods may be first used to determine whether the current user meets the conditions for matching a customer service group. For example, whether the user meets the conditions for matching a customer service group may include at least one of the following: the current user has completed identity verification; there is a corresponding customer service group in the region where the current user is located; the current user is not in the blacklist; the problem tags selected by the current user have a corresponding professional customer service group; the current user is a new user (no communication record between the current user and the customer service representative is found); the current user requests to change the customer service representative.

[0050] In the embodiments of the present application, when the system determines that the current user meets the conditions for matching a customer service group, various appropriate methods may be further used to determine the customer service group that matches the user at the current moment according to the problem tags selected by the user as the candidate customer service group to be selected. Optionally, the candidate customer service group to be selected may be one or more pre-divided customer service groups in the customer service resource library that can match the problem tags to a certain extent. Alternatively, the candidate customer service group to be selected may also be one or more customer service groups re-divided according to the problem tags.

[0051] After determining the candidate customer service groups, the system can further adopt a variety of appropriate screening logics to screen out the customer service that matches the user at the current moment from the candidate customer service groups as the candidate customer service. In one example, the screening conditions may include the online status of the customer service, the workload, the matching degree between the customer service skills and the problem tags, the role of the customer service in the group, etc. The screening process can be set according to the priorities corresponding to these screening conditions, and the customer service that matches the user at the current moment can be screened out from the candidate customer service groups according to the screening process as the candidate customer service. For example, the system can preferentially select the currently online customer service with a lighter workload as the candidate customer service.

[0052] In the above solution, when the user currently meets the conditions for matching the customer service group, first, according to the problem tags, the customer service group that matches the user at the current moment is initially determined; then, the customer service that matches the user at the current moment is further screened out from the candidate customer service groups as the candidate customer service. In this way, through the hierarchical progressive matching method, not only can the customer service that matches the user's problem be quickly and accurately screened out, improving the quality and efficiency of customer service, but also the calculation amount of screening and matching can be reduced, saving computing resources, thereby reducing the hardware requirements for scheme deployment and realizing the multi-scenario applicability of the scheme.

[0053] In one implementation manner, step S120 of screening out the candidate customer service that matches the user at the current moment from the customer service resource library further includes the following steps S1201 and / or step S1202: Step S1201, when the user is a new user, determine that the user currently meets the conditions for matching the customer service group; Step S1202, when the user is not a new user and the user does not meet the matching relationship with the user's first historical customer service, determine that the user currently meets the conditions for matching the customer service group, where the first historical customer service is the customer service who once answered the first question for the user, and the first question is a question associated with the current problem tag.

[0054] In the embodiments of the present application, a new user may be a user who has not communicated with the customer service in the current system. Such as a user who logs in to the customer service system for the first time. Exemplarily rather than restrictively, it can be determined whether the user is a new user according to the user's personal information such as ID, email, etc. and the service records stored in the system.

[0055] It can be understood that since new users usually do not have historical service records, it is necessary to re-match the target customer service for them. Therefore, in one example, if it is determined that the current user is a new user, the system can directly determine that the user currently meets the conditions for matching the customer service group, so as to execute the customer service matching process to match the customer service group for the new user. For example, when a newly registered user first uses the customer service system and selects the "account registration" problem tag, the system can directly determine that the user meets the conditions for matching the customer service group and match the corresponding customer service group according to the problem tag.

[0056] In another example, if it is determined that the current user is not a new user, it can be further determined whether the user and their first historical customer service (if any) meet the matching relationship. Specifically, the first historical customer service can be the customer service who has answered the first question for the user, and the first question is a question associated with the current question tag. That is to say, the first historical customer service is the customer service who has solved the same type of problem for the user. It can be understood that if there is a customer service who has solved the same type of problem for the user, then this customer service has the experience of communicating and handling the same type of problems with the current customer, and can answer the customer's questions more efficiently and accurately. Therefore, it can first be determined whether the user and the first historical customer service match at the current moment. Exemplarily, if it is determined that the user and the first historical customer service match at the current moment, then the first historical customer service can be directly used as the candidate customer service. If the user and the first historical customer service do not match at the current moment, then enter the matching process to match a new target customer service for the user.

[0057] In other words, for non-new users, the system can first determine whether the user has a first historical customer service. If so, it can be further determined whether the user and their first historical customer service meet the matching relationship. If it is determined that the user and the first historical customer service do not meet the matching relationship (for example, the first historical customer service is not online, has an overloaded workload, or is no longer responsible for this type of problem), the system can determine that the user currently meets the conditions for matching a customer service group, and then match a customer service group for the user.

[0058] Of course, if it is determined that an old user does not have a first historical customer service, that is, no customer service has answered the same type of question for the user before, it can also be determined that the user currently meets the conditions for matching a customer service group.

[0059] For example, when an old user selects the "bill query" question tag, in the case where it is determined that the user has not asked a bill query-related question before, or in the case where it is determined that the customer service who has handled the user's bill query problem is currently not online, busy, or has been transferred, the system can determine that the user meets the conditions for matching a customer service group, and then execute the process of matching a customer service group for the customer.

[0060] In the above solution, it can be flexibly determined whether the user meets the conditions for matching a customer service group according to whether the user is a new user and the matching relationship between the user and their first historical customer service, further improving the accuracy and efficiency of customer service allocation.

[0061] In one implementation manner, before executing step S1202, the customer service allocation method of this embodiment further includes the following steps: Step S1001, in the case where the user is not a new user, obtain the question tags that the user has selected before, and generate a historical tag set; Step S1002, when the first historical tag is the same as the problem tag currently selected by the user, determine the customer service who answered the question corresponding to the first historical tag for the user as the first historical customer service, where the first historical tag is any problem tag in the historical tag set; Step S1003, determine whether the first historical customer service is online and idle; Step S1004, when the first historical customer service is online and idle, determine that the user and the first historical customer service meet the matching relationship, and determine the first historical customer service as the candidate customer service matching the user at the current moment; Step S1005, when the first historical customer service is not online and idle, determine that the user and the first historical customer service do not meet the matching relationship.

[0062] In step S1001, for non-new users, the system can query the user's historical service records, obtain the problem tags previously selected by the user, and generate a historical tag set. The historical tag set can be a set of all problem tags selected by the user when using the customer service system in the past, reflecting the types of the user's historical consultation questions. For example, if a user has previously consulted questions such as "account security", "payment issues", "refund processing", etc., the system can form these problem tags into a historical tag set.

[0063] In step S1002, the system can compare the problem tag currently selected by the user with the problem tags in the historical tag set. If at least one problem tag (i.e., the first historical tag) that is the same as the current problem tag is found in the historical tag set, the customer service who previously answered the question corresponding to the first historical tag for the user can be determined as the first historical customer service. It can be understood that when the user has not consulted questions of the same type, the number of first historical customer services determined in this step can be 0; when the user has consulted questions of the same type, the number of first historical customer services determined in this step can be greater than or equal to 1.

[0064] For example, if the problem tag currently selected by the user is "account security", and the user's historical tag set also contains "account security", the system can determine the customer service who previously answered the account security question for the user as the first historical customer service.

[0065] In step S1003, the system can obtain the current status of the first historical customer service and determine whether it is online and idle. The online status can indicate that the customer service is currently logged in to the system and can accept assignments; the idle status indicates that the customer service currently has no user questions being processed or has a light workload and can accept new user assignments.

[0066] In one example, in step S1004, if it is determined that the first historical customer service is online and idle, the system can directly determine that the user and the first historical customer service meet the matching relationship, and determine the first historical customer service as the candidate customer service that matches the user at the current moment. In another example, in step S1004, if it is determined that the first historical customer service is online and idle, it can further be determined whether the difference between the time of the last conversation between the user and the first historical customer service and the current time is within a preset time. If it is within the preset time, it can be determined that the user and the first historical customer service meet the matching relationship, and the first historical customer service can be determined as the candidate customer service that matches the user at the current moment. Otherwise, it can be determined that the user and the first historical customer service do not meet the matching relationship. Thus, it proceeds to the step of matching a customer group for the user. The preset time can be set according to actual requirements. For example, the preset time is 10 minutes. If it is determined that the first historical customer service is online and idle, and the time of the last conversation between the user and the first historical customer service is no more than 10 minutes from the current time, it can be determined that the user and the first historical customer service meet the matching relationship. In this way, the continuity of service can be improved, allowing the customer service familiar with the user's historical problems to continue to provide services to the user, improving service efficiency and user satisfaction.

[0067] In step S1005, if it is determined that the first historical customer service is not online and idle (for example, the customer service is not online, or is dealing with other users' problems, or has an excessive workload, etc.), the system can determine that the user and the first historical customer service do not meet the matching relationship, and thus can proceed to the step of matching a customer group for the user.

[0068] The customer service allocation method provided in this embodiment can make full use of the user's historical service records, and preferentially allocate the user to the customer service that has answered the same type of questions for him / her. In this way, the continuity and consistency of service can be improved, and the user experience can be further enhanced. If there is no customer service that has answered the same type of questions for the user, or the customer service does not currently meet the conditions for serving the user, it can proceed to the step of matching a customer group for the user. In this way, a suitable customer service can be quickly and accurately allocated to the user in various scenarios.

[0069] In one implementation manner, step S121 determines the customer service group that matches the user at the current moment as the candidate customer service group, including the following steps: Step S1211, according to the user's basic information, screen out from the customer service resource library the customer service group responsible for the user group to which the user belongs at the current moment as the main customer service group of the user at the current moment; Step S1212, determine whether there is an online and idle customer service in the main customer service group; Step S1213, if so, determine the main customer service group as the candidate customer service group; Step S1214, if not, then select a secondary customer service group from the customer service resource library as the candidate customer service group, where the secondary customer service group is different from the primary customer service group.

[0070] In the embodiments of the present application, the basic information of the user can be any information reflecting the user's characteristics. The user characteristics may specifically include at least one of the following: user level, user location, product or service type used by the user, etc. In one example, the basic information of the user includes the user's location. Optionally, the user's IP address can be obtained, and the user's location can be determined based on the IP address. Alternatively, the port information (the website accessed by the user) of the user accessing the system can also be obtained, and the user's location can be determined based on the port information. The customer service resource library can be divided into customer service groups responsible for users in each region. For example, the customer service resource library can include a customer service group for Region A, a customer service group for Region B, and a customer service group for Region C. In step S1211, if the current user's location is Region A, the customer service group for Region A can be determined as the primary customer service group of the user at the current moment. In another example, the basic information of the user can include the user level, such as the general level and the VIP level. The user level can be determined according to the user ID and the corresponding relationship table between the user ID and the user level pre-stored in the system. The customer service resource library can include a general customer service group and a VIP customer service group. In step S1211, if the current user is a VIP customer, the VIP customer service group can be determined as the primary customer service group of the user at the current moment.

[0071] In step S1212, the system can obtain the working status of each customer service in the primary customer service group and determine whether there is a customer service currently online and idle. If there is a customer service in the primary customer service group currently online and idle, it means that the primary customer service group can provide services to the user; if there is no online and idle customer service in the primary customer service group, it means that the primary customer service group cannot provide services to the user currently, and an alternative customer service group needs to be found.

[0072] In step S1213, if it is determined that the primary customer service group includes an online and idle customer service, the system can determine the primary customer service group as the candidate customer service group. In this way, the customer service who best understands the characteristics and needs of the user group can provide services to the user, improving the pertinence and professionalism of the services.

[0073] In step S1214, if it is determined that there is no online and idle customer service in the primary customer service group, the system will select a secondary customer service group from the customer service resource library as the candidate customer service group. The secondary customer service group refers to the customer service group that can provide alternative services to the user when the primary customer service group is unavailable. In the embodiments of the present application, the secondary customer service group is different from the primary customer service group, but can have similar service capabilities or professional knowledge and can handle the problems corresponding to the problem tags selected by the user.

[0074] For example, if there is no online and idle customer service in the primary customer service group responsible for VIP user services, the system may select the senior customer service group in the general user service as the secondary customer service group. Another example is that if there is no online and idle customer service in the primary customer service group responsible for user services in a specific region, the system may select the customer service group for national general services as the secondary customer service group.

[0075] In the above solution, the customer service allocation method provided in this embodiment can, according to the basic information of the user, preferentially select the customer service group responsible for the user's group as the candidate customer service group, improving the pertinence and professionalism of the service. At the same time, when the primary customer service group is unavailable, the system can automatically select the secondary customer service group as a replacement, enabling users to obtain high-quality services in a timely manner and improving the availability and stability of the system.

[0076] In one implementation, step S122 screens out the customer service that matches the user at the current moment from the candidate customer service groups as the candidate customer service, specifically including the following steps: Step S1221, determine the primary customer service and secondary customer service of the user according to the correlation degree between the types of each customer service in the candidate customer service groups and the type of the problem label selected by the current user. Among them, the correlation degree between the type of the primary customer service and the type of the problem label selected by the current user is greater than or equal to the first correlation degree threshold, and the correlation degree between the type of the secondary customer service and the type of the problem label selected by the current user is greater than or equal to the second correlation degree threshold and less than the first correlation degree threshold; Step S1222, determine whether there is a primary customer service in the candidate customer service groups that is online and idle; Step S1223, if so, determine the candidate customer service according to the primary customer service that is online and idle; Step S1224, if not, determine the candidate customer service according to the secondary customer service that is online and idle.

[0077] In the embodiments of the present application, the types of customer service can be divided according to their professional skills, service fields they are good at, etc. For example, technical support customer service, after-sales service customer service, product consultation customer service, account security customer service, etc. The types of problem labels correspond to the types of customer service. For example, "account problem" corresponds to technical support customer service, and "product consultation" corresponds to product consultation customer service.

[0078] In step S1221, the system can obtain the type to which each customer service in the candidate customer service group belongs, compare it with the type of the problem label selected by the user, and use a preset correlation degree determination method (for example, a correlation degree comparison table between different types is pre-stored) to determine the correlation degree between the two. Furthermore, the primary customer service and secondary customer service of the user in the candidate customer service group can be determined according to the correlation degree. Specifically, the correlation degree between the type of the primary customer service and the type of the current problem label is greater than the correlation degree between the type of the secondary customer service and the type of the current problem label. That is to say, compared with the secondary customer service, the primary customer service may have better experience or higher ability in handling the current user's problem.

[0079] Both the first correlation degree threshold and the second correlation degree threshold can be set according to actual needs. For example, the first correlation degree threshold is equal to 100%, and the second correlation degree threshold is equal to 70%. For example, the correlation degree between the technical support type of customer service and the "technical support" type of problem is equal to 100%, the correlation degree between the after-sales service type of customer service and the "technical support" type of problem is 75%, and the correlation degree between the product consultation type of customer service and the "technical support" type of problem is 50%. If the problem label selected by the current user is "technical support", the technical support type of customer service in the candidate customer service can be determined as the user's primary customer service, and the after-sales service type of customer service in the candidate customer service can be determined as the user's secondary customer service.

[0080] It can be understood that if the correlation degree between the type to which the customer service belongs and the type of the problem label is higher, it means that the specialty of the customer service is more relevant to the problem to be solved by the user and is more suitable for handling this type of problem. Therefore, the primary customer service can be preferentially matched to the user. The secondary customer service also has the ability to handle the user's problem, so it can be used as the user's alternative customer service.

[0081] In a special example, when the number of customer service types is small, it can be directly determined whether the type to which the customer service in the candidate customer service group belongs is the same as the type of the problem label; if they are the same, the customer service is determined as the primary customer service of the current user; if they are different, the customer service is determined as the secondary customer service of the current user. For this case, the first correlation degree threshold is equal to 100%, and the second correlation degree threshold is equal to 0. For example, the customer service types only include pre-sales customer service and after-sales customer service. If the current problem label is "pre-sales consultation", the primary customer service of the current user is the pre-sales customer service, and the secondary customer service is the after-sales customer service. Another example is that if the problem label selected by the user is "account security", which belongs to the account type of problem, then the account type of customer service can be marked as the primary customer service, and the payment type of customer service or the product type of customer service can be marked as the secondary customer service.

[0082] In step S1222, the system can traverse the candidate customer service groups to determine whether there is a first-level customer service agent who is online and idle. If there is, it means that there is a customer service agent with a matching specialty who can provide services to the user; if not, it means that there is no currently available customer service agent with a matching specialty, and it is necessary to consider assigning a second-level customer service agent.

[0083] In step S1223, if it is determined that the candidate customer service groups include first-level customer service agents who are online and idle, the system can determine candidate customer service agents based on these first-level customer service agents. The advantage of doing this is to preferentially select customer service agents with a matching specialty to provide services to the user, improving the professionalism and quality of the service.

[0084] In step S1224, if it is determined that there are no first-level customer service agents who are online and idle in the candidate customer service groups, the system can determine candidate customer service agents based on the second-level customer service agents who are online and idle. Although the specialty areas of the second-level customer service agents do not exactly match the user's problem, they still have the basic ability to handle the problem and can provide services to the user.

[0085] The customer service assignment method provided in this embodiment can classify customer service agents into first-level customer service agents and second-level customer service agents according to the correlation degree between the specialty type of the customer service agent and the type of the problem label, and preferentially select first-level customer service agents with a matching specialty to provide services to the user, improving the professionalism and quality of the service. At the same time, when there are no available first-level customer service agents, the system will automatically select second-level customer service agents as a substitute, enabling the user to obtain services in a timely manner and improving the availability and stability of the system.

[0086] In one implementation manner, step S1223 determines candidate customer service agents based on the first-level customer service agents who are online and idle, and specifically includes the following steps: determining candidate customer service agents according to the role attributes of the first-level customer service agents who are online and idle in the candidate customer service groups. Among them, the role attributes include the main role and the backup role. In the case where there is a first-level customer service agent with the main role in the candidate customer service groups, the first-level customer service agent with the main role is determined as the candidate customer service agent; otherwise, the first-level customer service agent with the backup role is determined as the candidate customer service agent.

[0087] Specifically, the system can first obtain the role attributes of the first-level customer service agents who are online and idle in the candidate customer service groups. The role attribute refers to the responsibilities and positions of the customer service agents in the customer service group, including the main role and the backup role. The main role customer service agents can be the core members of the customer service group, with richer experience and higher professional skills, and are responsible for handling complex or important problems; the backup role customer service agents are usually the auxiliary members of the customer service group, with basic professional skills, and are responsible for handling routine or simple problems.

[0088] In one example, the customer service representatives in the primary role can be experts or senior personnel in a certain field, while the customer service representatives in the backup role can be support personnel, auxiliary personnel, or those with less experience. For example, the role attribute of a senior customer service specialist in the VIP customer service group is the primary role, while the role attribute of an intermediate customer service specialist can be the backup role.

[0089] In another example, the customer service representatives in the primary role can be those who often serve the customer group responsible for by this customer service group, and the customer service representatives in the backup role can be those who occasionally serve the customer group responsible for by this customer service group. Since a customer service representative can be assigned to different customer service groups, the role attribute of each customer service representative in different customer service groups can be different.

[0090] Exemplarily but not restrictively, if a customer service representative is assigned to multiple customer service groups, then the role attribute of this customer service representative is the primary role only in one of the customer service groups, and the role attributes in other customer service groups are all backup roles. For example, a certain customer service representative has language skills in both Country A and Country B. He mainly serves users in Country A and occasionally serves users in Country B. Then the role attribute of this customer service representative in the Country A customer service group is the primary role, and the role in the Country B customer service group is the backup role.

[0091] By considering the role attributes of customer service representatives, the system can further optimize the allocation of customer service representatives, enabling users to receive the most suitable service.

[0092] Specifically, if the candidate customer service group includes a first-level customer service representative in the primary role who is online and idle, the system can directly determine this first-level customer service representative in the primary role as the candidate customer service representative. This can give priority to selecting customer service representatives with rich experience and high skills to provide services to users, improving the quality and efficiency of services. If there is no first-level customer service representative in the primary role who is online and idle in the candidate customer service group, the system can determine a first-level customer service representative in the backup role as the candidate customer service representative. Although the experience and skills of customer service representatives in the backup role may not be as rich and high as those of customer service representatives in the primary role, they are still customer service representatives with the right expertise and can provide professional services to users.

[0093] The customer service allocation method provided in this embodiment can, according to the role attributes of first-level customer service representatives in the candidate customer service group, give priority to selecting first-level customer service representatives in the primary role to provide services to users, further improving the quality and efficiency of services. At the same time, when there is no available first-level customer service representative in the primary role, the system will automatically select a first-level customer service representative in the backup role as a substitute, enabling users to obtain professional and relevant services.

[0094] In one implementation manner, the customer service allocation method of this embodiment further includes the following steps: Step S1301: Obtain the historical service data sets of each candidate customer service. The historical service data sets include various service impact data, and each type of service impact data corresponds to a scoring variable. The scoring variable may include one or more of service satisfaction, response duration, service time, and service category.

[0095] Step S130: Determine the matching degree between each candidate customer service and the user according to the matching degree evaluation method corresponding to the problem label, including: Step S131: For each candidate customer service, determine the matching degree score between the customer service and the user according to the various service impact data of the customer service and the weight of each scoring variable corresponding to the current problem label.

[0096] Step S140: Determine the target customer service of the user according to the matching degree, which specifically includes the following steps: Step S141: Determine the priority of each candidate customer service according to the high and low of the matching degree score determined at the current moment, where the higher the matching degree score, the higher the priority; Step S142: Determine the candidate customer service with the highest priority at the current moment as the target customer service.

[0097] In the embodiment of the present application, the historical service data set can be obtained from the historical service records of the customer service. The historical service data may include the evaluation of the customer service by the user, the response speed of the customer service, the service duration, the types of problems processed, etc. Different scoring variables may have different importance for different problem labels. Therefore, the system can pre-allocate a weight related to the current problem label for each scoring variable. For example, for urgent problems, the weight of the response duration can be set relatively large; for complex problems, the weight of the service satisfaction can be set relatively large.

[0098] In step S1301, after determining the candidate customer service, the system can obtain the historical service data sets of each candidate customer service. The historical service data set is a data set recording the past service performance of the customer service, including various service impact data, such as user evaluation, response speed, service duration, types of problems processed, etc. Each type of service impact data corresponds to a scoring variable, which is used to quantify the performance of the customer service in a certain aspect.

[0099] Exemplarily, the scoring variable may include service satisfaction, response duration, service time, service category, etc.

[0100] Specifically, service satisfaction can refer to the satisfaction rating of users for customer service. For example, service satisfaction can be the chat satisfaction rating, which can be the score given by the customer to the current conversation after the session ends, measured (e.g., on a 1–5 star scale) and normalized to (actual score / 5) ∈ [0,1]. It can be understood that a high chat satisfaction score usually means that the customer has a high recognition of the solution and the interaction experience, and it is a direct indicator to measure the effect of customer service.

[0101] Response time can refer to the length of time it takes for the customer service to respond to the user's question. Specifically, it can be the time from when the customer initiates the session to the first response from the customer service, usually counted in seconds or minutes, and can be mapped to a normalized score through a threshold. A quick response can significantly improve the customer experience. Data shows that 95% of customers prefer timely responses, and are willing to wait for high-quality support even if the speed is slightly slower.

[0102] Service duration can refer to the total length of time the customer service provides services to the user. Specifically, it can be the complete duration of a single session from start to end, and can be mapped to [0,1] through a normalization formula (Min–Max normalization). Although being too long may affect efficiency, a moderate extension (e.g., 10–12 minutes) is often associated with high satisfaction, especially when the customer service provides detailed answers.

[0103] Service category can refer to the types of problems handled by the customer service. For example, each conversation can be labeled according to the topic or business type (such as product consultation, technical support, complaint handling), and the service performance under each label can be counted. Different classifications have different requirements for the customer service's capabilities. By classification, the assessment dimensions can be subdivided to achieve more accurate performance evaluation and allocation.

[0104] For example, the historical service dataset of candidate customer service A can include multiple data such as an average service satisfaction score of 4.8 (out of 5), an average response time of 30 seconds, an average service duration of 10 minutes, and the service category that the candidate is good at handling being account security issues, etc.

[0105] In step S131, for each candidate customer service, the system can calculate the matching score between the customer service and the user according to the historical service dataset of the candidate customer service. When calculating the matching score, the system can consider the weight of each rating variable corresponding to the current problem label. Different problem labels may have different degrees of emphasis on different rating variables, that is, different weights. Exemplarily, the system can also dynamically adjust the weights of each rating variable based on operational KPIs (such as conversion rate, satisfaction target).

[0106] For example, for the "Account Security" issue label, the system can set the weight of service satisfaction to 0.4, the weight of response time to 0.3, the weight of service time to 0.1, and the weight of service category to 0.2. For candidate customer service A, the system can calculate a matching score of 4.5 points (out of 5) based on its historical service dataset and the scoring variable weights.

[0107] Exemplarily, the matching score can be expressed using the following formula: ; where represents the matching score, represents service satisfaction, represents the first response time, represents the service duration, represents the service category, and w1, w2, w3, w4 represent weights.

[0108] In step S141, the system can determine the priority of each candidate customer service based on the calculated matching score. The higher the matching score, the higher the matching degree between the customer service and the user, and the higher the priority.

[0109] For example, if the matching score of candidate customer service A is 4.5 points, the matching score of candidate customer service B is 4.2 points, and the matching score of candidate customer service C is 4.8 points, then customer service C has the highest priority, followed by customer service A, and finally customer service B.

[0110] In step S142, the system can determine the target customer service for the user as the candidate customer service with the highest priority at the current moment. In the above example, the system can determine candidate customer service C as the target customer service and assign the user's problem to customer service C for handling.

[0111] The customer service allocation method provided in this embodiment can calculate the matching score between the customer service and the user based on the historical service data of the candidate customer service and the scoring variable weights corresponding to the problem label, select the customer service with the highest matching degree as the target customer service, further improve the accuracy of customer service allocation, and improve the quality of customer service. Moreover, it can also enable the solution to assign customer services with relatively high matching degrees to customers for different types of problems in various scenarios, improving the applicability of the solution.

[0112] In one implementation, step S140 determines the target customer service for the user according to the matching degree, and specifically includes the following steps: Step S143, when the first candidate customer service is included among the candidate customer services, determine the target customer service for the user according to the first candidate customer service, where the matching degree between the first candidate customer service and the user is greater than or equal to a preset matching degree threshold; Step S144, when the first candidate customer service is not included in each candidate customer service, transfer to the manual intervention process or the flexible queuing process.

[0113] In the embodiments of the present application, the matching degree threshold can be a standard preset by the system, which can be used to determine whether the matching degree between the customer service and the user has reached an acceptable level. The matching degree can be quantified by a matching degree score, and the matching degree threshold can be represented by a matching degree score threshold. An appropriate matching degree score threshold (hereinafter referred to as the score threshold) can be set according to actual needs.

[0114] Exemplarily, the system can calculate the matching degree scores between each candidate customer service and the user by using the method of step S141. It can first determine whether there is a candidate customer service with a matching degree score greater than or equal to the preset score threshold. If there is at least one candidate customer service with a matching degree score greater than or equal to the score threshold, the system can mark these candidate customer services as the first candidate customer services, and can determine the target customer service of the user according to the first candidate customer services. For example, if the number of the first candidate customer services is 1, the first candidate customer service can be directly used as the target customer service of the user. For another example, if there are multiple first candidate customer services, the first candidate customer service with the highest matching degree score can be selected as the target customer service of the user.

[0115] In a specific example, if the matching degree score threshold is set to 4.0 points (full score 5 points), the matching degree score of candidate customer service A is 4.5 points, the matching degree score of candidate customer service B is 3.8 points, and the matching degree score of candidate customer service C is 4.2 points, then customer service A and customer service C can be marked as the first candidate customer services, and the system can select customer service A with a higher matching degree from customer service A and customer service C as the target customer service.

[0116] In some special cases, for example, when the number of online customer services is small, the matching degree scores between the candidate customer services and the user may be relatively low, all less than the matching degree score threshold. That is, the first candidate customer service is not included in the candidate customer services. In this case, the system can trigger the manual intervention strategy to execute the manual allocation process or trigger the flexible queuing strategy to execute the flexible queuing process. For example, when the administrator is online and the customer is in urgent need of help, the manual allocation process can be triggered. And when the administrator is not online, the flexible queuing strategy can be triggered.

[0117] In the embodiments of the present application, the manual intervention strategy can refer to the strategy of manually allocating customer services for users by the customer service supervisor or the system administrator. That is, the customer service supervisor or the system administrator can, according to the actual situation and professional judgment, select the most suitable customer service among the currently online and idle customer services to provide services for the user.

[0118] In the embodiments of the present application, the elastic queuing strategy may refer to putting users into a queuing queue and waiting to be assigned when a suitable customer service agent (a candidate customer service agent with a matching score threshold greater than or equal to the score threshold) is available. Elastic queuing can dynamically adjust the position and priority of users in the queue according to factors such as the waiting time of users and the urgency of problems, ensuring that users can obtain services within a reasonable time.

[0119] For example, if the matching scores between all candidate customer service agents and the user are lower than 4.0 points, the system can trigger the manual intervention strategy, and the customer service supervisor can manually assign the most suitable customer service agent to the user according to the specific situation of the user and the characteristics of the problem. Or trigger the elastic queuing strategy to prompt the user to wait for a period of time. The user can be put into the queuing queue and wait to be assigned after a customer service agent with a higher matching degree becomes idle.

[0120] In the above solution, the customer service assignment method provided in this embodiment can, according to the matching threshold, ensure that the target customer service agent assigned to the user has a sufficiently high matching degree, improving the service quality and user satisfaction. At the same time, when there is no customer service agent with a sufficiently high matching degree available, the system can also trigger the manual intervention or elastic queuing strategy, enabling users to obtain high-quality services in special scenarios and improving user satisfaction.

[0121] Next, refer to Figure 2 the specific business process of the customer service system according to an embodiment of the present application shown in Figure 2 As shown, first, the user side can submit the problem to be solved through the front desk or APP of the customer service system. The customer service side can enter the problem into the system and classify the problem into a regular problem or a complex problem according to the type. If it is a regular problem, the system can display relevant links for the user to view through label classification. If the problem cannot be solved, the user can directly enter a consultation and initiate a request for artificial customer service. For complex problems, after classification by labels and detailed information, a request for artificial customer service can be directly initiated. The system can assign a target customer service agent to the user according to the customer service assignment method provided in the embodiments of the present application. The target customer service agent can solve the problem for the user through the customer service side. If it cannot be solved, the system can automatically enter the problem into the technical processing process. Specifically, the processing result can be obtained through methods such as replying by sales email, distributing for processing by various departments, and following up by sales review. Finally, the processing result can be informed to the user.

[0122] Next, refer to Figure 3a and Figure 3b the schematic diagrams of different partial processes of the customer service assignment method provided in another embodiment of the present application shown. Figure 3a It can be a schematic diagram of the basic assignment process of the customer service assignment method provided in another embodiment of the present application, Figure 3b It can be Figure 3aSchematic diagram of the "matching sub - process" in

[0123] As Figure 3a shown, first, the system can obtain the user's IP address or user identification (user ID). Then, it can be determined whether the user ID exists in the blacklist. If the user ID is in the blacklist, the process ends directly and exits the allocation. If the user ID is not in the blacklist, it is further determined whether the current allocation is a continuous allocation. Whether it is a continuous allocation or a non - continuous allocation, the process can transfer to the "matching sub - process" to match a target customer service for the user. If the match is successful, the process enters the "customer service reception" step, and the matched target customer service will receive the customer. If the match is unsuccessful, the process enters the "customer service message" step for subsequent processing.

[0124] As Figure 3b shown, the "matching sub - process" is specifically a process of matching a target customer service for the user. After the user accesses the system, it can be determined whether the problem type of the problem label currently selected by the user is the same as the type of the previous conversation, whether the previous conversation time is within 10 minutes, and whether the customer service received last time is online and idle. If these three conditions are met currently, the system can directly match the customer service (the first historical customer service) that received the current user last time to maintain the continuity of the service. If any of these three conditions is not met currently, the process can enter the next judgment process. First, it can be determined whether there is a customer service online in the user's main group (main customer service group). If there is a customer service online in the main group, the system can match the main group to handle the customer problem. If there is no customer service online in the main group, the system can match the secondary group (i.e., the secondary customer service group) to handle the customer problem.

[0125] After matching the main group, it can be further determined whether there is a main - type customer service (the user's primary customer service) online in the main group. If there is a main - type customer service online, it is further determined whether there is a main customer service (the primary customer service in the main role) online among the main - type customer services. If there is a main customer service online, the system can calculate the service quality scores of this batch of online main customer services and match the main customer service with the highest quality score to enable the customer to obtain the best - quality service. If there is no main customer service online, the system can match the secondary - type customer service (i.e., the user's secondary customer service). After matching the secondary - type customer service, the system can further determine whether there is a main customer service (the primary customer service in the main role) online among the secondary - type customer services. If there is, the system can calculate the service quality scores of this batch of online main customer services and match the main customer service with the highest quality score. If not, it can calculate the service quality scores of the secondary customer services online in the main - type customer services (the secondary customer services in the alternate role) and match the secondary customer service with the highest quality score.

[0126] Similarly, after matching the secondary group, the system can determine whether there is a primary type customer service (the user's first-level customer service) online in the secondary group. If there is a primary type customer service online, it further determines whether there is a primary customer service (the first-level customer service in the main role) online among the primary type customer services. If there is a primary customer service online, the system can calculate the service quality scores of this batch of online primary customer services and match the primary customer service with the highest quality score, so that the customer can obtain the best quality service. If there is no primary customer service online, the system can match the secondary type customer service (i.e., the user's second-level customer service). After matching the secondary type customer service, the system can further determine whether there is a primary customer service (the second-level customer service in the main role) online among the secondary type customer services. If there is, the system can calculate the service quality scores of this batch of online primary customer services and match the primary customer service with the highest quality score. If not, it can calculate the service quality scores of the secondary customer services (the second-level customer services in the backup role) online among the primary type customer services and match the secondary customer service with the highest quality score.

[0127] The above customer service allocation sub-process fully considers factors such as the continuity requirements of customers, the online status and idle situation of customer services, and the service quality of first-level customer services, and can provide customers with customer services that meet their needs as much as possible, providing customers with better quality and more efficient services.

[0128] The embodiment of the present application also provides a customer service allocation device. As Figure 4 shown, the customer service allocation device 400 includes: An acquisition module 410, configured to acquire the problem tags currently selected by the user; A screening module 420, configured to screen out candidate customer services that match the user at the current moment from the customer service resource library according to the problem tags when the current customer service allocation conditions are met; A matching module 430, configured to determine the matching degrees between each candidate customer service and the user respectively according to the matching degree evaluation method corresponding to the problem tags when the number of candidate customer services is multiple; A determination module 440, configured to determine the target customer service of the user according to the matching degree and the candidate customer services.

[0129] The embodiment of the present application also provides a customer service system. As Figure 5 shown, the customer service system 500 includes: a service front end 510 and a service back end 520, where the service front end 510 is used to receive the operation information of the user selecting problem tags, and the service back end 520 is used to implement the steps of the above customer service allocation method in response to the user's operation information, so that the target customer service provides services for the user.

[0130] Exemplarily, the main users of the service front-end 510 can be the customer groups to be served. The commonly used chat mode of the To B site can be adopted: self-service (intelligent customer service) + manual service (human customer service). For example, common problems such as policies and services can be solved through self-service and problem links. Complex problems such as function requests, service items, and high-intent purchase can be solved through self-service combined with human customer service. In a specific example, when accessing the human customer service, the user can select "product purchase" or "technical support" tags, so that the service back-end can preferentially match the customer service personnel corresponding to the tags according to different tags in the background. Exemplarily, the main users of the service back-end 520 can be the customer service groups of the system, and can include at least one of the following functional modules: Chat management module, used to manage multi-modal chat sessions in real time and support context association; Supervision and management module, used to monitor the customer service workflow and dialogue quality in real time, and assist the customer service to have a better dialogue when necessary; for example, the customer service supervisor can enter the ongoing session, view the real-time dialogue between the customer service and the customer, and can send messages in the dialogue box, and this message is only visible to the customer service. The customer service can better solve the customer's problems through the real-time messages sent by the customer service supervisor; File management module, used to store historical session data and support retrieval and viewing; Work order management module, used for the customer service to view and manage work order of sessions in various states; Team management module, used to configure the role permissions of the customer service team, optimize task allocation and efficiency analysis; Data dashboard module, used to visually analyze service data, generate dynamic reports and business insights; Setting center module, used to centrally configure the front-end tags and Q&A content without repeated development; Version management module, used to record system version changes, support audit tracking and hot update; Personal account management module, used to manage the online status and login / logout of personal accounts.

[0131] The embodiment of the present application also provides a terminal device. As Figure 6 shown, the terminal device 600 includes: at least one processor 610 ( Figure 6 only one processor is shown in the figure), a memory 620, and a computer program 630 stored in the memory 620 and executable on at least one processor 610. When the processor 610 executes the computer program 630, the steps of the above-mentioned customer service allocation method are implemented.

[0132] Figure 6The above are merely examples of terminal devices, which do not constitute a limitation on terminal devices. They may include more components than shown in the figures, or combine certain components, or have different components. The processor may be a central processing unit (CPU), or it may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0133] It should be noted that for the content such as information interaction and execution process between the above-mentioned devices / modules, since it is based on the same concept as the method embodiment of this application, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details will not be elaborated here.

[0134] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. Each functional module in the embodiment can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. In addition, the specific names of each functional module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working process of the modules in the above system can refer to the corresponding process in the foregoing method embodiment, and details will not be elaborated here.

[0135] The embodiment of this application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps in the above customer service allocation method can be implemented.

[0136] The embodiment of this application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above customer service allocation method.

[0137] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limiting them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A customer service allocation method, characterized in that, Including: Obtain the problem tags currently selected by the user; When the current customer service allocation conditions are met, according to the problem tags, screen out the candidate customer services that match the user at the current moment from the customer service resource library; When the number of the candidate customer services is multiple, according to the matching degree evaluation method corresponding to the problem tags, determine the matching degree between each candidate customer service and the user respectively; Determine the target customer service of the user according to the matching degree and the candidate customer services.

2. The customer service allocation method according to claim 1, wherein The screening out the candidate customer services that match the user at the current moment from the customer service resource library includes: When the user currently meets the conditions of the matching customer service group, according to the problem tags, determine the customer service group that matches the user at the current moment as the candidate customer service group; Screen out the customer services that match the user at the current moment from the candidate customer service group as the candidate customer services.

3. The customer service allocation method according to claim 2, wherein The screening out the candidate customer services that match the user at the current moment from the customer service resource library further includes: When the user is a new user, determine that the user currently meets the conditions of the matching customer service group; and / or When the user is not a new user and the user does not meet the matching relationship with the user's first historical customer service, determine that the user currently meets the conditions of the matching customer service group, where the first historical customer service is the customer service who answered the first question for the user, and the first question is a question associated with the current problem tags.

4. The customer service allocation method according to claim 3, characterized in that The method further includes: When the user is not a new user, obtain the problem tags previously selected by the user and generate a historical tag set; When the first historical tag is the same as the problem tag currently selected by the user, determine the customer service who answered the question corresponding to the first historical tag for the user as the first historical customer service, where the first historical tag is any problem tag in the historical tag set; Judge whether the first historical customer service is in an online and idle state; When the first historical customer service is in an online and idle state, determine that the user and the first historical customer service meet the matching relationship, and determine the first historical customer service as the candidate customer service that matches the user at the current moment; When the first historical customer service is not in an online and idle state, determine that the user and the first historical customer service do not meet the matching relationship.

5. The customer service allocation method according to claim 2, wherein The determining the customer service group that matches the user at the current moment as the candidate customer service group includes: According to the basic information of the user, screen out the customer service group responsible for the user group to which the user belongs at the current moment from the customer service resource library as the main customer service group of the user at the current moment; Judge whether there is an online and idle customer service in the main customer service group; If so, determine the main customer service group as the candidate customer service group; If not, screen out a secondary customer service group from the customer service resource library as the candidate customer service group, where the secondary customer service group is different from the main customer service group.

6. The customer service allocation method according to claim 2, wherein The screening out the customer services that match the user at the current moment from the candidate customer service group as the candidate customer services includes: Determine the primary customer service and secondary customer service of the user according to the correlation degree between the types of each customer service in the to-be-selected customer service group and the type of the problem label currently selected by the user, where the correlation degree between the type of the primary customer service and the type of the problem label currently selected by the user is greater than or equal to the first correlation degree threshold, and the correlation degree between the type of the secondary customer service and the type of the problem label currently selected by the user is greater than or equal to the second correlation degree threshold and less than the first correlation degree threshold; Judge whether there is a primary customer service in the to-be-selected customer service group who is online and idle; If so, determine the candidate customer service according to the primary customer service who is online and idle; If not, determine the candidate customer service according to the secondary customer service who is online and idle; 7. The customer service allocation method according to claim 6, wherein The determining the candidate customer service according to the primary customer service who is online and idle includes: Determine the candidate customer service according to the role attribute of the primary customer service who is online and idle in the to-be-selected customer service group, where the role attribute includes the main role and the backup role. In the case that there is a primary customer service with the main role in the to-be-selected customer service group, determine the primary customer service with the main role as the candidate customer service; otherwise, determine the primary customer service with the backup role as the candidate customer service.

8. The customer service allocation method according to any one of claims 1-7, characterized in that, The method further includes: Obtain the historical service data set of each candidate customer service, where the historical service data set includes various service impact data, and each service impact data corresponds to a scoring variable; The determining the matching degree between each candidate customer service and the user according to the matching degree evaluation method corresponding to the problem label includes: For each candidate customer service, determine the matching degree score between the customer service and the user according to the various service impact data of the customer service and the weight of each scoring variable corresponding to the current problem label; The determining the target customer service of the user according to the matching degree and the candidate customer service includes: Determine the priority of each candidate customer service according to the high and low of the matching degree score determined at the current moment, where the higher the matching degree score, the higher the priority; Determine the candidate customer service with the highest priority at the current moment as the target customer service.

9. The customer service allocation method according to any one of claims 1-7, characterized in that, The determining the target customer service of the user according to the matching degree and the candidate customer service includes: In the case that the first candidate customer service is included in each candidate customer service, determine the target customer service of the user according to the first candidate customer service, where the matching degree between the first candidate customer service and the user is greater than or equal to the preset matching degree threshold; The method further includes: In the case that the first candidate customer service is not included in each candidate customer service, transfer to the manual intervention process or the flexible queuing process.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the customer service allocation method according to any one of claims 1 to 9 are implemented.

Images