Customer service scheduling method and device, equipment and medium

Abstract

The invention relates to the technical field of business process optimization, and discloses a customer service scheduling method, which comprises the steps of receiving an incoming line request of a target customer, and performing customer group classification on the target customer according to customer information to obtain a customer group classification result; performing question classification on the consultation information to obtain question categories; determining a request priority according to the customer group classification result and the question category; detecting whether the consultation information is a historical consultation question, and if the consultation information is the historical consultation question, accessing an intelligent customer service channel; determining an intelligent customer service strategy according to the question category, the request priority and the customer group classification result; and screening out a target intelligent customer service from all the intelligent customer services according to the intelligent customer service strategy, and switching to a service mode corresponding to the question category for service. According to the invention, through customer group shunting, refined service is realized. Through the question category, the request priority and the customer group classification result, the long waiting time of the customer is avoided, the customer service efficiency is improved, and the labor cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of business process optimization, and particularly to a customer service scheduling method, device, equipment and medium. Background Art

[0002] At present, the scheduling of customer queuing and customer service resources generally adopts a fixed rule mode. The services that customers are going to handle are divided into multiple major service categories, and then further divided into multiple minor service categories. Finally, multiple queuing queues are formed according to the minor service categories. A fixed number of customer service resources are arranged in each queue, and a first-in-first-out queuing mechanism is adopted to process the service handling requests of customers one by one.

[0003] However, the existing customer queuing strategy and customer service resource scheduling strategy queue in the queuing queue according to service classification. The customer service resources in each queuing queue only handle the customers in the current queuing queue, resulting in the problem that high-quality customers wait for a long time, which affects the customer experience. Summary of the Invention

[0004] Embodiments of the present invention provide a customer service scheduling method, device, equipment and medium to solve the problem that customers wait for a long time in the prior art, which affects the customer experience.

[0005] A customer service scheduling method includes: Receiving an incoming request from a target customer, where the incoming request includes consultation information and customer information; Classifying the target customer into customer groups according to the customer information to obtain a customer group classification result; Classifying the consultation information to obtain a problem category; Determining a request priority corresponding to the incoming request according to the customer group classification result and the problem category; Detecting whether the consultation information is a historical consultation problem. If the consultation information is a historical consultation problem, then connecting the incoming request to the intelligent customer service channel; the historical consultation problem refers to all problems that the customer has consulted in a preset historical period before the current time point; Determining an intelligent customer service strategy according to the problem category, the request priority and the customer group classification result; Selecting a target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, and switching the target intelligent customer service to a service mode corresponding to the problem category to serve the target customer through the service mode.

[0006] A customer service scheduling device includes: A request receiving module, configured to receive an incoming request from a target customer, where the incoming request includes consultation information and customer information; A customer group classification module, configured to classify the target customers according to the customer information to obtain a customer group classification result; A question classification module, configured to classify the consultation information to obtain a question category; A request priority module, configured to determine a request priority corresponding to the incoming request according to the customer group classification result and the question category; An intelligent customer service channel module, configured to detect whether the consultation information is a historical consultation question. If the consultation information is a historical consultation question, the incoming request is connected to the intelligent customer service channel; the historical consultation question refers to all questions that the customer has consulted during a preset historical period before the current time point; A customer service strategy module, configured to determine an intelligent customer service strategy according to the question category, the request priority, and the customer group classification result; A customer service screening module, configured to screen out a target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, and switch the target intelligent customer service to a service mode corresponding to the question category to serve the target customer through the service mode.

[0007] A computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor is configured to execute the above-mentioned customer service scheduling method.

[0008] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned customer service scheduling method is implemented.

[0009] In the above-mentioned customer service scheduling method, device, equipment, and medium, in the customer service scheduling method of the present invention, through the consultation information and customer information included in the incoming request, the determination of the customer group classification result is realized, and the classification of the question category is realized. Furthermore, the determination of the request priority is realized, avoiding the problem of long waiting time for high-quality customers. By detecting whether the consultation information is a historical consultation question, the determination of the intelligent customer service channel is realized, reducing the labor cost, and further realizing the determination of the intelligent customer service strategy. Through the intelligent customer service strategy, the screening of the target intelligent customer service and the switching of the service mode are realized, thereby realizing the diversion of customers, improving the efficiency of customer service, and further enhancing customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0011] Figure 1 is a flowchart of a customer service scheduling method in an embodiment of the present invention; Figure 2 is a flowchart of step S20 of the customer service scheduling method in an embodiment of the present invention; Figure 3 is a flowchart of the customer service scheduling method in the first embodiment of the present invention; Figure 4 is a flowchart of step S803 of the customer service scheduling method in an embodiment of the present invention; Figure 5 is a flowchart of the customer service scheduling method in the second embodiment of the present invention; Figure 6 is a flowchart of the customer service scheduling method in the third embodiment of the present invention; Figure 7 is a flowchart of the customer service scheduling method in the fourth embodiment of the present invention; Figure 8 is a schematic block diagram of a customer service scheduling device in an embodiment of the present invention; Figure 9 is a schematic diagram of a computer device in an embodiment of the present invention. Detailed implementation manners

[0012] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the protection scope of the present invention.

[0013] In one embodiment, as Figure 1 shown, a customer service scheduling method is provided, including the following steps: S10: Receive an incoming request from a target customer, where the incoming request includes consultation information and customer information.

[0014] Understandably, the incoming call request can be a call request initiated by a customer to a service company based on a certain need. The call request can be a landline call, an Internet call, or the call request of the customer can be processed by a telephone platform. The consultation information can be voice, text, or picture. The customer information is the information used to characterize the customer's identity, for example, mobile phone number, platform account, etc. The target customer refers to the customer who initiates the incoming call request. Specifically, receive the incoming call request initiated by the target customer, and the incoming call request includes consultation information and customer information.

[0015] S20: Classify the target customer into customer groups according to the customer information to obtain a customer group classification result.

[0016] Understandably, the customer group classification result is used to characterize the customer group to which the target customer belongs. For example, the excellent customer group, the high-quality customer group, the growing customer group, the general customer group, and the new customer group.

[0017] Specifically, analyze the customer information to detect whether the customer information contains a customer group label. If the customer information contains a customer group label, then determine the target customer as the customer group classification result corresponding to the customer group label. If the customer information does not contain a customer group label, obtain a feature extraction rule, extract information features from the customer information through the feature extraction rule, and perform customer group classification on all feature information through a preset classification customer group label set, that is, determine the customer group label of each feature information, and then obtain the customer group classification result corresponding to the target customer through the customer group label.

[0018] S30: Classify the consultation information to obtain a problem category.

[0019] Understandably, the problem category is used to characterize the type of consultation information, for example, the assistance category, the consultation category, the complaint category, etc.

[0020] Specifically, when the consultation information is audio, identify the audio to obtain the corresponding recognized text. Then, perform intent recognition on the recognized text through an intent recognition model to obtain the customer's consultation question. Then, classify the customer's consultation question, that is, obtain a preset problem category, calculate the similarity between the customer's consultation question and each preset problem category, and then obtain the problem category. Or, obtain a problem classification model, and classify the customer's consultation question through the problem classification model to obtain the problem category. Among them, when the consultation information is text, directly perform recognition through the intent recognition model and then perform classification through the problem classification model.

[0021] When the consultation information is a picture, the picture recognition technology is used to recognize the consultation information to obtain the recognized text. Then, the intention recognition model is used to recognize the intention of the recognized text, and the customer consultation question can be obtained. Then, the customer consultation question is classified, that is, the preset question categories are obtained, and the similarity between the customer consultation question and each preset question category is calculated to obtain the question category. Alternatively, a question classification model is obtained, and the customer consultation question is classified by the question classification model to obtain the question category.

[0022] S40: According to the customer group classification result and the question category, determine the request priority corresponding to the incoming request.

[0023] Understandably, the request priority is used to represent the priority level of the service order of the target customer.

[0024] Specifically, obtain the priority evaluation rule, and use the priority evaluation rule to evaluate the priority of the customer group classification result and the question category. That is, first judge whether the question category is a complaint type. If so, determine the incoming request as the first priority. If not, then judge whether the customer group classification result is a preset customer group. If so, determine the incoming request as the priority level corresponding to the preset customer group. If not, obtain the consultation time corresponding to the question category to sort according to the consultation time, so as to obtain the request priority corresponding to the incoming request.

[0025] S50: Detect whether the consultation information is a historical consultation question. If the consultation information is a historical consultation question, connect the incoming request to the intelligent customer service channel; the historical consultation question refers to all questions that the customer has consulted in a preset historical period before the current time point.

[0026] Understandably, the historical consultation question refers to all questions asked by other customers before, that is, all questions that the customer has consulted in a preset historical period before the current time point. The preset historical period can be one month or one year, etc. The intelligent customer service channel refers to the connection channel replied by the intelligent robot.

[0027] Specifically, detect whether the consultation information is a historical consultation question, that is, calculate the similarity between the consultation information and all historical consultation questions, screen out the maximum value from all similarity results, and compare the maximum value with the preset threshold. If the maximum value is greater than the preset threshold, determine the consultation information as a historical consultation question and connect the incoming request to the intelligent customer service channel.

[0028] S60: Determine the intelligent customer service strategy according to the question category, the request priority, and the customer group classification result.

[0029] Understandably, the intelligent customer service strategy refers to the formulated service method for intelligent customer service.

[0030] Specifically, in the intelligent customer service channel, determine the channel corresponding to the customer group classification result, then screen out all preset service strategies through the request priority, and then screen out the intelligent customer service strategy from all preset service strategies according to the problem category, that is, match the problem category with the preset categories corresponding to all preset service strategies, so as to screen out the intelligent customer service strategy corresponding to the problem category.

[0031] S70: Screen out the target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, and switch the target intelligent customer service to the service mode corresponding to the problem category, so as to serve the target customer through the service mode.

[0032] Understandably, the target intelligent customer service refers to the intelligent robot used to reply to customer questions. The service mode refers to the service process associated with different problem types, and each problem type is associated with a service mode. That is, in this embodiment, different service modes will be set for different problem categories. For example, if the problem category is the assistance category, the associated service mode is the robot service mode; if the problem category is the consultation category, the associated service mode is the robot Q&A mode; if the problem category is the complaint category, the associated service mode is the robot coordination mode; in this way, different service modes can be used to serve different problem types better.

[0033] Specifically, screen out the target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, that is, screen out the target intelligent customer service from all intelligent customer services according to the service scope in the intelligent customer service strategy, and then switch the target intelligent customer service to the service mode corresponding to the problem category through the problem category, so as to serve the target customer through the target intelligent customer service in the service mode.

[0034] In the customer service scheduling method of the present invention, through the consultation information and customer information included in the incoming request, the determination of the customer group classification result is realized, and the classification of the problem category is realized. Furthermore, the determination of the request priority is realized, avoiding the problem of long waiting time for high-quality customers. By detecting whether the consultation information is a historical consultation question, the determination of the intelligent customer service channel is realized, reducing the labor cost, and then the determination of the intelligent customer service strategy is realized. Through the intelligent customer service strategy, the screening of the target intelligent customer service and the switching of the service mode are realized, thus realizing the diversion of customers, improving the efficiency of customer service, and further enhancing customer satisfaction.

[0035] In one embodiment, in step S20, that is, classifying the target customers according to the customer information to obtain a customer group classification result, including: S201, determining whether there is a customer group label in the customer information.

[0036] S202, when there is a customer group label, determining the customer group classification result corresponding to the target customer according to the customer group label.

[0037] S203, when there is no customer group label, extracting the feature information in the customer information, and classifying the feature information through a preset customer group label set to obtain a customer group classification result.

[0038] Understandably, the customer group label is used to represent the identifiers of different customer groups. For example, the special excellent customer group, the high-quality customer group, the growth customer group, the ordinary customer group, and the new customer group. The customer group classification result is used to represent the customer group to which the target customer belongs. The preset customer group label set refers to the features corresponding to each customer group for tagging.

[0039] Specifically, after receiving the incoming request, the customer information is parsed to identify whether there is a customer group label in the customer information. When there is a customer group label, the customer group classification result corresponding to the target customer is determined according to the customer group label, that is, the customer group to which the target customer belongs is determined through the customer group label. When there is no customer group label, the feature information in the customer information is extracted, that is, the customer information is parsed, and the features in the customer information are extracted through the feature extraction rules to obtain the feature information. Then, the preset customer group label set is obtained, and the extracted feature information is matched with the features corresponding to each customer group label, so as to determine the customer group with the most features corresponding to the feature information as the customer group classification result corresponding to the target customer.

[0040] In this embodiment, by determining whether there is a customer group label in the customer information, it is possible to judge whether the target customer is a new customer or an old customer. When there is a customer group label, the rapid determination of the customer group classification result is realized. When there is no customer group label, the extraction of the feature information is realized, so as to realize the identification of the customer group to which the target customer belongs, and further realize the determination of the customer group classification result.

[0041] In one embodiment, after step S50, that is, after detecting whether the consultation information is a historical consultation question, it further includes: S801, if the consultation information is not a historical consultation question, then connecting the incoming request to the artificial customer service channel.

[0042] S802, determining the artificial customer service strategy according to the question category, the request priority, and the customer group classification result.

[0043] S803. Screen out the first target customer service staff from the human customer service channels according to the human customer service strategy to serve the target customer.

[0044] Understandably, the human customer service channel refers to the channel through which human customer service serves customers. The human customer service strategy refers to the formulated service methods for human customer service. The first target customer service staff refers to the customer service staff screened out from all human customer services to serve the target customer.

[0045] Specifically, after detecting whether the consultation information is a historical consultation question, if the consultation information is not a historical consultation question, the incoming request is connected to the human customer service channel. Then, obtain the problem category, request priority, and customer group classification result corresponding to the target customer, and determine the human customer service strategy according to the problem category, request priority, and customer group classification result, that is, first screen out the service strategy corresponding to the customer group classification result, and then screen out the service strategy corresponding to the request priority from the service strategies corresponding to the customer group classification result. Finally, screen out the service strategy corresponding to the problem category from the service strategies corresponding to the request priority and determine it as the human customer service strategy. Next, screen out the first target customer service staff from the human customer service channel according to the human customer service strategy to serve the target customer, that is, screen out the customer service staff corresponding to the service scope through the service scope in the human customer service strategy, and screen out the customer service staff with a shorter waiting time for the target customer again. In this way, the first target customer service staff can be obtained. Among them, if there are multiple idle customer service staff, obtain the customer service staff skill information, and then screen out the first target customer service staff through the customer service staff skill information.

[0046] In this embodiment, when the consultation information is not a historical consultation question, the incoming request is connected to the human channel. Through the problem category, request priority, and customer group classification result, the determination of the human customer service strategy is realized, and then the screening of the first target customer service staff is realized.

[0047] In one embodiment, in step S803, that is, screening out the first target customer service staff from the human customer service channel according to the human customer service strategy, includes: S8031. Obtain the customer service mapping relationship, and determine the customer service group list according to the human customer service strategy and the customer service mapping relationship.

[0048] S8032. Obtain the reception quantity corresponding to each customer service staff in the customer service group list and the customer service staff skill information, and determine the first target customer service staff according to the reception quantity and the customer service staff skill information.

[0049] Understandably, the customer service mapping relationship refers to the corresponding relationship between the list and the service policy. The customer service group list is a statistical list of all human customer services.

[0050] Specifically, after obtaining the human customer service policy, obtain the customer service mapping relationship. According to the human customer service policy and the customer service mapping relationship, determine the customer service group list, that is, match the preset service policies in the human customer service policy and the customer service mapping relationship. When the match is successful, determine the customer service group list corresponding to the preset service policy in the customer service mapping relationship as the customer service group list corresponding to the human customer service policy. Then, obtain the reception quantity of each customer service staff and the customer service staff skill information from the customer service group list. Next, determine the first target customer service staff according to the reception quantity and the customer service staff skill information, that is, match the problem category and request priority corresponding to the human customer service policy with the customer service staff skill information and reception quantity of each customer service staff to screen out the first target customer service staff who meets the target customer from all customer service staff.

[0051] In this embodiment, through the customer service mapping relationship, the rapid determination of the customer service group list is realized, thereby realizing the determination of the reception quantity of each customer service staff and the determination of the customer service staff skill information, and further realizing the screening of the first target customer service staff.

[0052] In one embodiment, after step S70, that is, after screening out the target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service policy and switching the target intelligent customer service to the service mode corresponding to the problem category, it further includes: S901, obtain the reply waiting time of the target customer and the preset waiting time threshold.

[0053] S902, compare the reply waiting time with the preset waiting time threshold. If the reply waiting time exceeds the preset waiting time threshold, transfer from the intelligent customer service channel to the human customer service channel.

[0054] S903, determine the human customer service policy according to the problem category, the request priority, and the customer group classification result.

[0055] S904, screen out the second target customer service staff from the human customer service channel according to the human customer service policy to serve the target customer.

[0056] Understandably, the reply waiting time refers to the time period from when the customer asks each question until the intelligent customer service answers. The preset waiting time threshold refers to the maximum value of the waiting time set in advance. The second target customer service staff member refers to the customer service staff member who establishes a call with the target customer after the waiting time exceeds the threshold and transfers from the intelligent customer service channel to the manual customer service channel.

[0057] Specifically, after switching the target intelligent customer service to the service mode corresponding to the question category, obtain the reply waiting time of the target customer, that is, obtain the time when the target customer sends each consultation message, and obtain the time when the target intelligent customer service replies to each consultation, and calculate the difference between the two times to obtain the reply waiting time. Then, preset the waiting time threshold, and compare the reply waiting time with the preset waiting time threshold. If the reply waiting time exceeds the preset waiting time threshold, transfer from the intelligent customer service channel to the manual customer service channel. If the reply waiting time is less than or equal to the preset waiting time threshold, continue to use the intelligent customer service channel for reply. Among them, to avoid a certain question being difficult, it is also possible to determine whether the total reply waiting time of multiple questions exceeds the total waiting time threshold. For example, whether it exceeds 3 replies within 5 minutes. Further, obtain the question category, request priority, and customer group classification result corresponding to the target customer, and determine the manual customer service strategy according to the question category, request priority, and customer group classification result. That is, first screen out the service strategy corresponding to the customer group classification result, then, screen out the service strategy corresponding to the request priority from the service strategies corresponding to the customer group classification result, and finally, screen out the service strategy corresponding to the question category from the service strategies corresponding to the request priority, and determine it as the manual customer service strategy. Then, screen out the second target customer service staff member from the manual customer service channel according to the manual customer service strategy to serve the target customer. That is, screen out the customer service staff member corresponding to the service scope through the service scope in the manual customer service strategy, and screen out again the customer service staff member who makes the target customer wait for a shorter time. In this way, the second target customer service staff member can be obtained. Among them, if there are multiple idle customer service staff members, obtain the customer service staff member skill information, and then screen out the second target customer service staff member through the customer service staff member skill information.

[0058] In this embodiment, by comparing the reply waiting time with the preset waiting time threshold, the judgment of whether to transfer to the manual customer service channel is realized. Through the question category, request priority, and customer group classification result, the determination of the manual customer service strategy is realized, and further the screening of the second target customer service staff member is realized, as well as the service for the target customer is realized.

[0059] In one embodiment, after step S70, that is, after screening out the target intelligent customer service from all the intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy and switching the target intelligent customer service to the service mode corresponding to the problem category, the following steps are further included: S905, obtain the reply voice of the target intelligent customer service and a preset number threshold, and count the number of repetitions of the reply voice.

[0060] S906, compare the number of repetitions with the preset number threshold, and when the number of repetitions is greater than or equal to the preset number threshold, transfer from the intelligent customer service channel to the manual customer service channel.

[0061] S907, determine the manual customer service strategy according to the problem category, the request priority, and the customer group classification result.

[0062] S908, screen out the third target customer service staff from the manual customer service channel according to the manual customer service strategy to serve the target customer.

[0063] It can be understood that the reply voice refers to the content replied by the intelligent customer service to the question asked by the target customer. The preset number threshold refers to the maximum value of the repetition of the reply content of the intelligent customer service. The third target customer service staff refers to the customer service staff who establishes a call with the target customer after transferring from the intelligent customer service channel due to the number of repetitions of the reply content exceeding the threshold. The number of repetitions refers to the number of times the content of the reply voice is the same.

[0064] Specifically, after switching the target intelligent customer service to the service mode corresponding to the problem category, obtain the reply voice of the target intelligent customer service, convert the reply voice into a reply text, and calculate the similarity of multiple consecutive reply texts of the target intelligent customer service to determine whether the reply texts are repeated. When the reply texts are repeated, count the number of consecutive repetitions of the reply texts and determine it as the repetition count of the reply voice. In another embodiment, directly calculate the similarity of multiple consecutive reply voices of the target intelligent customer service to determine whether the reply voices are the same. When the reply voices are the same, count the number of times the reply voices are the same and determine it as the repetition count of the reply voice. Then, obtain a preset count threshold, compare the repetition count with the preset count threshold, and when the repetition count is greater than or equal to the preset count threshold, transfer from the intelligent customer service channel to the manual customer service channel. When the repetition count is less than the preset count threshold, continue to use the intelligent customer service channel for answering. Further, obtain the problem category, request priority, and customer group classification result corresponding to the target customer, and determine the manual customer service strategy according to the problem category, request priority, and customer group classification result, that is, first screen out the service strategy corresponding to the customer group classification result, then, screen out the service strategy corresponding to the request priority from the service strategies corresponding to the customer group classification result, and finally, screen out the service strategy corresponding to the problem category from the service strategies corresponding to the request priority and determine it as the manual customer service strategy. Next, screen out the second target customer service staff from the manual customer service channel according to the manual customer service strategy to serve the target customer, that is, screen out the customer service staff corresponding to the service scope through the service scope in the manual customer service strategy, and screen out the customer service staff with a shorter waiting time for the target customer again. In this way, the third target customer service staff can be obtained. Among them, if there are multiple idle customer service staff, obtain the customer service staff skill information, and then screen out the third target customer service staff through the customer service staff skill information.

[0065] In this embodiment, through the repetition count and the preset count threshold, the judgment of whether to transfer to the manual customer service channel is realized. Through the problem category, request priority, and customer group classification result, the determination of the manual customer service strategy is realized, and then the screening of the third target customer service staff is realized, as well as the service for the target customer.

[0066] In one embodiment, after step S70, that is, after serving the target customer, it further includes: S1001, record the consultation service data of the target customer, and determine the actual consultation problem of the target customer through the consultation service data.

[0067] S1002. When the actual consultation question is a historical consultation question, update and train all the intelligent customer services corresponding to the historical consultation question according to the consultation service data, so as to update the service mode corresponding to the historical consultation question.

[0068] S1003. When the actual consultation question is not a historical consultation question, determine that the actual consultation question is a new consultation question, allocate an intelligent customer service for the new consultation question, and perform iterative training on all the intelligent customer services corresponding to the new consultation question according to the consultation service data, so as to add a service mode corresponding to the new consultation question to the intelligent customer service.

[0069] Understandably, the consultation service data refers to all the questions asked by the target customer this time, and the reply voices of all the questions. The actual consultation question refers to the intention of the target customer's inquiry this time.

[0070] Specifically, after serving the target customer, record the consultation service data of the target customer. Then, determine the actual consultation question of the target customer through the consultation service data, obtain the intention recognition model, and perform intention recognition on the consultation service data through the intention recognition model, so as to obtain the intention of the target customer, and determine the recognized intention as the actual consultation question of the target customer. Further, judge whether the actual consultation question is a historical consultation question, that is, match the actual consultation question with all historical consultation questions one by one, and when the similarity is greater than the preset threshold, determine that the actual consultation question is a historical consultation question, otherwise, determine that the actual consultation question is not a historical consultation question. When the actual consultation question is a historical consultation question, update and train all the intelligent customer services corresponding to the historical consultation question according to the consultation service data, so as to update the service mode corresponding to the historical consultation question. When the actual consultation question is not a historical consultation question, determine that the actual consultation question is a new consultation question, allocate an intelligent customer service for the new consultation question, and perform iterative training on all the intelligent customer services corresponding to the new consultation question according to the consultation service data, that is, use the customer consultation as the training data and the artificial reply voice as the label to train the newly added intelligent customer service until convergence, so as to add a service mode corresponding to the new consultation question to the intelligent customer service.

[0071] In this embodiment, through the recorded consultation service data, the determination of the actual consultation question is realized, and further the judgment of whether the actual consultation question is a historical consultation question is realized. When the actual consultation question is a historical consultation question, through the consultation service data, the update of all the intelligent customer services corresponding to the historical consultation question and the update of the service mode are realized. When the actual consultation question is not a historical consultation question, the addition of all the intelligent customer services corresponding to the historical consultation question and the addition of the service mode are realized.

[0072] It should be understood that the sequence numbers of the steps in the above embodiments do not indicate the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0073] In one embodiment, a customer service scheduling device is provided, which corresponds one-to-one with the customer service scheduling method in the above embodiment. As Figure 8 shown, the customer service scheduling device includes a request receiving module 10, a customer group classification module 20, a problem classification module 30, a request priority module 40, an intelligent customer service channel module 50, a customer service strategy module 60, and a customer service screening module 70. The detailed description of each functional module is as follows: The request receiving module 10 is configured to receive an incoming request from a target customer, and the incoming request includes consultation information and customer information; The customer group classification module 20 is configured to classify the target customer according to the customer information to obtain a customer group classification result; The problem classification module 30 is configured to classify the consultation information to obtain a problem category; The request priority module 40 is configured to determine a request priority corresponding to the incoming request according to the customer group classification result and the problem category; The intelligent customer service channel module 50 is configured to detect whether the consultation information is a historical consultation problem. If the consultation information is a historical consultation problem, the incoming request is connected to the intelligent customer service channel; the historical consultation problem refers to all problems that the customer has consulted in a preset historical period before the current time point; The customer service strategy module 60 is configured to determine an intelligent customer service strategy according to the problem category, the request priority, and the customer group classification result; The customer service screening module 70 is configured to screen out a target intelligent customer from all intelligent customers in the intelligent customer service channel according to the intelligent customer service strategy, and switch the target intelligent customer to a service mode corresponding to the problem category to serve the target customer through the service mode.

[0074] Optionally, the customer group classification module 20 includes: A customer group label detection unit, configured to determine whether there is a customer group label in the customer information; A label existence unit, configured to determine a customer group classification result corresponding to the target customer according to the customer group label when there is a customer group label; A label non-existence unit, configured to extract feature information from the customer information and perform customer group classification on the feature information through a preset customer group label set to obtain a customer group classification result when there is no customer group label.

[0075] Optionally, the device further includes: An artificial customer service channel module, configured to connect the incoming request to the artificial customer service channel if the consultation information is not a historical consultation question; A first artificial service strategy module, configured to determine an artificial customer service strategy according to the problem category, the request priority, and the customer group classification result; A first target customer service staff module, configured to screen out a first target customer service staff from the artificial customer service channel according to the artificial customer service strategy to serve the target customer.

[0076] Optionally, the first target customer service staff module includes: A customer service group list unit, configured to obtain a customer service mapping relationship and determine a customer service group list according to the artificial customer service strategy and the customer service mapping relationship; A customer service staff determination unit, configured to obtain the reception quantity corresponding to each customer service staff and the customer service staff skill information in the customer service group list, and determine a first target customer service staff according to the reception quantity and the customer service staff skill information.

[0077] Optionally, the device further includes: A waiting time acquisition module, configured to acquire the reply waiting time of the target customer and a preset waiting time threshold; A waiting time comparison module, configured to compare the reply waiting time with the preset waiting time threshold, and if the reply waiting time exceeds the preset waiting time threshold, transfer from the intelligent customer service channel to the artificial customer service channel; A second artificial service strategy module, configured to determine an artificial customer service strategy according to the problem category, the request priority, and the customer group classification result; A second target customer service staff module, configured to screen out a second target customer service staff from the artificial customer service channel according to the artificial customer service strategy to serve the target customer.

[0078] Optionally, the device further includes: A repetition times acquisition module, configured to acquire the reply voice of the target intelligent customer service and a preset times threshold, and count the repetition times of the reply voice; A repetition times comparison module, configured to compare the repetition times with the preset times threshold, and when the repetition times is greater than or equal to the preset times threshold, transfer from the intelligent customer service channel to the artificial customer service channel; A third artificial service strategy module, configured to determine an artificial customer service strategy according to the problem category, the request priority, and the customer group classification result; A third target customer service staff module, configured to screen out a third target customer service staff from the manual customer service channels according to the manual customer service strategy to serve the target customer.

[0079] Optionally, the device further includes: A consultation service data recording module, configured to record the consultation service data of the target customer, and determine the actual consultation problem of the target customer through the consultation service data; An intelligent customer service update and training module, configured to, when the actual consultation problem is a historical consultation problem, update and train all the intelligent customer services corresponding to the historical consultation problem according to the consultation service data, so as to update the service mode corresponding to the historical consultation problem; An intelligent customer service iterative training module, configured to, when the actual consultation problem is not a historical consultation problem, determine that the actual consultation problem is a new consultation problem, allocate an intelligent customer service to the new consultation problem, and perform iterative training on all the intelligent customer services corresponding to the new consultation problem according to the consultation service data, so as to add a service mode corresponding to the new consultation problem to the intelligent customer service.

[0080] The present application further provides a Figure 9 computer device as shown, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor is configured to execute the customer service scheduling method described above.

[0081] For the specific definitions of the computer device, the processor, and its respective units and modules, reference may be made to the definitions of the customer service scheduling method in the foregoing text, which will not be elaborated herein. Each module in the foregoing processor may be implemented in whole or in part by software, hardware, and their combination. Understandably, the processor includes a processor, a memory, a network interface, and a database connected through a device bus. Each module of the processor may be embedded in or independent of the processor in the form of hardware, or stored in the memory in the form of software, so that the processor can call and execute the operations corresponding to the foregoing respective modules. Among them, the processor is configured to provide computing and control capabilities. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating device, a computer program, and a database. The internal memory provides an environment for the operation of the operating device and the computer program in the non-volatile storage medium. The database is used to store the data used in the customer service scheduling method in the foregoing embodiments. The network interface is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements a customer service scheduling method.

[0082] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned customer service scheduling method is implemented.

[0083] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned method embodiments. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0084] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above-mentioned functions can be allocated to different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0085] The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention 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 described in the foregoing embodiments, or perform equivalent replacements for 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 various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A customer service scheduling method, characterized in that, Including: Receiving an incoming request from a target customer, where the incoming request includes consultation information and customer information; Classifying the target customer into a customer group according to the customer information to obtain a customer group classification result; Classifying the consultation information to obtain a problem category; Determining a request priority corresponding to the incoming request according to the customer group classification result and the problem category; Detecting whether the consultation information is a historical consultation problem. If the consultation information is a historical consultation problem, then connecting the incoming request to the intelligent customer service channel; The historical consultation problem refers to the problems that all customers have consulted during a preset historical period before the current time point; Determining an intelligent customer service strategy according to the problem category, the request priority, and the customer group classification result; Selecting a target intelligent customer service from all intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, and switching the target intelligent customer service to a service mode corresponding to the problem category to serve the target customer through the service mode.

2. The customer service scheduling method according to claim 1, wherein, After detecting whether the consultation information is a historical consultation problem, it further includes: If the consultation information is not a historical consultation problem, then connecting the incoming request to the manual customer service channel; Determining a manual customer service strategy according to the problem category, the request priority, and the customer group classification result; Selecting a first target customer service staff from the manual customer service channel according to the manual customer service strategy to serve the target customer.

3. The customer service scheduling method according to claim 2, wherein The selecting a first target customer service staff from the manual customer service channel according to the manual customer service strategy includes: Obtaining a customer service mapping relationship, and determining a list of customer service groups according to the manual customer service strategy and the customer service mapping relationship; Obtaining the reception quantity corresponding to each customer service staff in the list of customer service groups and the customer service staff skill information, and determining the first target customer service staff according to the reception quantity and the customer service staff skill information.

4. The customer service scheduling method according to claim 1, wherein After serving the target customer through the service mode, it further includes: Obtaining the reply waiting time of the target customer and a preset waiting time threshold; Comparing the reply waiting time with the preset waiting time threshold. If the reply waiting time exceeds the preset waiting time threshold, then transferring from the intelligent customer service channel to the manual customer service channel; Determining a manual customer service strategy according to the problem category, the request priority, and the customer group classification result; Selecting a second target customer service staff from the manual customer service channel according to the manual customer service strategy to serve the target customer.

5. The customer service scheduling method according to claim 1, wherein After serving the target customer through the service mode, it further includes: Obtaining the reply voice of the target intelligent customer service and a preset number threshold, and counting the repetition times of the reply voice; Comparing the repetition times with the preset number threshold. When the repetition times are greater than or equal to the preset number threshold, transferring from the intelligent customer service channel to the manual customer service channel; Determining a manual customer service strategy according to the problem category, the request priority, and the customer group classification result; Screen out a third target customer service staff from the artificial customer service channels according to the artificial customer service strategy to serve the target customer.

6. The customer service scheduling method according to any one of claims 1 to 5, characterized in that, After serving the target customer, it further includes: Recording the consultation service data of the target customer, and determining the actual consultation problem of the target customer through the consultation service data; When the actual consultation problem is a historical consultation problem, update and train all the intelligent customer services corresponding to the historical consultation problem according to the consultation service data to update the service mode corresponding to the historical consultation problem; When the actual consultation problem is not a historical consultation problem, determine that the actual consultation problem is a new consultation problem, allocate an intelligent customer service for the new consultation problem, and perform iterative training on all the intelligent customer services corresponding to the new consultation problem according to the consultation service data to add a service mode corresponding to the new consultation problem to the intelligent customer service.

7. The customer service scheduling method according to claim 1, characterized in that The classifying the target customer into customer groups according to the customer information to obtain a customer group classification result includes: Determining whether there is a customer group label in the customer information; When there is a customer group label, determining the customer group classification result corresponding to the target customer according to the customer group label; When there is no customer group label, extracting the feature information in the customer information, and performing customer group classification on the feature information through a preset customer group label set to obtain a customer group classification result.

8. A customer service scheduling device, characterized in that, It includes: A request receiving module, configured to receive an incoming request from a target customer, where the incoming request includes consultation information and customer information; A customer group classification module, configured to classify the target customer into customer groups according to the customer information to obtain a customer group classification result; A problem classification module, configured to classify the consultation information to obtain a problem category; A request priority module, configured to determine the request priority corresponding to the incoming request according to the customer group classification result and the problem category; An intelligent customer service channel module, configured to detect whether the consultation information is a historical consultation problem. If the consultation information is a historical consultation problem, then connect the incoming request to the intelligent customer service channel; The historical consultation problem refers to all problems that the customer has consulted in a preset historical period before the current time point; A customer service strategy module, configured to determine an intelligent customer service strategy according to the problem category, the request priority, and the customer group classification result; A customer service screening module, configured to screen out a target intelligent customer service from all the intelligent customer services in the intelligent customer service channel according to the intelligent customer service strategy, and switch the target intelligent customer service to a service mode corresponding to the problem category to serve the target customer through the service mode.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is configured to execute the customer service scheduling method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the customer service scheduling method according to any one of claims 1 to 7.

Images