Session processing method and device, server and storage medium

Abstract

The invention provides a session processing method and device, a server and a storage medium, relates to the technical field of communication, and solves the technical problem of relatively low session processing effectiveness caused by the fact that an idle customer service selected by the server may not accurately answer a problem proposed by a user. The method comprises the following steps: acquiring a historical session record in a target session in response to a session processing request of a user account; based on the historical session record, a label of each customer service in a plurality of customer services and a session state of each customer service, determining a recommendation score of each customer service; and determining the customer service with the highest recommendation score as a target customer service, and accessing the target customer service to the target session, so that the target customer service processes the target session.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a session processing method, apparatus, server, and storage medium. Background Art

[0002] Currently, when a user account consults a question during a session, it can request a human customer service to handle the current session, and the server can select an idle customer service to handle the consultation session.

[0003] However, in the above method, since different users consult different questions, the idle customer service selected by the server may not be able to accurately answer the questions raised by the users, resulting in low effectiveness of session processing. Summary of the Invention

[0004] This application provides a session processing method, apparatus, server, and storage medium, which solves the technical problem that since different users consult different questions, the idle customer service selected by the server may not be able to accurately answer the questions raised by the users, resulting in low effectiveness of session processing.

[0005] In a first aspect, this application provides a session processing method, including: in response to a session processing request of a user account, obtaining a historical session record in a target session, where the session processing request is used to request a human customer service to process the target session, the target session is the current session of the user account, and the historical session record includes the questions consulted by the user account; based on the historical session record, labels of each customer service among multiple customer services, and the session status of each customer service, determining a recommended score for each customer service, where a recommended score of a customer service is used to represent the effectiveness degree of the customer service in processing the target session, and a label of a customer service is used to represent the type of questions corresponding to the customer service; determining the customer service with the highest recommended score as the target customer service, and connecting the target customer service to the target session so that the target customer service processes the target session.

[0006] Optionally, the session status of a customer service includes the current number of sessions and the idle time. Determining the recommended score of the first customer service based on the historical session record, the label of the first customer service, and the session status of the first customer service may specifically include: based on the historical session record and the label of the first customer service, determining a matching score of the first customer service, where the matching score of the first customer service is used to represent the matching degree between the first customer service and the user account; based on the number of sessions of the first customer service and the idle time of the first customer service, determining an idle score of the first customer service, where the idle score of the first customer service is used to represent the idle degree of the first customer service; determining the recommended score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service.

[0007] Optionally, determining the recommendation score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service may specifically include: determining the recommendation score of the first customer service based on a first weight, the matching score of the first customer service, a second weight, and the idle score of the second customer service, where the first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommendation score.

[0008] Optionally, the session processing method further includes: when the number of sessions of each customer service is equal to the session number threshold, obtaining the sorting of at least one session to be processed, and the problem types of the at least one session to be processed; when the priority of the target problem type is higher than the priority threshold, adjusting the serial number of the target session to the target position, where the target position is: before the serial number where the problem type with a priority lower than the first priority is located among the problem types of the at least one session to be processed, the first priority is the priority of the target problem type, and the target problem type is the problem type included in the historical session record.

[0009] Optionally, the session processing method further includes: obtaining a first question sent by the user account; determining a label of the user account based on the identifier of the user account; determining at least one reply message from a database based on the label of the user account and the first question; and sending the at least one reply message to the user account.

[0010] In a second aspect, the present application provides a session processing apparatus, including: an obtaining module, a determining module, and a processing module; the obtaining module is configured to obtain a historical session record in a target session in response to a session processing request of a user account, where the session processing request is used to request an artificial customer service to process the target session, the target session is the session currently being carried out by the user account, and the historical session record includes the question consulted by the user account; the determining module is configured to determine the recommendation score of each customer service based on the historical session record, the label of each customer service among a plurality of customer services, and the session status of each customer service, where the recommendation score of a customer service is used to represent the effectiveness of the customer service in processing the target session, and the label of a customer service is used to represent the problem type corresponding to the customer service; the determining module is further configured to determine the customer service with the highest recommendation score as the target customer service; the processing module is configured to connect the target customer service to the target session so that the target customer service processes the target session.

[0011] Optionally, the session state of a customer service agent includes the current number of sessions and the idle time. The determination module is specifically configured to determine the matching score of the first customer service agent based on the historical session record and the label of the first customer service agent. The matching score of the first customer service agent is used to represent the matching degree between the first customer service agent and the user account. The determination module is further specifically configured to determine the idle score of the first customer service agent based on the number of sessions of the first customer service agent and the idle time of the first customer service agent. The idle score of the first customer service agent is used to represent the idle degree of the first customer service agent. The determination module is further specifically configured to determine the recommendation score of the first customer service agent based on the matching score of the first customer service agent and the idle score of the first customer service agent.

[0012] Optionally, the determination module is further specifically configured to determine the recommendation score of the first customer service agent based on a first weight, the matching score of the first customer service agent, a second weight, and the idle score of the second customer service agent. The first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommendation score.

[0013] Optionally, the obtaining module is further configured to obtain the sorting of at least one session to be processed and the problem types of the at least one session to be processed when the number of sessions of each customer service agent is equal to the session number threshold. The processing module is further configured to adjust the serial number of the target session to the target position when the priority of the target problem type is higher than the priority threshold. The target position is: before the serial number where the problem type with a priority lower than the first priority is located among the problem types of the at least one session to be processed. The first priority is the priority of the target problem type, and the target problem type is the problem type included in the historical session record.

[0014] Optionally, the session processing device further includes a sending module. The obtaining module is further configured to obtain the first question sent by the user account. The determination module is further configured to determine the label of the user account based on the identifier of the user account. The determination module is further configured to determine at least one reply message from the database based on the label of the user account and the first question. The sending module is configured to send the at least one reply message to the user account.

[0015] In a third aspect, the present application provides a server, including: a processor and a memory configured to store processor-executable instructions; wherein, the processor is configured to execute the instructions to implement any one of the optional session processing methods in the first aspect above.

[0016] In a fourth aspect, the present application provides a computer-readable storage medium, on which instructions are stored. When the instructions in the computer-readable storage medium are executed by a server, the server is enabled to execute any one of the optional session processing methods in the first aspect above.

[0017] The session processing method, apparatus, server, and storage medium provided by this application. The server can respond to a session processing request of a user account, obtain the historical session record in the target session, and then determine the recommended score for each of multiple customer service representatives based on the historical session record, the label of each customer service representative, and the session status of each customer service representative. The customer service representative with the highest recommended score is determined as the target customer service representative, and the target customer service representative is connected to the target session to process the target session. In this application, since the historical session record includes the questions consulted by the user account and the label of each customer service representative is used to represent the problem type corresponding to each customer service representative, the server determines the recommended score for each customer service representative based on the historical session record, the label of each customer service representative, and the session status of each customer service representative, which can improve the accuracy of determining the effectiveness of each customer service representative in processing the target session. Also, since the target customer service representative is the one with the highest recommended score, after the server connects the target customer service representative to the target session, the target customer service representative can process the target session quickly and accurately. Description of the Drawings

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art.

[0019] Figure 1 Schematic diagram of the network architecture of the session processing system provided by the embodiment of this application;

[0020] Figure 2 Schematic flowchart of a session processing method provided by the embodiment of this application;

[0021] Figure 3 Schematic flowchart of another session processing method provided by the embodiment of this application;

[0022] Figure 4 Schematic flowchart of another session processing method provided by the embodiment of this application;

[0023] Figure 5 Schematic flowchart of another session processing method provided by the embodiment of this application;

[0024] Figure 6 Schematic flowchart of another session processing method provided by the embodiment of this application;

[0025] Figure 7 Schematic diagram of the internal structure of a server provided by the embodiment of this application;

[0026] Figure 8 Schematic flowchart of another session processing method provided by the embodiment of this application;

[0027] Figure 9Schematic structural diagram of a session processing device provided by an embodiment of the present application;

[0028] Figure 10 Schematic structural diagram of another session processing device provided by an embodiment of the present application. Detailed implementation manners

[0029] The session processing method, device, server, and storage medium provided by embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0030] In the description of the present application, terms such as "first" and "second" in the specification and the accompanying drawings are used to distinguish different objects, rather than to describe a specific order of the objects. For example, the first weight and the second weight are used to distinguish different weights, rather than to describe a specific order of the weights.

[0031] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes other steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0032] It should be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0033] In the present application, the "and / or" includes any one of the two methods or the use of both methods simultaneously.

[0034] In the description of the present application, unless otherwise specified, the meaning of "a plurality of" refers to two or more.

[0035] The emergence of online customer service systems based on web sessions has gradually replaced traditional customer service tools. As a new type of professional web customer service tool, it offers a good user experience and is mainly used for convenient, fast, and instant communication between various website visitors and enterprises. With the rapid development of technology updates and the Internet, the online customer service system is no longer limited to the single function of instant messaging. It can also perform website traffic statistics, user relationship management, and collection of track information of visiting users. Along with the rise of the Internet, there are more and more online customer service systems, and their functions are becoming more complete and powerful. The online customer service system serves visiting users, and enterprises need to configure professional customer service personnel to provide service support. At the moment when a visitor initiates a transfer to a human agent, the session automatic distribution system distributes these sessions to each logged-in customer service personnel according to the established rules. As the online customer service system plays an increasingly important role in various fields, the number of visiting users is increasing day by day, and it is necessary to quickly and reasonably allocate customer service personnel to provide service support. The session automatic distribution system can also be embedded in any online customer service system. This patent mainly uses the session automatic distribution algorithm to enable the online customer service system to more quickly and reasonably allocate customer service for service support.

[0036] The online customer service system relies on the website as a medium to provide instant communication page communication technology between Internet users and internal website personnel, and belongs to the basic components of network marketing. In the process of the rapid development of the Internet, the number of users is increasing day by day, and the number of consultation people, consultation questions, and supporting customer service personnel is also increasing synchronously. How to quickly and reasonably allocate sessions to customer service, do a good job in answering questions and service support, improve user experience, and improve work efficiency will become particularly important.

[0037] In the current mainstream session automatic distribution system, usually when a user creates a session and initiates a transfer to a human agent, the user is required to check the type of consultation question, and then query whether there are online support personnel in the corresponding question group. If so, the customer service is connected to the session; if not, the session is added to the queue for waiting. In this traditional mode scenario, in the link where the user initiates a question consultation, the true problems and core appeal points of the user cannot be accurately grasped. Therefore, when the customer service connects to the session, the primary task is to confirm the true intention of the user, which greatly reduces the customer service support effect. In extreme cases, the user needs to create a session again and again, transfer to a human agent until the correct support customer service is connected to solve the user's problem, and the user experience becomes very poor, and the service effect of the customer service will also become low.

[0038] As described in the background art, in the related art, the idle customer service selected by the server may not be able to accurately answer the questions raised by the user, resulting in low effectiveness of session processing. Based on this, the embodiments of the present application provide a session processing method, apparatus, server, and storage medium. Since the historical session record includes the questions consulted by the user account and the labels of each customer service are used to characterize the question types corresponding to each customer service, the server determines the recommended score of each customer service based on the historical session record, the labels of each customer service, and the session status of each customer service, which can improve the accuracy of determining the effectiveness of each customer service in processing the target session. Also, since the target customer service is the customer service with the highest recommended score, after the server intervenes the target customer service in the target session, the target customer service can quickly and accurately process the target session.

[0039] A session processing method, apparatus, server, and storage medium provided by the embodiments of the present application can be applied to a session processing system, such as Figure 1 shown. The session processing system includes a terminal 101 and a server 102. Usually, in practical applications, the connections between the above devices can be wireless connections. For the convenience of intuitively representing the connection relationships between the devices, Figure 1 solid lines are used for illustration in

[0040] Among them, the terminal 101 can send a session processing request to the server, and the server 102 can respond to the session processing request, obtain the historical session record in the target session, and then determine the recommended score of each customer service based on the historical session record, the labels of each customer service among multiple customer services, and the session status of each customer service, and determine the customer service with the highest recommended score as the target customer service, and connect the target customer service to the target session.

[0041] The session processing method, apparatus, server, and storage medium provided by the embodiments of the present application are applied to the scenario of session processing. When the server receives a session processing request of a user account, it can obtain the historical session record of the current session of the user account, and then determine the recommended score of each customer service based on the historical session record, the labels of each customer service, and the session status of each customer service, and intervene the customer service with the highest recommended score in the target session.

[0042] Combined with the above Figure 1 shown session processing system, as Figure 2 shown, when the session processing method provided by the embodiments of the present application is applied to the server 102, the session processing method may include S101 - S103.

[0043] S101. In response to a session processing request of a user account, the server obtains the historical session record in the target session.

[0044] Among them, the session processing request is used to request a human customer service to process the target session, where the target session is the session currently being conducted by the user account, and the historical session record includes the questions consulted by the user account.

[0045] It should be understood that before sending the session processing request, the user account needs to first create the target session, where the target session is a conversation between the user account and the server. The user account can consult questions with the server through the target session, and the server can return reply information to the user account.

[0046] In the embodiment of the present application, the server in the target session can also be understood as an intelligent customer service.

[0047] It can be understood that the historical session record may also include the reply information of the server.

[0048] In an optional implementation manner, the terminal can initiate the session processing request in response to a click operation of the user account on a certain control.

[0049] In another optional implementation manner, when the number of sessions in the target session is greater than or equal to the number threshold, the server can send a pop-up window on the page of the target session. The pop-up window is used to prompt the user to initiate the session processing request, and the user account can click on the pop-up window to send the session processing request.

[0050] S102. The server determines the recommended score of each customer service based on the historical session record, the label of each customer service among multiple customer services, and the session status of each customer service.

[0051] Among them, the recommended score of a customer service is used to represent the effectiveness of the customer service in processing the target session, and the label of a customer service is used to represent the type of questions corresponding to the customer service.

[0052] It can be understood that the type of questions corresponding to a customer service is the type of questions that the customer service is good at processing.

[0053] Optionally, the label of a customer service can also be used to represent the field that the customer service is good at.

[0054] It should be understood that the session status of a customer service is used to represent the busy degree of the customer service, and the session status can be the status of being in the process of processing a session or not processing a session.

[0055] It can be understood that when the session status of the customer service is the "processing session" status, it indicates that the customer service is relatively busy and may not be able to handle new sessions. At this time, the effectiveness of the customer service in handling the target session is relatively low; when the session status of the customer service is the "unhandled session" status, it indicates that the customer service is relatively idle and can handle the target session more attentively. At this time, the effectiveness of the customer service in handling the target session is relatively high.

[0056] In the embodiment of the present application, the server can determine the similarity between the problems included in the historical session record and the problem types corresponding to the customer service. When the similarity is relatively high, it indicates that the customer service can answer the questions of the user account more accurately, and the effectiveness of the customer service in handling the target session is relatively high; when the similarity is relatively low, it indicates that the customer service may not be able to answer the questions of the user account accurately, and the effectiveness of the customer service in handling the target session is relatively low.

[0057] In an optional implementation manner, when the session status of the customer service is the "processing session" status, the server can determine that the recommended score of the customer service is 0. When the session status of the customer service is the "unhandled session" status, the server determines the recommended score of the customer service based on the similarity between the problems included in the historical session record and the problem types corresponding to the customer service.

[0058] S103. The server determines the customer service with the highest recommended score as the target customer service and connects the target customer service to the target session so that the target customer service can handle the target session.

[0059] It should be understood that when the recommended score of the customer service is the highest, it indicates that the effectiveness of the customer service in handling the target session is relatively high, and the customer service can quickly and accurately answer the questions of the customer. At this time, the server can connect the customer service to the target session.

[0060] The technical solutions provided by the above embodiments can at least bring the following beneficial effects: As can be seen from S101 - S103, the server can respond to the session processing request of the user account, obtain the historical session records in the target session, and then determine the recommended score for each customer service based on the historical session records, the labels of each customer service among multiple customer services, and the session status of each customer service. The customer service with the highest recommended score is determined as the target customer service, and the target customer service is connected to the target session to process the target session. In the embodiments of the present application, since the historical session records include the questions consulted by the user account, and the label of each customer service is used to represent the problem type corresponding to each customer service, the server determines the recommended score for each customer service based on the historical session records, the label of each customer service, and the session status of each customer service, which can improve the accuracy of determining the effectiveness of each customer service in processing the target session. Also, since the target customer service is the customer service with the highest recommended score, after the server connects the target customer service to the target session, the target customer service can process the target session quickly and accurately.

[0061] In an implementation manner of the embodiments of the present application, the session status of the one customer service includes the current number of sessions and the idle time. Combining Figure 2 , as Figure 3 shown, the server determines the recommended score of the first customer service based on the historical session records, the label of the first customer service, and the session status of the first customer service, which can be specifically implemented through S1021 - S1023.

[0062] S1021. The server determines the matching score of the first customer service based on the historical session records and the label of the first customer service.

[0063] Among them, the matching score of the first customer service is used to represent the matching degree between the first customer service and the user account.

[0064] In the embodiments of the present application, the server can determine the questions included in the historical session records, and then determine the similarity between the questions included in the historical session and the label of the first customer service as the matching score of the first customer service.

[0065] In an optional implementation manner, the server can use the cosine similarity based on the semantic vector model to determine the similarity between the questions included in the historical session and the label of the first customer service.

[0066] In another optional implementation manner, the server can also determine the same questions between the questions included in the historical session and the label of the first customer service, and then determine the ratio between the number of the same questions and the number of the labels of the first customer service as the similarity between the questions included in the historical session and the label of the first customer service.

[0067] In another alternative implementation, the server may determine the similarity between the user profile and the user profile of the first customer service agent as the matching score of the first customer service agent. Specifically, the similarity between the tags included in the user profile and the tags of the first customer service agent is determined as the matching score of the first customer service agent.

[0068] In the embodiments of the present application, multiple tags are stored in the server, and the server may tag the user account or the customer service agent.

[0069] It should be noted that the server may tag the same user with the same tag multiple times, that is, the number of the same tag type for one user may be multiple.

[0070] Exemplarily, assume that the tags of the user account one are A, B, and C, the number of times of tag A is 10, the number of times of tag B is 20, and the number of times of tag C is 10.

[0071] S1022. The server determines the idle score of the first customer service agent based on the number of sessions of the first customer service agent and the idle time of the first customer service agent.

[0072] Among them, the idle score of the first customer service agent is used to represent the idle degree of the first customer service agent.

[0073] It can be understood that a customer service agent can handle multiple sessions simultaneously. When the number of sessions being processed by the customer service agent is less, it indicates that the idle degree of the customer service agent is higher. When the number of sessions being processed by the customer service agent is more, it indicates that the idle degree of the customer service agent is lower.

[0074] It should be understood that the idle time of the customer service agent is the time when the customer service agent is in the state of not processing sessions.

[0075] In an alternative implementation, the server may set different idle time intervals, and each idle time interval corresponds to a score.

[0076] Optionally, the server determines that the idle score of a customer service agent satisfies: Formula 1:

[0077] Formula 1:

[0078]

[0079] Among them, F represents the idle score, x represents the number of sessions, and y represents the idle time.

[0080] S1023. The server determines the recommended score of the first customer service agent based on the matching score of the first customer service agent and the idle score of the first customer service agent.

[0081] In an alternative implementation, the server may determine the larger value between the matching score and the idle score as the recommended score for the first customer service representative.

[0082] In another alternative implementation, the server may also determine the sum of the matching score and the idle score as the recommended score for the first customer service representative.

[0083] In an alternative implementation, the server may first perform normalization processing on the matching score and the idle score, and then determine the recommended score based on the normalized matching score and idle score.

[0084] In the embodiments of the present application, the historical conversation record and the label of the first customer service representative can determine the matching degree between the field that the customer service is good at and the problem consulted by the user. The number of conversations and the idle time can determine the idle degree of the first customer service representative. Since the matching score is used to represent the matching degree between the customer service and the user account, and the idle score is used to represent the idle degree of the first customer service representative, the recommended score determined by the server based on the matching score and the idle score can accurately represent the effectiveness of the first customer service representative in handling the target conversation.

[0085] Combined Figure 3 , as Figure 4 shown, the above-mentioned server determines the recommended score of the first customer service representative based on the matching score of the first customer service representative and the idle score of the first customer service representative, which may specifically include S1023a.

[0086] S1023a. The server determines the recommended score of the first customer service representative based on the first weight, the matching score of the first customer service representative, the second weight, and the idle score of the second customer service representative.

[0087] Among them, the first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommended score.

[0088] It should be understood that the first weight is used to represent the proportion of the matching score of a customer service representative, and the second weight is used to represent the weight of the idle score of a customer service representative.

[0089] In an alternative implementation, the server may determine the sum of the first product (i.e., the product of the matching score of the first customer service representative and the first weight) and the second product (i.e., the product of the idle score of the first customer service representative and the second weight) as the recommended score of the first customer service representative.

[0090] In the embodiments of the present invention, since the server determines the recommended score of the first customer service representative according to the first weight, the matching score of the first customer service representative, the second weight, and the idle score of the second customer service representative, the accuracy of determining the recommended score can be improved.

[0091] CombinedFigure 2 , as Figure 5 shown, this session processing method further includes S104 - S105.

[0092] S104. When the number of sessions of each customer service agent is equal to the session number threshold, the server obtains the sorting of at least one pending session and the problem types of at least one pending session.

[0093] It can be understood that when the number of sessions of each customer service agent is equal to the session number threshold, it means that all customer service agents are unable to handle the target session at this time. At this time, the server can add the target session to the queue.

[0094] It should be understood that the queue includes at least one pending session that is queuing.

[0095] It can be understood that the problem type of each pending session is the problem type of the problems included in the historical session record of the pending session.

[0096] In the embodiments of the present application, the server can divide the problems into multiple problem types and then determine the priority corresponding to each problem type.

[0097] In an alternative implementation, the server can determine the serial numbers of the at least one pending session based on the chronological order of the session processing requests corresponding to each pending session.

[0098] S105. When the priority of the target problem type is higher than the priority threshold, the server adjusts the serial number of the target session to the target position.

[0099] Wherein, the target position is: before the serial number of the problem type with a priority lower than the first priority among the problem types of the at least one pending session, the first priority is the priority of the target problem type, and the target problem type is the problem type included in the historical session record.

[0100] It should be understood that when the priority of the target problem type is higher than the priority threshold, it means that the priority of the target problem type is relatively high. The server can determine that the target session needs to be processed preferentially. At this time, the server can move the queuing serial number of the target session forward so as to allocate a human customer service agent to the target session faster.

[0101] It should be understood that when the priority of a certain problem type (such as the first problem type) is higher than that of the target problem type, it means that the priority of the first problem type is relatively low. At this time, the server can adjust the serial number of the target session to before the pending session corresponding to the first problem type.

[0102] It can be understood that when the priority of the target question type is lower than the priority threshold, it indicates that the priority of the target question type is relatively low. At this time, the server can add the target session to the queuing queue based on the time of the session processing request corresponding to the target session.

[0103] In another alternative implementation, the server can also determine the sequence number of each user account in the queuing queue based on the queuing score of each user account in the queuing queue.

[0104] Optionally, the server determines that the queuing score of the user account satisfies Formula 2:

[0105] Formula 2:

[0106]

[0107] Where Sk represents the queuing score of user account k, a represents the scaling parameter of the queuing duration, z represents the queuing duration of user account k, e z represents the weight of the queuing duration, n represents the number of identical tags in the tags of user account k and the tags of the customer service, Ai represents the number of tag i in the tags included in user account k, Ak represents the total number of tags of user account k, Mi represents the number of user accounts including tag i in the queuing queue, and N represents the total number of tag types of users included in the queuing queue.

[0108] Exemplarily, assume that the tags included in the database include A, B, C, D, E, F, G, and the queuing queue includes user account one, user account two, and user account three. The tags included in user account one are A, B, C, D, E, where the number of tag A is 10, the numbers of tags B and C are 20 respectively, and the numbers of tags C, D, and E are 50 respectively, and the queuing duration is 50s; the tags included in user account two are A, B, C, F, where the number of tag A is 5, the number of tag B is 20, and the numbers of tags C and F are 15 respectively, and the queuing duration is 30s; the tags included in user account three are A, C, D, F, G, where the number of tag A is 15, and the numbers of tags C, D, F, and G are 18 respectively, and the queuing duration is 13s, and the tags of customer service 1 are A, B, C, M, N, H.

[0109] It can be determined that the same tags between the tags of the user account 1 and those of the customer service are A, B, and C. The number of tag A in the tags included in the user account 1 is 10, the number of tag B in the tags included in the user account 1 is 20, the number of tag C in the tags included in the user account 1 is 20, the total number of tags of the user account 1 is 200, the number of user accounts including tag A in the queuing queue is 3, the number of user accounts including tag B in the queuing queue is 2, the number of user accounts including tag C in the queuing queue is 3, and the types of tags of the users included in the queuing queue are A, B, C, D, E, F, and G.

[0110] In the embodiment of the present application, when all customer services are not idle to process the target session, the server adds the target session to the queuing queue, and then based on the priority of the problem type included in the historical session record of the target session, when the priority of the target problem type is relatively high, the server adjusts the target session to before the sessions to be processed with a priority lower than that of the target problem type, which can improve the processing efficiency of the target session.

[0111] Combined Figure 2 , as Figure 6 shown, before the server obtains the historical session record, the session processing method further includes S106 - S109.

[0112] S106. The server obtains the first problem sent by the user account.

[0113] It should be understood that after creating the target session, the user account can send a consulting question in the target session.

[0114] S107. The server determines the tags of the user account based on the identifier of the user account.

[0115] It should be understood that the tags of the user account are used to represent the keywords (or problem types) of the questions that the user account often asks during the historical time period.

[0116] Optionally, the tags of one user account can also be the role of the user account.

[0117] Exemplarily, the role of the user account can be an intelligent driving engineer, a business hall operator, etc.

[0118] In the embodiment of the present application, a user portrait library can be stored in the server. The user portrait library includes user portraits corresponding to each user account among multiple user accounts, and the user portrait of one user account can be the tags of each user account.

[0119] Optionally, the user portrait of one user account can also include the classification of the user account and the account information of each user account.

[0120] In an alternative implementation, the server may extract keywords based on the user information of the user account and the historical session record of the user account, and generate a user profile of the user account.

[0121] S108. The server determines at least one reply message from the database based on the label of the user account and the first question.

[0122] It should be understood that the database includes reply information corresponding to each of the multiple questions and labels corresponding to each question. The server may determine the answer corresponding to the question or the question including the keyword based on one question or the keyword of one question.

[0123] In the embodiment of the present application, the server may search the database for the answers corresponding to the label of the user account and the first question respectively.

[0124] In an alternative implementation, the server may extract the keywords included in the first question, and then search the database for the reply information corresponding to the keywords.

[0125] S109. The server sends at least one reply message to the user account.

[0126] Specifically, the server sends the at least one reply message to the terminal corresponding to the user account, so that the terminal can display the at least one reply message in the target session.

[0127] In an alternative implementation, a reply message may include both the keywords of the first question and the label of the user account. The server may determine the matching degree between each reply message and the label of the user account and the first question, and then determine N reply messages with the highest matching degrees, and send the N reply messages to the user account.

[0128] Optionally, the server may determine the matching degree of each reply message based on the number of keywords of the first question included in a reply message and the number of labels of the user account.

[0129] Optionally, the server may generate a page with the reply messages other than the N reply messages from the at least one reply message, and then send a link to the page to the user account, so that the user account can independently select to view the page.

[0130] It can be understood that the server may determine the first question and the at least one reply message as the historical session record.

[0131] In an alternative implementation, the server may update the tags and classification of the user account in the user profile library based on the first question sent by the user account this time.

[0132] In the embodiments of the present application, after the server obtains the first question sent by the user account, it first determines the tags of the first user account, and then determines the reply information from the database based on the tags of the user account and the first question, which can more accurately determine a reply with a higher matching degree with the user account, and can improve the effectiveness of the intelligent customer service in answering questions for the user account.

[0133] As Figure 7 shown, in an implementation manner of the embodiments of the present application, the server executing the embodiments of the present application may include a consulting user module, an intelligent customer service module, an artificial customer service module, a user profile library, and a database.

[0134] The consulting user module can obtain the source of the user and the question consulted by the user, and then classify the question consulted by the user; and can update the user profile of the user account in the user profile library based on the question consulted by the user account this time.

[0135] The intelligent customer service module is used to recommend questions to the user account and retrieve answers in the knowledge base according to the question consulted by the user account; the intelligent customer service module can also transfer the target session to the artificial customer service module in response to the session processing request of the user account.

[0136] The artificial customer service module stores the business types of each customer service, is used to detect online artificial customer services, group the customer services, and record the service records of each artificial customer service. The artificial customer service module can obtain the basic allocation rules and consultation priorities from the database to determine the customer service assigned to each user account and the queuing sequence number.

[0137] The user profile library stores the user information, user tags of multiple user accounts, and the historical questions consulted by each user account.

[0138] The database includes a knowledge base, the consultation priorities of questions, basic allocation rules, and question classifications.

[0139] In an implementation manner of the embodiments of the present application, the server may include a session allocation module and a queuing module. The session allocation module is used to allocate artificial customer services to user accounts, and the queuing module is used to determine the sequence number of the user account in the queuing sequence. As Figure 8 shown, the session processing method includes S201 - S207.

[0140] S201. Transfer to artificial.

[0141] Specifically, in response to a session processing request of a user account, the server may transfer the target session to a human operator.

[0142] S202. Session allocation module.

[0143] Specifically, based on the session allocation module, a human customer service is allocated to the user account.

[0144] When a customer service agent is connected, S203 is executed.

[0145] S203. Customer service agent connects to the session.

[0146] Specifically, the customer service agent with the highest recommended score is connected to the target session.

[0147] When no customer service agent is connected, S204 is executed.

[0148] S204. Queuing module.

[0149] Specifically, based on the queuing module, the serial number of the user account in the queuing sequence is determined.

[0150] S205. Customer service agent actively connects to the session.

[0151] Specifically, the customer service agent can actively view and connect to the target session.

[0152] After S203 and S205, S206 is executed.

[0153] S206. User problem handling.

[0154] Specifically, the customer service agent can answer the questions consulted by the user account.

[0155] S207. User service evaluation.

[0156] Specifically, after the target session ends, the server can obtain the evaluation of the user account.

[0157] The embodiments of the present application can divide functional modules of a server or the like according to the above method examples. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. It should be noted that the division of modules in the embodiments of the present application is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0158] In the case of dividing each functional module corresponding to each function, Figure 9 shows a possible structural schematic diagram of the session processing device involved in the above embodiments, as Figure 9As shown in the figure, the session processing device 20 may include: an acquisition module 201, a determination module 202, and a processing module 203.

[0159] The acquisition module 201 is configured to obtain the historical session record in the target session in response to a session processing request of a user account, where the session processing request is used to request a human customer service to process the target session, the target session is the session currently being carried out by the user account, and the historical session record includes the questions consulted by the user account.

[0160] The determination module 202 is configured to determine the recommended score of each customer service based on the historical session record, the label of each customer service among multiple customer services, and the session status of each customer service, where the recommended score of a customer service is used to characterize the effectiveness of the customer service in processing the target session, and the label of a customer service is used to characterize the problem type corresponding to the customer service.

[0161] The determination module 202 is further configured to determine the customer service with the highest recommended score as the target customer service.

[0162] The processing module 203 is configured to connect the target customer service to the target session so that the target customer service processes the target session.

[0163] Optionally, the session status of a customer service includes the current number of sessions and the idle time.

[0164] The determination module 202 is specifically configured to determine the matching score of the first customer service based on the historical session record and the label of the first customer service, where the matching score of the first customer service is used to characterize the matching degree between the first customer service and the user account.

[0165] The determination module 202 is further specifically configured to determine the idle score of the first customer service based on the number of sessions of the first customer service and the idle time of the first customer service, where the idle score of the first customer service is used to characterize the idle degree of the first customer service.

[0166] Optionally, the determination module 202 is further specifically configured to determine the recommended score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service.

[0167] Optionally, the determination module 202 is further specifically configured to determine the recommended score of the first customer service based on a first weight, the matching score of the first customer service, a second weight, and the idle score of the second customer service, where the first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommended score.

[0168] The acquisition module 201 is further configured to obtain the sorting of at least one pending session and the problem types of the at least one pending session when the number of sessions of each customer service is equal to the session number threshold.

[0169] The processing module 203 is further configured to, when the priority of the target problem type is higher than the priority threshold, adjust the sequence number of the target session to a target position, where the target position is: before the sequence number where the problem type with a priority lower than the first priority among the problem types of the at least one session to be processed is located, and the first priority is the priority of the target problem type, and the target problem type is the problem type included in the historical session record.

[0170] Optionally, the session processing device further includes a sending module 204.

[0171] The obtaining module 201 is further configured to obtain a first problem sent by the user account.

[0172] The determining module 202 is further configured to determine a label of the user account based on the identifier of the user account.

[0173] The determining module 202 is further configured to determine at least one reply message from the database based on the label of the user account and the first problem.

[0174] The sending module 204 is configured to send the at least one reply message to the user account.

[0175] In the case of adopting an integrated unit, Figure 10 FIG. shows a possible structural schematic diagram of the session processing device involved in the above embodiments. As Figure 10 shown, the session processing device 30 may include: a processing module 301 and a communication module 302. The processing module 301 may be configured to control and manage the actions of the session processing device 30. The communication module 302 may be configured to support the communication of the session processing device 30 with other entities. Optionally, as Figure 10 shown, the session processing device 30 may further include a storage module 303 for storing the program code and data of the session processing device 30.

[0176] Among them, the processing module 301 may be a processor or a controller. The communication module 302 may be a transceiver, a transceiver circuit or a communication interface, etc. The storage module 303 may be a memory.

[0177] Among them, when the processing module 301 is a processor, the communication module 302 is a transceiver, and the storage module 303 is a memory, the processor, the transceiver and the memory may be connected through a bus. The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc.

[0178] It should be understood that in various embodiments of the present application, the magnitude of the serial numbers of the above processes does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0179] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0180] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0181] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more integrated media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a Solid State Disk (SSD)), etc.

[0183] As described above, the above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A session processing method, characterized in that, including: In response to a session processing request of a user account, obtain historical session records in a target session, where the session processing request is used to request an artificial customer service to process the target session, the target session is the session currently being conducted by the user account, and the historical session records include questions consulted by the user account; Based on the historical session records, labels of each of multiple customer services, and the session status of each of the customer services, determine a recommended score for each of the customer services, where a recommended score of a customer service is used to represent the effectiveness of the customer service in processing the target session, and a label of a customer service is used to represent the type of question corresponding to the customer service; Determine the target customer service as the customer service with the highest recommended score, and connect the target customer service to the target session so that the target customer service processes the target session.

2. The session processing method according to claim 1, wherein The session status of a customer service includes the current number of sessions and the idle time. Determining the recommended score of the first customer service based on the historical session records, the label of the first customer service, and the session status of the first customer service includes: Based on the historical session records and the label of the first customer service, determine a matching score of the first customer service, where the matching score of the first customer service is used to represent the matching degree between the first customer service and the user account; Based on the number of sessions of the first customer service and the idle time of the first customer service, determine an idle score of the first customer service, where the idle score of the first customer service is used to represent the idle degree of the first customer service; Determine the recommended score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service.

3. The session processing method according to claim 2, wherein The determining the recommended score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service includes: Based on a first weight, the matching score of the first customer service, a second weight, and the idle score of the second customer service, determine the recommended score of the first customer service, where the first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommended score.

4. The session processing method according to claim 1, characterized in that The method further includes: When the number of sessions of each of the customer services is equal to a session number threshold, obtain the sorting of at least one session to be processed, and the types of questions of the at least one session to be processed; When the priority of the target question type is higher than a priority threshold, adjust the serial number of the target session to a target position, where the target position is: before the serial number where the question type with a priority lower than the first priority among the question types of the at least one session to be processed is located, the first priority is the priority of the target question type, and the target question type is the type of question included in the historical session records.

5. The session processing method according to any one of claims 1-4, characterized in that, The method further includes: Obtain a first question sent by the user account; Based on the identifier of the user account, determine the label of the user account; Based on the label of the user account and the first question, determine at least one reply message from a database; Send the at least one reply message to the user account.

6. A session processing device, characterized in that, including: an obtaining module, a determining module, and a processing module; The obtaining module is configured to obtain historical conversation records in a target conversation in response to a conversation processing request of a user account, where the conversation processing request is used to request an artificial customer service to process the target conversation, the target conversation is the conversation currently being carried out by the user account, and the historical conversation records include questions consulted by the user account; The determining module is configured to determine a recommended score for each customer service based on the historical conversation records, tags of each customer service among multiple customer services, and the conversation status of each customer service, where a recommended score of a customer service is used to characterize the effectiveness of the customer service in processing the target conversation, and a tag of a customer service is used to characterize the type of question corresponding to the customer service; The determining module is further configured to determine the customer service with the highest recommended score as the target customer service; The processing module is configured to connect the target customer service to the target conversation so that the target customer service processes the target conversation.

7. The session processing device according to claim 6, wherein The conversation status of a customer service includes the current number of conversations and the idle time; The determining module is specifically configured to determine a matching score of a first customer service based on the historical conversation records and the tag of the first customer service, where the matching score of the first customer service is used to characterize the matching degree between the first customer service and the user account; The determining module is further specifically configured to determine an idle score of the first customer service based on the number of conversations of the first customer service and the idle time of the first customer service, where the idle score of the first customer service is used to characterize the idle degree of the first customer service; The determining module is further specifically configured to determine the recommended score of the first customer service based on the matching score of the first customer service and the idle score of the first customer service.

8. The conversation processing device according to claim 7, wherein The determining module is further specifically configured to determine the recommended score of the first customer service based on a first weight, the matching score of the first customer service, a second weight, and the idle score of the second customer service, where the first weight is the weight corresponding to the matching score, and the second weight is the weight corresponding to the recommended score.

9. The conversation processing device according to claim 6, wherein The obtaining module is further configured to obtain a sorting of at least one pending conversation and the question types of the at least one pending conversation when the number of conversations of each customer service is equal to a conversation number threshold; The processing module is further configured to adjust the sequence number of the target conversation to a target position when the priority of the target question type is higher than a priority threshold, where the target position is: before the sequence number where the question type with a priority lower than a first priority among the question types of the at least one pending conversation is located, the first priority is the priority of the target question type, and the target question type is the question type included in the historical conversation records.

10. The session processing device according to any one of claims 6-9, characterized in that, The conversation processing device further includes a sending module; The obtaining module is further configured to obtain a first question sent by the user account; The determining module is further configured to determine a tag of the user account based on the identifier of the user account; The determining module is further configured to determine at least one reply message from a database based on the tag of the user account and the first question; The sending module is configured to send the at least one reply message to the user account.

11. A server, characterized in that, The server includes: a processor; a memory configured to store executable instructions of the processor; wherein the processor is configured to execute the instructions to implement the session processing method according to any one of claims 1-5.

12. A computer-readable storage medium having instructions stored thereon, characterized in that, When the instructions in the computer-readable storage medium are executed by the server, the server is enabled to execute the session processing method according to any one of claims 1-5.

Images