Data processing method, medium and equipment for online customer service system

Abstract

A data processing method, medium and device for an online customer service system, the method comprising: in response to receiving an access request of a user to an online customer service system, allocating a manual customer service to the user; creating a session including the manual customer service, the user and a robot customer service, and setting the session to be in a hosting state, so as to respond to a message input by the user in the session through the robot customer service; and in response to a received takeover request of the manual customer service, switching the session from the hosting state to a takeover state so as to respond to a message input by the user in the session through the manual customer service.

Description

Technical Field

[0001] This application relates to the technical field of online customer service, and particularly to a data processing method, medium, and device for an online customer service system. Background Art

[0002] With the rapid development and popularization of Internet technology, the customer service mode of enterprises has gradually shifted from traditional offline services to online services. As an important bridge for enterprises to communicate with customers, the importance of the online customer service system has become increasingly prominent. The online customer service system can not only provide enterprises with a more efficient and convenient service method, but also reduce operating costs and improve customer satisfaction. Robot customer service has the advantages of fast response speed, 24-hour uninterrupted service, and handling a large number of common problems, and has gradually become an important part of the online customer service system. Although robot customer service performs well in handling common problems, for some complex problems, human customer service still has irreplaceable advantages. In related technologies, generally, robot customer service is first used to provide services to users, and when the robot customer service cannot solve the problem, it is then transferred to human customer service. However, during the period of waiting to transfer to human, the user's needs cannot be responded to in a timely manner, resulting in low customer satisfaction. Summary of the Invention

[0003] In a first aspect, an embodiment of this application provides a data processing method for an online customer service system, the method including: in response to receiving an access request from a user to the online customer service system, allocating a human customer service to the user; creating a session including the human customer service, the user, and a robot customer service, and setting the session to a managed state to respond to messages input by the user in the session through the robot customer service; in response to receiving a takeover request from the human customer service, switching the session from the managed state to a takeover state to respond to messages input by the user in the session through the human customer service.

[0004] In a second aspect, an embodiment of this application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in any embodiment of this application is implemented.

[0005] In a third aspect, an embodiment of this application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the method described in any embodiment of this application is implemented.

[0006] In a fourth aspect, an embodiment of this application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method described in any embodiment of this application is implemented.

[0007] In the embodiments of the present application, when an access request from a user to an online customer service system is received, a human customer service is directly assigned to the user, a three-party session including the human customer service, the user, and a robot customer service is created, and the session is set to a managed state to respond to the messages input by the user in the session through the robot customer service. Since the response speed of the robot customer service is relatively fast, the user's needs can be efficiently responded to after the user accesses the online customer service system. In addition, since a human customer service is directly assigned to the user when the user accesses, when the robot customer service cannot solve the user's problem, the user does not need to wait for the process of transferring from the robot customer service to the human customer service, but can directly be served by the human customer service, thereby improving the response efficiency of the human customer service to the user's needs when the robot customer service cannot solve the user's problem. Moreover, the session created before the human customer service starts to take over the session includes the three parties of the human customer service, the user, and the robot customer service. In this way, when it is necessary to take over the session, the human customer service can directly and quickly understand the user's needs and the response situation of the robot customer service through the session, instead of having to understand the user's needs and the response situation of the robot customer service after being successfully assigned to the user, which is conducive to the human customer service quickly taking over the session and realizing seamless switching between the robot customer service and the human customer service. In summary, the embodiments of the present application can improve the response efficiency of the online customer service system to the user, reduce the waiting time of the user during the service process, and thus improve the user satisfaction.

[0008] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings here are incorporated into the specification and constitute a part of the present application. These drawings show embodiments consistent with the present application and are used together with the specification to explain the technical solutions of the present application.

[0010] Figure 1 is a flowchart of the working process of an online service system in the related art.

[0011] Figure 2A and Figure 2B is a schematic diagram of an application scenario of the embodiments of the present application.

[0012] Figure 3 is a flowchart of a data processing method for an online customer service system according to the embodiments of the present application.

[0013] Figure 4A is a schematic diagram of a chat interface in the managed state according to the embodiments of the present application.

[0014] Figure 4B is a schematic diagram of a chat interface in the takeover state according to the embodiments of the present application.

[0015] Figure 5 It is a schematic diagram of specific messages and their tags in an embodiment of the present application.

[0016] Figure 6 It is a schematic diagram of summary information in an embodiment of the present application.

[0017] Figure 7 It is a schematic diagram of a session list in an embodiment of the present application.

[0018] Figure 8 It is a schematic diagram of a chat interface after re-hosting in an embodiment of the present application.

[0019] Figure 9 It is a general flowchart of an embodiment of the present application.

[0020] Figure 10 It is a block diagram of a data processing device for an online customer service system in an embodiment of the present application.

[0021] Figure 11 It is a schematic diagram of a computer device in an embodiment of the present application. Detailed Description of the Invention

[0022] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0023] The terms used in the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items. Additionally, the term "at least one" as used herein represents any one of a plurality or any combination of at least two of a plurality.

[0024] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0025] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present application and to make the above-mentioned objects, features, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0026] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties. And the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0027] Figure 1 The working flowchart of an online service system in the related art is shown, mainly including the following steps:

[0028] Step S102, the user accesses the online customer service system. After the user accesses, a session can be established between the robot customer service and the user, and the user can input messages through this session, such as voice messages or text messages.

[0029] Step S104, the robot customer service provides services to the user. The robot customer service can determine the user's needs according to the message input by the user, such as complaints, refunds, querying the logistics status, seeking technical support, etc., and respond to the message input by the user.

[0030] Step S106, during the service process of the robot customer service, the online customer service system can detect in real time whether it is necessary to switch to a human customer service. If so, step S108 is executed; otherwise, it jumps to step S116. For example, when the robot customer service fails to solve the user's problem, recognizes that the user has dissatisfaction, or when the user requests the human customer service to provide services, it can be determined that it is necessary to switch to the human customer service.

[0031] Step S108, assign a human customer service to the user.

[0032] Step S110, transfer to the human customer service. After the transfer is successful, a session can be established between the human customer service and the user, and the chat record between the robot customer service and the user can be sent to the human customer service through this session, so that the human customer service can quickly understand the user's needs that have not been met by the robot customer service.

[0033] Step S112, the human customer service provides services to the user. The human customer service can respond to the user's needs according to the obtained chat record.

[0034] Step S114, determine whether the service of the human customer service has ended. If the user's problem has been solved, the service of the human customer service ends, and the current online customer service process ends. Otherwise, return to Step S112 and continue to provide service to the user by the human customer service.

[0035] Step S116, determine whether the service of the robot customer service has ended. If the user's problem has been solved, the service of the robot customer service ends, and the current online customer service process ends. Otherwise, return to Step S104 and continue to provide service to the user by the robot customer service.

[0036] Figure 2A and Figure 2B shows a schematic diagram of an actual application scenario. As Figure 2A shown, after the user accesses the online customer service system, a session can be established between the user and the robot customer service (hereinafter referred to as Session 1). The user can input messages in Session 1 to express their own needs, such as searching for an order. The robot customer service can respond to the user by replying to the message. Through Figure 2A it can be seen that the robot customer service fails to meet the user's needs. Therefore, the user can switch to the human customer service through the "transfer to human customer service" function on the interface. As Figure 2B shown, after the switch is successful, a session can be established between the user and the human customer service (hereinafter referred to as Session 2), and the chat record between the user and the robot customer service in Session 1 can be sent to Session 2. The human customer service can continue to provide service to the user based on the chat record received in Session 2.

[0037] It can be seen from the above embodiments that in the related art, the robot customer service is first used to provide service to the user, and when the robot customer service cannot solve the problem, it is transferred to the human customer service. In the related art, usually an upper limit on the number of users that can access the human customer service is set. After the number of users accessing the human customer service reaches the above upper limit, other users need to wait for the service of the users already accessing the human customer service to be completed before they can continue to access the human customer service. For example, assume that the upper limit on the number of users that can access each human customer service is 3. Then, when a fourth user also needs to access the human customer service M, the fourth user needs to wait for the human customer service M to complete the service for one of the first three users before accessing the human customer service M. That is, in the above method when transferring to the human customer service, users often have to wait for an uncertain period of time, and the user's needs cannot be responded to in a timely manner during the waiting process, resulting in a low user satisfaction.

[0038] Based on this, the embodiments of the present application provide a data processing method for an online customer service system. Refer to Figure 3 , the method includes:

[0039] Step S202: In response to receiving an access request from a user to the online customer service system, assign a human customer service agent to the user;

[0040] Step S204: Create a session including the human customer service agent, the user, and the robot customer service agent, and set the session to a managed state to respond to the messages input by the user in the session through the robot customer service agent;

[0041] Step S206: In response to receiving a takeover request from the human customer service agent, switch the session from the managed state to the takeover state to respond to the messages input by the user in the session through the human customer service agent.

[0042] The implementation details of the embodiments of the present application will be illustrated by way of example below.

[0043] In step S202, in response to receiving an access request from a user to the online customer service system, assign a human customer service agent to the user. That is, each user is assigned a human customer service agent when accessing the online customer service system.

[0044] The online customer service system refers to a system that provides customer service through the Internet, including various forms such as text and voice. The online customer service system includes a user side, a human customer service side, and a service side. The user side includes a chat interface through which the user can input and send messages, and can also display the reply messages received from the robot customer service agent or the human customer service agent in real time. The service side can deploy various algorithms and models to implement the functions of the robot customer service agent. These algorithms and models include natural language processing (NLP) algorithms, machine learning models, deep learning models, etc., for understanding user problems, generating reply content, and performing intelligent recommendations, etc. When the user side sends a message, the service side will receive these messages and perform corresponding processing according to the content and type of the messages. For simple and common problems, the service side will directly generate a reply through the robot customer service agent and send it to the user side; for complex or problems that require human intervention, the service side will forward the messages to the human customer service side. The human customer service side also provides a chat interface through which the human customer service agent can communicate and interact with the user in real time.

[0045] In some embodiments, the user side may include a functional component for accessing the online customer service system. By operating this functional component, an access request to the online customer service system can be generated. In response to receiving the above access request, a human customer service agent can be assigned to the user.

[0046] The number of human customer service agents can be greater than or equal to 1. The method of assigning a human customer service agent to the user in the case where the number of human customer service agents is greater than 1 will be illustrated by way of example below.

[0047] Optionally, a human customer service can be randomly selected from multiple human customer services and assigned to the user. This method is simple and efficient and is applicable to scenarios where the user's needs are relatively simple and the skill differences among customer services are not significant.

[0048] Optionally, the human customer service that has served the user can be assigned to the user. If a human customer service has served a certain user, then compared with other human customer services that have not served the user, this human customer service may have a better understanding of the user's characteristics and needs, so that it can enter the state faster and provide more efficient services for the user. In addition, in some scenarios that require continuous services, through this assignment method, the trouble of the user repeating the description of problems and needs can be reduced, and the user satisfaction can be improved.

[0049] Optionally, the human customer service with the highest matching degree with the user's user information among multiple human customer services can be assigned to the user. Among them, the user information includes but is not limited to user level (such as VIP user, non-VIP user), consultation question category, historical consultation information, and / or user preferences, etc. Specifically, a customer service information database can be maintained to record customer service information such as the professional field, historical service records, and skill level of each human customer service. The matching degree between the customer service information and the user information can be determined through a matching algorithm, and a human customer service can be assigned to the user according to the above matching degree. Taking the user information including the user level and the customer service information including the skill level as an example, a human customer service with a higher skill level can be assigned to a user with a higher user level, so as to ensure the service quality for users with a higher user level. Taking the user information including the consultation question category and the customer service information including the professional field as an example, the human customer service whose professional field matches the consultation question category can be assigned to the user, so that the human customer service can better understand and solve the problems consulted by the user and provide professional services for the user.

[0050] Optionally, the human customer service with the fewest number of taken-over sessions among multiple human customer services can be assigned to the user. In the embodiments of the present application, after the user is assigned to a human customer service, the session of this user may be in a trusteeship state or a takeover state. The trusteeship state means that the human customer service entrusts the session to the robot customer service and provides services for the user through the robot customer service. The takeover state means that the human customer service takes over the robot customer service to provide services for the user. When the session is in the trusteeship state, the one actually providing services for the user is not the human customer service but the robot customer service. Therefore, this session does not consume the energy of the human customer service. When the session is in the takeover state, the human customer service needs to provide actual services for the user. Therefore, this session will consume the energy of the human customer service. Therefore, assigning the human customer service with the fewest number of taken-over sessions to the user can reasonably allocate the energy of each human customer service, reduce the situation that the same human customer service serves too many users at the same time and cannot respond to the user's needs in time, thereby improving the service quality and user satisfaction.

[0051] Optionally, multiple human customer service representatives can be polled, and the polled human customer service representative can be assigned to the user. In this embodiment, each human customer service representative has the opportunity to be assigned to the user, avoiding the situation where some human customer service representatives are overly busy while others are idle, and improving the fairness of the assignment. In addition, polling assignment can ensure a relatively balanced workload of human customer service representatives and improve the stability of the online customer service system.

[0052] In addition to the above-listed assignment methods, other methods can also be used to assign human customer service representatives to users, or a combination of at least two of the above methods can be used to assign human customer service representatives to users. This application will not list them one by one.

[0053] In step S204, a session including a human customer service representative, a user, and a robot customer service representative can be created, and the session can be set to a managed state. When the session is in the managed state, the robot customer service representative can respond to the messages input by the user in the session, and there is no need for the human customer service representative to respond to the messages input by the user. In the managed state, the session messages of the session can include the messages input by the user and the first response messages obtained by the robot customer service representative's response to the messages input by the user, and the above session messages can be displayed in the chat interface of the user side.

[0054] Since when a human customer service representative is successfully assigned to the user at the beginning, the session of this user is in the managed state and does not consume the energy of the human customer service representative, therefore, in this application, there is no need to set an upper limit on the number of users that each human customer service representative can access. As long as it is determined through the assignment algorithm that a certain human customer service representative is assigned to the user, a session including this user, the robot customer service representative, and the corresponding human customer service representative is created. In this way, the user does not need to wait for the user service of the previously accessed human customer service representative to be completed before accessing the human customer service representative, reducing the waiting time of the user. For example, when the human customer service representative M has accessed 3 users, the number of sessions including this human customer service representative M is 3. After the fourth user accesses the online customer service system, a session including the fourth user, the robot customer service representative, and the human customer service representative M can continue to be created, so that the number of sessions including this human customer service representative M increases to 4. In addition, although it is the robot customer service representative that provides services to the user at this time, the above session messages can still be displayed in the chat interface of the human customer service representative. In this way, it is convenient for the human customer service representative to quickly browse the chat records between the robot customer service representative and the user when needed, so as to quickly understand the user's needs and monitor the service quality of the robot customer service representative. Figure 4AThe figure shows a schematic diagram of the chat interface of a human customer service. After a user accesses the online customer service system, Session 1 is created. Among them, Session 1 includes a human customer service, a user, and a robot customer service. However, at this time, since Session 1 is in a managed state, the human customer service can choose not to participate in the service first, but let the robot customer service serve the user. Therefore, the chat record only includes the messages input by the user and the messages replied by the robot customer service, and does not include the messages replied by the human customer service.

[0055] It should be noted that the operation of allocating a human customer service to the user in step S202 can be executed first, and after the allocation is successful, a session including a human customer service, a user, and a robot customer service can be created. Or, the operation of allocating a human customer service to the user in step S202 and the operation of creating a session including a human customer service, a user, and a robot customer service in step S204 can also be executed in parallel. When creating a session, a session including the user and the robot customer service can be created first, and after successfully allocating a human customer service to the user, the human customer service can be added to the session. In the second method, if a human customer service has not been successfully allocated to the user when a session including the user and the robot customer service is successfully created, the operation of allocating a human customer service to the user can be executed while the robot customer service responds to the messages input by the user in the session. In this way, there is no need to wait for the successful allocation of the human customer service before the robot customer service responds to the messages input by the user, which improves the response speed of the robot customer service and further improves the response efficiency of the online customer service system to the user.

[0056] In some embodiments, when the session is in a managed state, specific messages in the session messages can be marked. When the session is in a managed state, the online customer service system is responsible for managing and maintaining the relevant data and status of the session. At this time, the robot customer service is communicating with the user, providing preliminary services and answers. The online customer service system will automatically save all the message records in the session, including the messages sent by the user, the replies of the robot customer service, etc., and record the status information of the session, such as the start time of the session, the current stage, the duration of the session, etc., and mark specific messages in the session messages according to preset rules. Among them, specific messages include but are not limited to at least one of the following:

[0057] Information related to the user's needs; for example, the needs, questions, or expectations clearly put forward by the user. See Figure 5 , the user sends a message "I can't find my order". The above message indicates that the user's need is to find the order, and the online customer service system will mark this message as information related to the user's needs so that the human customer service can quickly understand the user's core needs after taking over the conversation.

[0058] Information related to the transaction number; where the transaction number can be an order number, customer number, product number, work order number, express waybill number, etc., which are numbers or codes used to uniquely identify a service matter or service entity. The specific category of the transaction number specifically depends on the nature and requirements of the transaction. See Figure 5 , the user sends the message "123456789", and this number is the order number. The online customer service system will mark this message as information related to the order number so that the human customer service can quickly determine which order the user's needs are related to after taking over the conversation.

[0059] Information related to user satisfaction; for example, the user's satisfaction label, dissatisfaction, keywords, and satisfaction score. The user may use some specific keywords in the conversation to express their satisfaction or dissatisfaction, such as "thank you", "satisfied", "solved", "bad review", "complaint", etc. The system can identify and mark these keywords to help the human customer service quickly understand the user's emotional state after taking over the conversation and adopt corresponding service strategies to serve the user. In addition, the satisfaction label can be dynamically adjusted according to the user's emotional changes, for example, changing from "neutral" to "satisfied" or "dissatisfied". See Figure 5 , the user sends the message "Can you do it? If not, get someone else", indicating that the user has dissatisfaction and the user satisfaction has decreased. The above message can be marked so that the human customer service can quickly understand the user's emotion.

[0060] Information related to the processing progress of the robot customer service. For example, the answer status of the robot customer service to the user's question, processing steps, difficulties encountered, etc. See Figure 5 The robot customer service replies to the user "Not found yet. There may be an input error. Please check and try again". The online customer service system will mark this message as information related to the processing progress of the robot customer service so that the human customer service can understand the current processing situation of the robot customer service.

[0061] In Figure 5 the example shown, specific messages are marked with a specific background color. It can be understood that this marking method is only for illustrative purposes. In other examples, specific messages can also be marked by using bold font, increasing the font size, changing the character color, adding underlines, etc., so that specific messages are more prominent. In addition, different categories of specific messages can be marked with different marking methods to facilitate the human customer service to distinguish and quickly locate various categories of specific messages. For example, in Figure 5 the example shown, the information related to user satisfaction is highlighted with an orange background, and other categories of specific messages are highlighted with a yellow background.

[0062] In step S206, if a takeover request sent by a human customer service is received, the above session can be switched from the managed state to the takeover state. In some embodiments, an operation control can be included in the chat interface of the human customer service. The human customer service can send a takeover request by operating the operation control. Among them, when the session is in the managed state, the human customer service can monitor the service quality of the robot customer service through the chat interface, and actively send a takeover request when it is determined that the service quality does not meet the user's expectations. Or, when the session is in the managed state, the online customer service system can also automatically monitor the service quality of the robot customer service. In response to detecting that the service quality of the robot customer service does not meet the user's expectations, a prompt message can be output to prompt the human customer service to take over the session. The human customer service can send a takeover request after receiving the prompt message.

[0063] In the takeover state, the robot customer service no longer responds to the messages input by the user in the session, but the human customer service responds to the messages input by the user in the session to obtain a second response message. In the takeover state, the session messages of the session can include the messages input by the user, the first response message, and the second response message. The above session messages can be displayed on the chat interface of the client and the chat interface of the human customer service respectively. After the human customer service clicks the Figure 4A "I want to take over" button shown, session 1 can be switched to the takeover state. As Figure 4B shown, in the takeover state, the human customer service replies to the messages input by the user, and the chat interface of the human customer service can simultaneously display the messages sent by the user, the messages sent by the robot customer service, and the messages sent by the human customer service.

[0064] In some embodiments, in response to switching the session from the managed state to the takeover state, summary information of the session messages can also be displayed in the chat interface of the human customer service. The summary information is a brief summary of the chat process between the robot customer service and the user, aiming to help the human customer service quickly understand the key content and progress of the conversation. The summary information can include but is not limited to the following content:

[0065] Overview of user questions, used to briefly describe the questions or needs raised by the user.

[0066] Key points of the robot customer service's reply, used to summarize the content of the robot customer service's reply to the user's questions.

[0067] User's emotional state, used to label the user's emotional state according to the user's emotional expression in the conversation, such as "satisfied", "dissatisfied", "neutral", etc.

[0068] Key information extraction, used to extract key information in the conversation, such as order number, product model, fault description, etc.

[0069] Conversation progress, which is used to briefly describe the progress of the conversation, such as "The problem has not been solved", "The robot customer service has provided a solution", etc.

[0070] In the chat interface of the human customer service, the summary information can be displayed in a prominent manner, such as at the top or sidebar of the chat window. The summary information can be presented in a concise text form or visualized through icons, labels, etc. For example, the user's emotional state can be represented by emojis of different colors, and the key information can be displayed in the form of labels.

[0071] Through the above embodiments, when the human customer service takes over the conversation, the present application can help the human customer service quickly understand the chat process between the robot customer service and the user by displaying the summary information of the conversation messages, improving the service efficiency and quality.

[0072] See Figure 6 , the summary information is "The order with order number 123456789 cannot be found, the problem has not been solved, and the user's emotion has escalated". This summary information includes the problem overview (the order cannot be found), the order number (123456789), the progress of the conversation (the problem has not been solved), and the user's emotional state (the user's emotion has escalated).

[0073] In practical applications, the same human customer service may be assigned to one or more users. For each generated conversation including this human customer service, the conversation including this human customer service can be added to the conversation list of this human customer service, and the conversation list of this human customer service can be displayed in the chat interface of the human customer service. As Figure 7 shown, a certain human customer service is assigned to user U1, user U2, and user U3. Therefore, the conversation list of this human customer service includes three conversations, namely the conversation including this human customer service, user U1, and the robot customer service (i.e., conversation 1), the conversation including this human customer service, user U2, and the robot customer service, and the conversation including this human customer service, user U3, and the robot customer service. The conversation list including the above three conversations can be displayed in the chat interface of this human customer service, so as to facilitate the human customer service to comprehensively understand the situations of the users it serves. It can be understood that the illustration in the figure is only for exemplary purposes, and in practical applications, the conversation list may include only one conversation, only two conversations, or more than three conversations.

[0074] In some embodiments, the arrangement order of each conversation in the conversation list can be determined based on specific conditions. The above specific conditions include but are not limited to at least one of the following:

[0075] Creation time of the session; for example, in the session list, the session with an earlier creation time is placed before the session with a later creation time. Since the session is created when the user accesses the online customer service system, by placing the session with an earlier creation time in a more prominent position in the session list, it is convenient for the human customer service to distinguish the time when each user accesses the online customer service system, so as to be able to give priority to serving the users who access the online customer service system earlier, reduce the waiting time of the users, and improve the user experience.

[0076] Event categories involved in the session; for example, in the session list, the session involving specific event categories is placed before the session involving non-specific event categories. Among them, specific event categories may include, but are not limited to, event categories related to account security, event categories related to user complaints, and / or event categories related to system failures. In addition to the above-listed specific event categories, according to actual needs, other relatively important or urgent event categories can also be set as specific event categories. By placing the session involving specific event categories in a more prominent position in the session list, it is convenient for the human customer service to give priority to processing important and urgent sessions.

[0077] User satisfaction during the session. For example, if the user satisfaction in Session 1 is lower than that in Session 2, then in the session list, Session 1 is placed before Session 2. Among them, user satisfaction can be determined by recognizing the user's emotions. For example, an emotion recognition model can be used to score the user's emotions. The higher the score, the more excited the user's emotions, which means the lower the user satisfaction. Or, user satisfaction can be determined based on the user's rating of the service quality. The higher the user's rating of the service quality, the higher the user satisfaction. Or, user satisfaction can be determined based on the service duration of the robot customer service. Generally, users expect problems to be solved as soon as possible. If the robot customer service provides services to the user for a long time, it may mean that the problems raised by the user have not been solved for a long time, and at this time the user satisfaction may decrease. Therefore, the service duration of the robot customer service may be inversely related to user satisfaction.

[0078] In addition to the above-listed conditions, the sorting order of each session in the session list can also be determined based on other conditions, or based on multiple conditions jointly. This is not listed one by one here.

[0079] In the related art, after switching from a robot customer service to a human customer service, usually the human customer service provides services to the user until the service ends. However, after the human customer service has handled complex problems, the user may raise some simple problems that could have been solved by the robot customer service. These simple problems occupy the resources of the human customer service, resulting in a waste of resources. To solve the above problems, in the case where the session is in the takeover state, in response to receiving a hosting request from the human customer service, the session can be switched back from the takeover state to the hosting state. As Figure 8 shown, an operation control (such as "Re-host" in the figure) may be included on the chat interface of the human customer service. The human customer service can click on this operation control to re-host the session. After hosting, the robot customer service can replace the human customer service to provide services to the user and respond to the messages sent by the user. For example, in Figure 8 the example shown, when the user consults "How to cook the ingredients in the order", the human customer service can click "Re-host" to host the session to the robot customer service. The robot customer service can reply with the message "Add water to the pot, put in the ingredients, bring to a boil over high heat, and then turn to low heat and continue to cook for 30 minutes."

[0080] In some embodiments, in the case where the session is in the takeover state, if the human customer service identifies that the message input by the user can be responded to by the robot customer service, a hosting request can be actively sent to re-host the session to the robot customer service. Or, in the case where the session is in the takeover state, the online customer service system can detect whether the robot customer service can respond to the message input by the user in the session. If so, a prompt message can be output to prompt the human customer service to host the session. After receiving the prompt message, the human customer service can send a hosting request to re-host the session to the robot customer service.

[0081] In some embodiments, status indication information can also be displayed in the chat interface of the human customer service to indicate whether the current state of the session is the hosting state or the takeover state.

[0082] Figure 9 shows the overall process of an embodiment of the present application, including the following steps:

[0083] Step S302, user access: The user accesses the online customer service system and initiates a consultation.

[0084] Step S304, the robot customer service initially responds to the user and assigns a human customer service to the user at the same time.

[0085] Step S306, service quality monitoring: The online customer service system monitors in real time whether situations such as changes in the user's emotions or the robot customer service's inability to answer questions occur, which require switching to the human customer service. If so, execute step S308, otherwise, execute step S310.

[0086] Step S308, intelligent switch: If it is detected that the user's emotion escalates, the robot customer service cannot answer the question, etc., automatically prompt the human customer service to take over, and execute Step S312.

[0087] Step S310, the robot customer service continues to answer questions.

[0088] Step S312, human intervention: The human customer service takes over the robot customer service to continue providing services to the user.

[0089] Step S314, the human customer service solves the user's problem.

[0090] Step S316, problem identification, determine whether the user's problem is a problem that the robot customer service can solve. If so, execute Step S3318, otherwise execute Step S320.

[0091] Step S318, the online customer service system prompts the human customer service to hand over the problem to the robot for processing, and execute Step S322.

[0092] Step S320, the human customer service continues to process.

[0093] Step S322, intelligent switch: The human customer service hands over the problem to the robot for processing, releasing human resources.

[0094] The embodiments of the present application have the following advantages:

[0095] (1) When the user accesses the online customer service system, a human customer service is allocated in a timely manner. At this time, it is still the robot customer service that provides services to the user. The human customer service can obtain the chat record between the robot customer service and the user through the chat interface. In the case of the user's high emotion, the robot's inability to answer questions, etc., automatically prompt the human customer service to take over. On the one hand, the human customer service is allocated in advance, and there is no need for routing dispatching and queuing waiting when the robot customer service cannot solve the problem, improving the response efficiency of the human customer service and reducing the user's waiting time; on the other hand, the human customer service can monitor the communication process between the robot customer service and the user through the chat interface, facilitating a quick entry into the takeover state and further improving the response efficiency of the human customer service.

[0096] (2) After the human solves the problem, if the robot can answer the new problem, actively prompt the human customer service to hand it over, releasing the human energy to serve other conversations, reducing resource waste and alleviating the pressure on the human customer service.

[0097] In summary, through the solution of the embodiments of the present application, the human customer service can not only effectively monitor the service situation of the robot customer service, respond to the user's needs in real time, but also take over when the robot customer service can solve the questions by itself, alleviating the work pressure of the human customer service, and finally realizing seamless human-machine collaboration, thereby significantly improving user satisfaction and reducing customer service costs.

[0098] See Figure 10 , an embodiment of the present application further provides a data processing device for an online customer service system, and the device includes:

[0099] An allocation module, configured to allocate a human customer service for the user in response to receiving an access request from the user to the online customer service system;

[0100] A creation module, configured to create a session including the human customer service, the user, and a robot customer service, and set the session to a managed state, so as to respond to a message input by the user in the session through the robot customer service;

[0101] A switching module, configured to switch the session from the managed state to a takeover state in response to receiving a takeover request from the human customer service, so as to respond to a message input by the user in the session through the human customer service.

[0102] An embodiment of the present application further provides a computer device, which at least includes a memory, a processor, and a computer program stored on the memory and executable on the processor. Wherein, when the processor executes the program, the method described in any of the foregoing embodiments is implemented.

[0103] Figure 11 FIG. shows a more specific schematic diagram of the hardware structure of a computer device provided by an embodiment of the present application. The device may include: a processor 502, a memory 504, an input / output interface 506, a communication interface 508, and a bus 510. Wherein, the processor 502, the memory 504, the input / output interface 506, and the communication interface 508 are communicatively connected to each other inside the device through the bus 510.

[0104] The processor 502 may be implemented in a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is configured to execute relevant programs to implement the technical solutions provided by the embodiments of the present application. The processor 502 may further include a graphics card, and the graphics card may be an Nvidia titan X graphics card or a 1080Ti graphics card, etc.

[0105] The memory 504 may be implemented in the form of a Read Only Memory (ROM), a Random Access Memory (RAM), a static storage device, a dynamic storage device, etc. The memory 504 may store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of the present application through software or firmware, the relevant program codes are stored in the memory 504 and called and executed by the processor 502.

[0106] The input / output interface 506 is used to connect to the input / output module to implement information input and output. The input / output module may be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Among them, the input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0107] The communication interface 508 is used to connect to a communication module (not shown in the figure) to implement communication interaction between this device and other devices. Among them, the communication module may implement communication in a wired manner (such as USB, network cable, etc.) or in a wireless manner (such as mobile network, Wi-Fi, Bluetooth, etc.).

[0108] The bus 510 includes a path for transmitting information between various components of the device (such as the processor 502, the memory 504, the input / output interface 506, and the communication interface 508).

[0109] It should be noted that although the above device only shows the processor 502, the memory 504, the input / output interface 506, the communication interface 508, and the bus 510, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary for implementing the solutions of the embodiments of the present application, and do not necessarily include all the components shown in the figure.

[0110] The embodiments of the present application provide a computer program product, including a computer program, which when executed by a processor implements the method described in any embodiment of the present application.

[0111] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method described in any of the foregoing embodiments.

[0112] A computer-readable medium includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information accessible by a computer device. As defined herein, a computer-readable medium does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0113] Each embodiment in this application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description of the method embodiments. The device embodiments described above are only illustrative. The modules described as separate components may or may not be physically separated. When implementing the solutions of the embodiments of this application, the functions of the modules can be implemented in one or more software and / or hardware. It is also possible to select some or all of the modules according to actual needs to achieve the purpose of the solutions of this embodiment. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0114] The above is only the specific implementation manner of the embodiments of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the embodiments of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the embodiments of this application.

Claims

1. A data processing method for an online customer service system, the method comprising: In response to receiving a user's access request to the online customer service system, assigning a manual customer service to the user; Creating a session including the human customer service, the user, and the robot customer service, and setting the session to a managed state so that the robot customer service can respond to a message input by the user in the session; In response to receiving a takeover request from the manual customer service, the session is switched from the managed state to the taken over state, so that the manual customer service responds to the message input by the user in the session.

2. According to the method of claim 1, allocating manual customer service to the user comprises: Assigning a human customer service randomly selected from a plurality of human customer service representatives to the user; or Assigning a human customer service representative who has served the user in the past to the user; or Assigning a human customer service representative who has the highest matching degree with the user information of the user among multiple human customer service representatives to the user; or Assigning a human customer service representative who has taken over the least number of sessions among multiple human customer service representatives to the user; or Multiple manual customer service representatives are polled, and the polled manual customer service representatives are assigned to the user.

3. The method according to claim 1, further comprising: The conversation message of the conversation is displayed in the chat interface of the manual customer service; wherein, When the session is in a managed state, the session message includes a message input by the user and a first response message obtained by the robot customer service in response to the message input by the user; When the session is in the takeover state, the session message includes the message input by the user, a first response message obtained by the robot customer service in response to the message input by the user, and a second response message obtained by the manual customer service in response to the message input by the user.

4. The method according to claim 3, wherein displaying the conversation message of the conversation in the chat interface of the manual customer service comprises: Adding the conversation to the conversation list of the manual customer service; Displaying the conversation list in the chat interface of the manual customer service; In response to the human customer service selecting the session from the session list, the session messages of the session are displayed in the chat interface of the human customer service.

5. The method according to claim 4, wherein the session list includes at least one session, and the order of arrangement of the at least one session in the session list is determined based on at least one of the following: The time the session was created; The categories of events the session involves; User satisfaction during the session.

6. The method according to claim 3, further comprising: When the session is in a managed state, a specific message in the session message is marked; the specific message includes at least one of the following: Information related to user needs; Information related to the transaction number; Information related to user satisfaction; Information related to the processing progress of the robot customer service.

7. The method according to claim 3, further comprising: In response to switching the session from the hosting state to the taking-over state, summary information of the session messages is displayed in the chat interface of the manual customer service.

8. The method according to claim 1, further comprising: When the session is in a managed state, in response to detecting that the service quality of the robot customer service does not meet the user's expectations, a prompt message is output to prompt the human customer service to take over the session.

9. The method according to claim 1, further comprising: When the session is in the takeover state, in response to receiving a hosting request from the manual customer service, the session is switched from the takeover state to the hosting state again.

10. The method according to claim 9, further comprising: When the session is in a takeover state, in response to detecting that the robot customer service can respond to a message input by the user in the session, prompt information is output to prompt the human customer service to host the session.

11. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method according to any one of claims 1 to 10.

12. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 10 when executing the computer program.

13. A computer program product, comprising a computer program, which implements the method according to any one of claims 1 to 10 when executed by a processor.

Images