CN113724036A

CN113724036A - Method and electronic equipment for providing question consultation service

Info

Publication number: CN113724036A
Application number: CN202110867164.4A
Authority: CN
Inventors: 薛晶; 梁宇荣; 简剑; 王德胜; 狄兆龙; 李晓婧; 齐英辉; 王涛; 李明; 黄剑冰
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2021-07-29
Filing date: 2021-07-29
Publication date: 2021-11-30

Abstract

The embodiment of the application discloses a method and electronic equipment for providing problem consultation service, wherein the method comprises the following steps: receiving a target question to be consulted by a user in a target session; judging whether the target question needs to be answered through intelligent customer service resources or not; and if so, determining answer content for answering the target question through the intelligent customer service resource, wherein the answer content is generated by simulating a manual customer service resource to answer the target question according to a pre-established knowledge model. Through the embodiment of the application, the frequency of manual work transfer can be reduced, and the cost of manual work transfer is reduced.

Description

Method and electronic equipment for providing question consultation service

Technical Field

The present application relates to the field of intelligent customer service technologies, and in particular, to a method and an electronic device for providing a problem consultation service.

Background

In the customer service system, manual customer service resources can be provided to provide problem consultation services for users. However, in practical applications, there may be situations where a user initiates a consultation request, and a human customer service resource seat is not on-line or is busy all over, in which case some systems may provide consultation services to the user through intelligent customer service resources (e.g., "customer service robots," etc.). The customer service robot system depends on a pre-established knowledge base, and in the prior art, configuration personnel are usually required to configure and maintain the knowledge base, including adding information such as titles and contents of specific problems to the knowledge base. Thus, after receiving the consultation questions of the user, the customer service robot can search the content which can be used for answering the specific consultation questions from the knowledge base in a way of title matching and the like and feed back the content to the user.

However, in the implementation manner in the prior art, there are often problems that the answers fed back by the customer service robot are more stereotyped, and the expression manner, the content and the like are not rich enough. Therefore, in the process of communicating with the user through the customer service robot, conditions such as consultation interruption and the like may occur due to reasons such as poor user experience and the like, so that resource waste is caused, the function of the customer service robot cannot be really and effectively exerted, and finally, the customer service resource still needs to be changed to provide consultation services for the user.

Disclosure of Invention

The application provides a method and electronic equipment for providing problem consultation service, which can reduce the labor conversion frequency and reduce the labor conversion cost.

The application provides the following scheme:

a method of providing issue advisory services, comprising:

receiving a target question to be consulted by a user in a target session;

judging whether the target question needs to be answered through intelligent customer service resources or not;

and if so, determining answer content for answering the target question through the intelligent customer service resource, wherein the answer content is generated by simulating a manual customer service resource to answer the target question according to a pre-established knowledge model.

A method of providing issue advisory services, comprising:

receiving a target question to be consulted by a user in a target session;

providing an operation option for initiating a help request to an intelligent customer service resource in an interface where a target session window is positioned, wherein the target session window is used for a manual customer service resource to have a conversation with a problem consultant user;

after receiving the help request of the artificial customer service resources through the operation options, calling the intelligent customer service resources to obtain answer content of the target question, wherein the answer content is generated by simulating the artificial customer service resources to answer the target question according to a pre-established knowledge model.

An apparatus for providing a question consultation service, comprising:

a target question receiving unit for receiving a target question to be consulted by a user in a target session;

the judging unit is used for judging whether the target question needs to be answered through intelligent customer service resources or not;

and the answer content determining unit is used for determining answer content for answering the target question through the intelligent customer service resource if needed, and the answer content is generated by simulating a manual customer service resource to answer the target question according to a pre-established knowledge model.

An apparatus for providing a question consultation service, comprising:

the system comprises an operation option providing unit, a question consultant user and a service interaction unit, wherein the operation option providing unit is used for providing an operation option for initiating a help request to an intelligent customer service resource in an interface where a target session window is positioned, and the target session window is used for a dialog between an artificial customer service resource and a question consultant user;

and the calling unit is used for calling the intelligent customer service resource to obtain answer content of the target question after receiving the help request of the artificial customer service resource through the operation options, wherein the answer content is generated by simulating the artificial customer service resource to answer the target question according to a pre-established knowledge model.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any of the preceding claims.

An electronic device, comprising:

one or more processors; and

a memory associated with the one or more processors for storing program instructions that, when read and executed by the one or more processors, perform the steps of the method of any of the preceding claims.

According to the specific embodiments provided herein, the present application discloses the following technical effects:

according to the method and the device, after a question consultation request of a user is received, whether intervention of intelligent customer service resources is needed or not can be judged, if yes, answer content used for answering the target question through the intelligent customer service resources can be determined, and the answer content is generated by simulating artificial customer service resources to answer the target question according to a pre-established knowledge model. That is to say, in the embodiment of the present application, although the intelligent customer service resource provides the consultation service for the user, the answer content provided by the intelligent customer service resource is generated by simulating the answering manner and the like when the manual customer service resource answers the corresponding question, so that the dialog requirement of the user in the daily communication scene is closer to the answer content answered by the intelligent customer service resource, the answer content answered by the intelligent customer service resource can have similar performance when the manual customer service answers, and is more easily accepted by the user, the frequency of manual operation is reduced, and the cost of manual operation is reduced.

The specific knowledge model can be established according to the original text content of the historical conversation data generated by the user related to the target organization in the process of using the associated communication system, and the specific content generated in the historical conversation data has the characteristics of relatively divergent solution scheme and the like, so that the richness of the content of the knowledge base is favorably improved.

Of course, it is not necessary for any product to achieve all of the above-described advantages at the same time for the practice of the present application.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

FIG. 1 is a schematic diagram of a system architecture provided by an embodiment of the present application;

FIG. 2 is a flow chart of a first method provided by an embodiment of the present application;

3-1, 3-2 are schematic diagrams of knowledge contents in a knowledge base provided by the embodiment of the application;

FIGS. 4-1 and 4-2 are schematic views of interfaces provided in embodiments of the present application;

FIG. 5 is a flow chart of a second method provided by embodiments of the present application;

FIG. 6 is a schematic diagram of a first apparatus provided by an embodiment of the present application;

FIG. 7 is a schematic diagram of a second apparatus provided by an embodiment of the present application;

fig. 8 is a schematic diagram of an electronic device provided in an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments that can be derived from the embodiments given herein by a person of ordinary skill in the art are intended to be within the scope of the present disclosure.

In order to facilitate understanding of the technical solutions provided in the embodiments of the present application, it should be first described that, in a process of providing goods or services to a user by a merchant in various ways such as online or offline, the merchant may also generally provide various channels for the user to perform a problem inquiry. For example, a hotline telephone channel, an online customer service channel in a merchandise object information system, an instant messaging channel, an offline self-service terminal channel, and the like may be included. If a user encounters a specific problem during the process of purchasing or using a specific commodity or service, the user can initiate consultation with the corresponding merchant through the various channels. Correspondingly, the merchant side can be provided with manual customer service resources such as seats corresponding to various channels and the like so as to provide consultation services for the user in a manual mode.

In order to facilitate management of merchants to various consultation channels and provide services for consultation requests of various consultation channels by the same artificial customer service resource, some service providers develop intelligent customer service systems. By using the intelligent customer service system, merchants can uniformly route the consultation requests of various channels, and the manual customer service resources can provide services for the consultation requests of various channels only through one workbench without switching among various different terminal devices or application programs, so that the service efficiency of the customer service resources can be effectively improved.

The intelligent customer service system can be generally integrated into a specific communication system, particularly an instant communication system which is dedicated to providing services such as intelligent office and business communication for enterprise users, and the like. For a specific business user such as a merchant, the communication system may be installed for employees of the business, so that communication between employees, between employees and customers (suppliers, etc.), etc. may be performed through the communication system.

In addition, under the condition of integrating the intelligent customer service system, a specific employee in an enterprise can also be provided with an 'artificial customer service' account, the employee can be enabled to have an 'agent' identity in a mode of logging in the 'artificial customer service' account, and the employee can provide consultation services for users in the 'agent' identity through operation options such as 'working in duty' in an interface. Further still, the intelligent customer service system may deploy consulting portals in a variety of channels, such as web pages, detail pages of associated merchandise object information systems, related social networking systems, and so forth. Thus, in the course of a particular user (typically a consumer user of a merchant offering goods or services, etc.) purchasing, using a particular goods object or service, a consultation may be initiated through a variety of channels if problems are encountered, including a return problem, etc. Then, as the merchant deploys the instant messaging system integrated with the intelligent customer service system, the consultation problems of respective channels can be uniformly routed, and the processing such as seat distribution and the like can be carried out. If a certain consultation request is distributed to a certain 'seat', the intelligent customer service system can establish conversation connection between the 'seat' and a consumer user, so that the two can communicate. It should be noted that, at this time, the "agent" may have a conversation with the consumer user through the instant messaging system integrated with the intelligent customer service system, and the consumer user may have a conversation with the "agent" through the original channel (the channel used when the consultation is initiated) (the intelligent customer service system may communicate with the specific channel party).

Of course, a specific "seat" may provide the counseling service for employees of an organization (including internal employees, or employees having outsourced, practice, etc. relationships) in addition to providing the counseling service for the external consumer users, etc. That is, when the employee of the organization has some problems in the work, the employee may also initiate a consultation to the "seat", and at this time, the employee may directly initiate a consultation to the "seat" through the communication system associated with the intelligent customer service system, and so on.

In summary, if an organization (including the aforementioned merchant that provides goods objects or services) deploys a communication system integrated with an intelligent customer service system, session data is generated during the use of the communication system by the relevant users. Such session data may occur between employees and employees, or between employees and customers, or between "agents" and consumer users, "agents" and employees, and so forth. Moreover, the session data includes a large amount of content related to the question consultation and answering. Of course, if the specific organization does not deploy the intelligent customer service system, but provides the consulting services for the users through a plurality of different channels by the manual customer service resources, a large amount of session data can be generated in the process of providing the consulting services. The session data can be useful information for users who need to consult new problems which can be generated subsequently, and therefore, the session data is valuable for constructing a knowledge base.

Based on the above analysis, in the embodiment of the present application, session data generated by a user associated with a specific organization during using a communication system may be collected, and a knowledge model of the organization may be generated by learning the collected session data. Specifically, the knowledge model may be trained according to the acoustic information in the session data. The so-called original sound information is the original text content in the actual historical conversation data. For example, in a session, a consumer user consults a return related question, responds with a human customer service, and may use the question and the textual content of the response to train a knowledge model. The knowledge model may then be maintained and updated as more session data is generated. In this way, the knowledge model generated in this way can be used to provide specific problem consulting services for the user.

By the method, as long as the related communication system is deployed in the organization, and the session data is generated and accumulated for a period of time, the knowledge base of the organization can be automatically established and maintained and updated, so that the process of manually configuring the knowledge base is replaced, and the occupation of the manpower cost in the process of establishing and maintaining the knowledge base is saved. Moreover, such knowledge base may be built and used as the communications system is used by the users associated with the organization without the need to first order how all the people would ask questions and how to answer.

The specific knowledge model may be trained according to the acoustic content in the historical conversation data, the trained knowledge model may take a consultation question provided by a specific user as an input, the output result may be answer content for answering the question, and the answer content may be generated by simulating an artificial customer service resource to answer the question. Therefore, the generated answer content is closer to the conversation requirement of the user in the daily communication scene. For example, in the process of providing the consultation service for a consumer user, the dialogue expression between the consumer user and the agent is more preferable to spoken language, and the expression mode is more flexible, so that the knowledge model can be trained by using the acoustic information, the content responded by the customer service robot can have similar expression when the customer service is manually responded, the customer service robot is more easily accepted by the user, the manual conversion frequency is reduced, and the manual conversion cost is reduced. Under the condition that the artificial customer service resources are allocated to the question consultation request of the user, the intelligent customer service resources can be used for hosting when the artificial customer service resources leave temporarily, or the intelligent customer service resources can be used for answering the user questions when the artificial customer service resources can not answer the user questions; in the process, the intelligent customer service resources can simulate the manual customer service resources, so that the perception of the user for switching between the manual customer service resources and the intelligent customer service resources can be reduced to a certain extent, and the user experience is improved.

Moreover, the original text content generated in the specific historical conversation data also has the characteristics of more divergent solutions and the like, for example, regarding the problems of 'how to remove stains' and the like, if a knowledge base is manually configured, only a unique standard answer is generally configured; however, in the actual session data, the specific session content is generated in the actual consultation process of the specific user, the specific questioning method, the problem details, and the like may be various (for example, stains may be further subdivided into a plurality of types, including oil stains, fruit stains, and the like), the given solutions are also more divergent, may include some relatively professional solutions, may also include some solutions which are not necessarily certified professionally but can actually solve some problems, and the like. Therefore, the content richness of the knowledge base can be favorably improved.

In addition, by the method, data generated by the user in the conversation process can help other users to solve the problem, so that a one-way user help seeking mode is converted into a mutual help mode among multiple users, and the tool of the conversation data content of the user is realized.

From the perspective of system architecture, referring to fig. 1, an embodiment of the present application mainly relates to an intelligent customer service system, which may include a server and a client, where the server is mainly used for collecting data, creating, updating, maintaining, and the like of a knowledge model; the client can be integrated in communication systems such as an instant messaging system (of course, the client can also be an independently operated client), a specific client can be provided for the use of customer service resources of an organization, the organization can obtain the use right of the intelligent customer service system through purchasing and the like, and then the specific customer service resources can provide consultation services for users through the intelligent customer service system. The specific users include consumer users outside the organization, or employee users inside the organization. A consumer user outside an organization can initiate a consultation request through various channels, a staff user inside the organization can directly initiate the consultation request through a related communication system, and the intelligent customer service system can perform uniform routing, so that customer service resources can provide services for the user consultation requests in various channels through the intelligent customer service system. The specific customer service resources may include a "customer service robot" resource, and in the embodiment of the present application, the "customer service robot" resource may provide services for the counselor user by using a knowledge model. The specific knowledge model can be created, updated and maintained after the acoustic information of the historical conversation data generated in the communication system process is collected, analyzed and processed, so that the customer service robot can simulate artificial customer service resources to answer the target question, and the given answer content is closer to the state of the user in a daily conversation scene with the artificial customer service, thereby improving the user experience. Of course, the specific intelligent customer service system can also be directly integrated into various individual customer service channel systems.

The following describes in detail a specific technical solution provided in an embodiment of the present application.

Example one

First, the embodiment provides a method for providing a problem consultation service from the perspective of a service end of an intelligent customer service system, and referring to fig. 2, the method may include:

s201: and receiving a target question which the user needs to consult in the target session.

For example, in a specific implementation, when a certain user browses information of a certain commodity object through an electronic commerce information system, a problem consultation request may be initiated through an operation option such as "customer service" in a detail page, and at this time, the customer service system may create a corresponding session for the request of the user, and so on. After a session is created, the user can enter target questions for specific consultation through a window of the session.

S202: and judging whether the target question needs to be answered through the intelligent customer service resource.

After receiving a target question to be consulted by a user through the target session, it may be determined whether an intervention of an intelligent customer service resource is required, that is, whether the target question needs to be answered by the intelligent customer service resource. In the specific implementation, there may be a plurality of specific determination methods. For example, if the target session is a session to which no artificial customer service resource has been allocated, that is, a newly created session, etc., at this time, it may be determined whether there is an available artificial customer service resource currently, and if so, the artificial customer service resource may be allocated preferentially, otherwise, if there is no available artificial customer service resource currently (for example, the current time may not be within the service time range of the artificial customer service resource, or the artificial customer service resource may be fully busy, etc.), it may be determined that the target question needs to be answered by the intelligent customer service resource.

Or, the target session may be a session to which a target artificial customer service resource is already allocated, that is, assuming that a user initiates a problem consultation request and the system allocates a certain artificial customer service resource to the user to serve the user, the intelligent customer service resource may also intervene at an appropriate time during a session between the user and the artificial customer service resource. Specifically, it may be determined whether the target artificial customer service resource cannot answer the target question, and if so, it is determined that the target question needs to be answered by the intelligent customer service resource. That is, in the process of providing services for the user by using the specific artificial customer service resources, a situation that the target question of the user cannot be answered may occur, and at this time, the user may be provided with services by switching to the intelligent customer service resources.

Specifically, there may be various ways to determine whether the target artificial customer service resource cannot answer the target question. For example, it is determined whether the target artificial customer service resource is unread or does not answer the target question for a long time or the target question is still not solved after a plurality of rounds of interaction, and if so, it may be determined that the target artificial customer service resource cannot answer the target question. If the target question proposed by the user is unread for a long time, the target human customer service resource may be busy in handling other things and cannot answer the question of the user, and at the moment, the intelligent customer service resource can intervene. If the current target artificial customer service resource does not answer the target question proposed by the user for a long time or the target question still cannot be solved after multiple rounds of interaction, the current target artificial customer service resource may not have related knowledge or the related knowledge is relatively deficient. At this time, the intelligent customer service resource can also intervene to replace the current manual customer service resource to return to the target problem of the user. In order to determine whether the target problem is still not solved after multiple rounds of interaction, relevant operation options may be provided at the user side of the consultant, for example, whether the problem of the user is solved may be asked, the user may perform feedback by clicking the corresponding operation option, and accordingly, whether the target problem of the user is solved may be determined according to the feedback result of the user, and so on. Or, in another mode, whether the target problem of the user is solved or not can be judged by a mode of natural language understanding and the like of conversation contents between the user and the target artificial customer service resource, and the like.

In another mode, in order to avoid interference of automatic intervention of the intelligent customer service resource on a normal service process of the intelligent customer service resource, an operation option for actively initiating a help request to the intelligent customer service resource may be provided at the target artificial customer service resource side, so that if the target artificial customer service resource needs to leave temporarily in a process of providing a consultation service for a user, or does not know how to answer a current target question, help may be initiated to the intelligent customer service resource through the operation option. Correspondingly, if a help request initiated by the target artificial customer service resource to the intelligent customer service resource is received, it can also be determined that the target artificial customer service resource cannot answer the target question. In particular, a corresponding operation entry may be provided in a client interface associated with the artificial customer service resource, for example, a "helper panel" may be provided at the right side of the session interface, and the artificial customer service resource may actively initiate a help request to the intelligent customer service resource through the helper panel. Or, in another mode, when the intelligent customer service resource judges that the artificial customer service resource cannot answer the current target question, prompt information can be provided for the artificial customer service resource to inquire whether the artificial customer service resource needs the help of the intelligent customer service resource, and if the artificial customer service resource selects 'need', the intelligent customer service resource intervenes in the session between the artificial customer service resource and the artificial customer service resource.

S203: and if so, determining answer content for answering the target question through the intelligent customer service resource, wherein the answer content is generated by simulating a manual customer service resource to answer the target question according to a pre-established knowledge model.

If the intelligent customer service resources are needed to answer the target questions, possible answer modes used when the artificial customer service resources answer the target questions can be simulated according to the pre-established knowledge model, and then answer contents are generated. Furthermore, the intelligent customer service resource can provide feedback information for the user by using the answer content.

There are many ways to create a knowledge model in relation to a specific knowledge model. For example, in one manner, as described above, after an organization (including a merchant that provides goods objects or services, or other type of corporate group, etc.) obtains the usage rights of the intelligent customer service system, the intelligent customer service system can collect the original content of the historical session data generated by the relevant users of the organization during the usage of the associated communication system, and use the data to establish and update the knowledge model. The related communication system, that is, the communication system integrated with the intelligent customer service system, may be an instant communication system, and the like.

The specific historical session data may include: the artificial customer service resources of the target organization provide session data generated in the process of problem consultation service for the consultant user. A consultant user herein may include a user external to an organization, for example, an organization being a merchant offering a good object or service, and a particular consultant user may include a consumer user of such a good object or service. Or it may also include staff users inside the organization, for example, staff may initiate consultation to the manual customer service resource through the intelligent customer service system if some problems are encountered during daily work, including some hardware devices in the company are out of order, etc. In addition, because the intelligent customer service system is integrated in the communication system, the communication system can have the functions of instant communication and the like, and staff users related to the target organization can communicate in daily work through the communication tool, for example, some project groups are established, project discussion is carried out in the groups together, and the like. Therefore, some conversation data can be generated in the daily communication process between the staff users. In the embodiment of the present application, the session data may also be collected to construct a specific knowledge base.

After specific historical session data is collected, the knowledge model can be trained using the collected historical session data. Wherein, when training a specific knowledge model, the user acoustic information in the historical conversation data can be preserved. That is, the content, expression, and the like of the user performing a conversation with the other party during the history session are retained as knowledge content for providing a service for subsequent user consultation. The historical conversation data can comprise text data and voice data, if the historical conversation data is voice data, the specific acoustic information can comprise not only the content, the expression mode and the like of specific conversations, but also the voice characteristic information of the user, and therefore when the mode of answering questions by the artificial customer service resources is simulated, simulation of the voice of the specific artificial customer service resources can be achieved, specific voice content can be generated, and the like.

In a specific implementation manner, when the knowledge model is generated, keyword extraction and statistics may be performed on various questions, and corresponding question and answer content may be provided for each specific keyword. The keywords may be in the form of word pairs (word _ pair), and information such as objects and actions may be embodied in one keyword pair. For example, as shown in fig. 3-1, a plurality of key term pairs may be extracted, and information such as the number of times each key term pair appears may also be given. Where "price | change" represents that the problem involved in a particular session is associated with a modified price, "new order | represents that the problem involved in a particular session is associated with a new order, and so on. In addition, the corresponding question and answer of each key word pair may also be provided, for example, as shown in fig. 3-2, the question and answer may all retain the original voice information of the user, that is, the original text and the expression of the user when the user asks or answers, specifically, the original text of the question is "change price", and the corresponding original text of the answer may be "good, slightly lower", and so on. The original text content is expressed in a spoken language biased mode, the knowledge model is trained by directly utilizing the original text content, the situations that intelligent customer service resources are too detailed when answering questions can be avoided, conversation habits in the daily communication process of people and people are better met, and even a user cannot perceive that the intelligent customer service resources are in conversation.

It should be noted that in the problem consulting scenario described in the embodiment of the present application, a merchant generally provides consulting services for consumer users or employees in an enterprise, and therefore, the knowledge model may be constructed by taking an organization as a unit. That is, each organization may correspond to a different knowledge model, which is established based on session data generated by the associated user during use of the communication system associated with the intelligent customer service system.

It should be noted that the specific knowledge model may also be created in other manners, for example, it may also be implemented in a configuration manner, but unlike the configuration of the knowledge base in the prior art, in the embodiment of the present application, the specific knowledge base may be knowledge content related to simulating human customer service resource to answer the question, rather than merely configuring a tailored "standard answer" for the specific question.

Because the knowledge model is established in advance, when the current target question needs to be answered through the intelligent customer service resources, the knowledge model can be used for simulating the 'kiss mouth' of the manual customer service resources for answering the target question to generate corresponding answer content. Such answer content can be used to answer the target question consulted by the current user.

For example, as shown in fig. 4-1, suppose that the target question input by the counselor user for a certain merchandise object is "i need a vertical interface", in the prior art, the question may not be included in the knowledge base configured manually, or the expression of the user may not be standardized enough to be understood by the ordinary "customer robot", so that the reply made by the ordinary "customer robot" may be "parent, your question" in turn. Alternatively, even if a normal "customer service robot" can give a reply, it may be stiff. In the embodiment of the application, the knowledge model can be created according to the original voice information of the user in the historical conversation data, so that the counselor user can be more likely to understand by the 'customer service robot' in a random expression mode, the kiss of the artificial intelligence customer service in answering the question can be simulated, and the knowledge content which is not so stiff and is more suitable for conversation in the daily communication occasion of the user is provided. For example, as shown in fig. 4-2, for the same user consultation question, the "customer service robot" in the embodiment of the present application may give a response of "you turn that interface over, that is, up, and do not buy". Such reply content would allow the consultant user to have the experience of directly conversing with the human customer service resource. Of course, in specific implementation, the reply content made by the customer service robot may also be marked, for example, as shown in fig. 4-2, before the specific reply content, a prompt message such as "model training by historical voice" may be provided.

As mentioned above, the intelligent customer service resource may be involved in a specific session during a dialog between the user and the target human customer service resource, so as to provide help for the human customer service resource or the user. In this case, the problem of switching from artificial customer service resources to intelligent customer service resources is involved, and therefore, in a preferred embodiment of the present application, the knowledge content in the specific knowledge model may further include language habit information of the target artificial customer service resources. At this time, the specific answer content may be generated by simulating a manner in which the artificial customer service resource (not specifically referred to a specific artificial customer service resource) answers the target question and a language habit of the target artificial customer service resource. For example, the original answer content may be generated by first simulating the manner in which the human customer service resource answers the target question according to a knowledge model. Then, according to the language habit of the current target artificial customer service resource, the original answer content is adjusted, for example, some words are replaced by words which the current target artificial customer service resource likes to use, or emoticons which the current target artificial customer service resource likes to use are added in the original answer content, and so on.

That is, different man-made customer service resources may have different language habits, including some word habits, language styles (e.g., some prefer pretty, some prefer sincere, etc.), and so on. The language habit information can be reflected in historical conversation data of the artificial customer service resources. Therefore, in the process of creating the knowledge model by using the historical conversation data, the statistical analysis of the historical conversation data can be performed by taking specific artificial customer service resources as units, and the language habit information of each artificial customer service resource can be learned from the statistical analysis. Therefore, if a specific artificial customer service resource is allocated to the current target session and switching to an intelligent customer service resource is required, answer content generated by the specific knowledge model can simulate the artificial customer resource to answer the target question and can also realize simulation of language habits of the current specific artificial customer service resource. By the method, the perception of the user on the switching event from the artificial customer service resource to the intelligent customer service resource can be further reduced, and the user experience is improved.

In addition, in practical application, the human customer service resource may respond to the user's question by text input or by voice in the process of providing the user with the question consultation service. For the former, the answer content may be directly output to the target session. And for the latter, the generated answer content can be converted into voice content and then output to the target session. Alternatively, in a preferred embodiment, the knowledge content in the specific knowledge model may further include: the voice feature information of the target artificial customer service resource is such that, if the target artificial customer service has a dialogue with a user in a voice manner in the historical dialogue data of the target session, a voice content may be generated according to the answer content for outputting the voice content in the target session, wherein a specific voice content may be generated by simulating the target artificial customer service resource to dictate the answer content according to the knowledge model.

That is to say, different artificial customer service resources have different respective voice features, specifically may include voiceprint features and the like, and such features may also be embodied by the historical voice conversation data of the specific artificial customer service resources, so that when the historical voice conversation data is used for learning and training of the knowledge model, the original voice original sound information may also be retained. Therefore, the voice characteristic information of the specific artificial customer service resource can be learned by taking the artificial customer service resource as a unit. Furthermore, when the intelligent customer service system takes over or replaces the previous target artificial customer service resource to answer the target question of the user, the voice, tone and the like of the target artificial customer service resource can be simulated, so that the generated voice content can be heard more like the previous target artificial customer service resource dictating in person. Therefore, the perception of the user on the switching event from the manual customer service resource to the intelligent customer service resource can be further reduced.

Furthermore, the knowledge content in the specific knowledge base may further include: the question category information that a plurality of artificial customer service resources in the associated target organization are respectively good at answering (for example, the question category information can be specifically determined according to high-frequency keywords and the like in historical conversation data associated with the artificial customer service resources). At this time, after it is determined that the current target artificial customer service resource cannot answer the current target question, the intelligent customer service resource cannot determine the answer content capable of answering the target question, and then other artificial customer service resources adept to answer the target question can be determined according to the question category information to which the target question belongs, and the other artificial customer service resources can be accessed into the target session, so that the other artificial customer service resources can answer the target question. That is to say, in the process of constructing the knowledge model, specific session data can be analyzed in the dimension of the artificial customer service resources, and information such as the field where the artificial customer service resources are good at is obtained. Therefore, if the intelligent customer service resources can not solve the user problems and the like, the intelligent customer service resources can also be used for switching other manual customer service resources for the user. For example, in the process of serving a certain user, the artificial customer service resource a finds that the user cannot answer a certain question presented by the user, and at this time, may request intervention of a "customer service robot", and after the intervention of the "customer service robot", finds that there is no content directly corresponding to the question presented by the user in the knowledge base, may forward the artificial customer service resource B, which may be good at answering the question of the user, to the user, and so on. Therefore, the capacity of the customer service robot can not only search answers for the questions provided by the user in the modes of title matching and the like, but also help the user to switch more suitable manual customer service resources aiming at the specific questions provided by the user, so that the capacity of the customer service robot is expanded.

In addition, the knowledge content in the specific knowledge model may further include question scene information to which various answer contents are applicable. For example, if some answer content includes a keyword used in some more formal occasions, it may be determined that the answer content is more suitable for use in scenes such as work, and the like. Therefore, when answer content is generated for the current target question, the questioning scene information of the current consultation requirement can be determined according to the submission time, the submission place and the like of the current target question. For example, if the current time is a work time of a work day, it may be determined that the target user is on work, in a work scene, and so on. Correspondingly, the artificial customer service resources can be simulated to answer the target questions under the questioning scene according to the pre-established knowledge model to generate specific answer contents. For example, if it is determined that the current target user is in a work scene, feedback results may be provided to the current target user according to answer content more suitable for use in the work scene, and so on.

It should be noted that, in practical applications, the specific users having the requirement of consulting questions may further include: employee users associated with the target organization. That is, the employee user may also initiate a consultation with the organization's customer service resources if they encounter problems during the day-to-day work. In this case, the consultation can also be directly initiated to the customer service robot, or the customer service robot intervenes when the artificial customer service resource is not on line or is fully busy or the problem which cannot be solved is encountered in the process of communicating with the artificial customer service resource, and the like. The specific process may be the same as when facing a user outside the organization, such as a consumer user, and will not be described further herein.

However, the processing method is different from the processing method when facing users outside the organization, because the session data can be collected in the process of communicating the human customer service resources with the consultant users and also in the process of daily work of the employees in the organization, when the knowledge base is established, the knowledge content can be established from the dimension of the employee users, and the specific knowledge content can include information of fields, problem types and the like which are good for the specific employee users respectively. For example, if an employee user is found to often use certain key words in a conversation with other users, it may be determined that the employee user has relatively rich knowledge in the domain corresponding to the key words, and so on. In this way, the "customer service robot" may specifically use the knowledge content in the knowledge base to answer the consultation question provided by the employee user, or may forward other manual customer service resources to the employee user, and may also use the knowledge content in the knowledge base to recommend another employee user who can answer the consultation question to the employee user, in addition to using the knowledge content in the knowledge base to answer the consultation question provided by the employee user. That is, for a staff user inside an organization, when he encounters a problem in some way, the "customer service robot" can tell him who to find.

The significance of the above mode is that: in some large organizations, the number of employees is large, and it is almost impossible for each employee to know each other, especially employees in different departments; however, in daily work, there may be a problem with a certain employee, and even another employee may be able to answer the problem. In this case, if two employees know each other, the two employees can communicate with each other directly, but if they do not know each other and do not know which employee can solve the problem, the two employees cannot communicate with each other. For example, employee a in a certain legal department may have some problems with its own computer equipment, while employees in a technical department may help to solve the problems, but may initiate a consultation with a "customer service robot" because they do not know employees in other departments. In the consultation process, if the problem that the customer service robot cannot answer is met, or the problem needs to be manually solved on the employee work station site, another employee B can be recommended to the current employee A according to the knowledge content in the knowledge base, so that the employees A, B can communicate with each other by themselves, or the employee B can be directly transferred to the employee A, and the like.

In summary, according to the embodiment of the present application, after a question consultation request of a user is specifically received, it may be determined whether intervention by an intelligent customer service resource is required, and if so, answer content for answering the target question through the intelligent customer service resource may be determined, where the answer content is generated by simulating an artificial customer service resource to answer the target question according to a pre-established knowledge model. That is to say, in the embodiment of the present application, although the intelligent customer service resource provides the consultation service for the user, the answer content provided by the intelligent customer service resource is generated by simulating the answering manner and the like when the manual customer service resource answers the corresponding question, so that the dialog requirement of the user in the daily communication scene is closer to the answer content answered by the intelligent customer service resource, the answer content answered by the intelligent customer service resource can have similar performance when the manual customer service answers, and is more easily accepted by the user, the frequency of manual operation is reduced, and the cost of manual operation is reduced.

Example two

The second embodiment provides a method for providing question consulting service from the perspective of an intelligent customer service system client associated with artificial customer service resources, and referring to fig. 5, the method may include:

s501: receiving a target question to be consulted by a user in a target session;

s502: providing an operation option for initiating a help request to an intelligent customer service resource in an interface where a target session window is positioned, wherein the target session window is used for a manual customer service resource to have a conversation with a problem consultant user;

s503: after receiving the help request of the artificial customer service resources through the operation options, calling the intelligent customer service resources to obtain answer content of the target question, wherein the answer content is generated by simulating the artificial customer service resources to answer the target question according to a pre-established knowledge model.

For the parts not described in detail in the second embodiment, reference may be made to the description in the first embodiment, and details are not repeated here.

It should be noted that, in the embodiments of the present application, the user data may be used, and in practical applications, the user-specific personal data may be used in the scheme described herein within the scope permitted by the applicable law, under the condition of meeting the requirements of the applicable law and regulations in the country (for example, the user explicitly agrees, the user is informed, etc.).

In correspondence to the first embodiment, an apparatus for providing a problem consultation service is further provided in the first embodiment of the present application, and referring to fig. 6, the apparatus may include:

a target question receiving unit 601 for receiving a target question that a user needs to consult in a target session;

a determining unit 602, configured to determine whether the target question needs to be answered by an intelligent customer service resource;

an answer content determining unit 603, configured to determine, if necessary, answer content for answering the target question through the intelligent customer service resource, where the answer content is generated by simulating an artificial customer service resource to answer the target question according to a pre-established knowledge model.

Wherein the target session comprises a session to which no human customer service resource has been allocated;

in this case, the determining unit may be specifically configured to:

and if no artificial customer service resource is available currently, determining that the target question needs to be answered through the intelligent customer service resource.

Or, the target session comprises a session to which a target artificial customer service resource has been allocated;

in this case, the determining unit may be specifically configured to:

and judging whether the target artificial customer service resource can not answer the target question, if so, determining that the target question needs to be answered through the intelligent customer service resource.

Specifically, the determining unit may be specifically configured to:

and judging whether the target artificial customer service resource is unread or does not answer the target question for a long time or the target question still cannot be solved after multiple rounds of interaction, and if so, determining that the target artificial customer service resource cannot answer the target question.

Alternatively, the determining unit may be specifically configured to:

and if a help request initiated by the target artificial customer service resource to the intelligent customer service resource is received, determining that the target artificial customer service resource cannot answer the target question.

In an alternative implementation, the knowledge content in the knowledge model may include: language habit information of the target artificial customer service resource;

at this time, the answer content is generated by simulating the mode of answering the target question by the artificial customer service resource and the language habit of the target artificial customer service resource according to the knowledge model.

Alternatively, the knowledge content in the knowledge model may also include: voice characteristic information of the target artificial customer service resource;

if the target artificial customer service has a dialogue with the user in a voice mode in the historical dialogue data of the target conversation, the device further comprises:

and the voice content generating unit is used for generating voice content according to the answer content so as to output the voice content in the target session, wherein the voice content is generated by simulating the target artificial customer service resource to dictate the answer content according to the knowledge model.

In addition, the knowledge content in the knowledge model may further include: a plurality of artificial customer service resources in the associated target organization are respectively adept at solving question category information;

the apparatus may further include:

the artificial customer service resource determining unit is used for determining other artificial customer service resources which are adept to answer the target question according to the question category information to which the target question belongs if the answer content cannot be determined;

and the switching unit is used for switching the other artificial customer service resources into the target session so as to answer the target question by the other artificial customer service resources.

In addition, the apparatus may further include:

the question scene determining unit is used for determining a question scene corresponding to the target question;

at this time, the answer content is generated by simulating the artificial customer service resources to answer the target question in the questioning scene according to a pre-established knowledge model.

In particular, the knowledge model may be established based on historical session data textual content generated by users associated with the target organization during use of the associated communication system.

Wherein the historical session data comprises: the artificial customer service resources of the target organization mechanism provide session data generated in the process of problem consultation service for the consultant user, and/or the session data generated in the process of daily communication of the staff users related to the target organization mechanism.

Corresponding to the second embodiment, the embodiment of the present application further provides an apparatus for providing a problem consultation service, and referring to fig. 7, the apparatus may include:

a target question receiving unit 701 for receiving a target question that a user needs to consult in a target session;

an operation option providing unit 702, configured to provide an operation option for initiating a help request to an intelligent customer service resource in an interface where a target session window is located, where the target session window is used for a dialog between an artificial customer service resource and a question consultant user;

the invoking unit 703 is configured to invoke the intelligent customer service resource to obtain answer content of the target question after receiving the help request of the artificial customer service resource through the operation option, where the answer content is generated by simulating that the artificial customer service resource answers the target question according to a pre-established knowledge model.

In addition, the present application also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method described in any of the preceding method embodiments.

And an electronic device comprising:

one or more processors; and

a memory associated with the one or more processors for storing program instructions that, when read and executed by the one or more processors, perform the steps of the method of any of the preceding method embodiments.

Where fig. 8 illustrates an architecture of an electronic device, for example, device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, an aircraft, and the like.

Referring to fig. 8, device 800 may include one or more of the following components: processing component 802, memory 804, power component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814, and communication component 816.

The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing element 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods provided by the disclosed solution. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operation at the device 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The power component 806 provides power to the various components of the device 800. The power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 800.

The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the device 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed state of the device 800, the relative positioning of components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

Communications component 816 is configured to facilitate communications between device 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, or a mobile communication network such as 2G, 3G, 4G/LTE, 5G, etc. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast associated information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communications component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the device 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

In an exemplary embodiment, a non-transitory computer-readable storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the device 800 to perform the methods provided by the present disclosure is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present application may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments of the present application.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The method and the electronic device for providing problem consultation service provided by the application are introduced in detail, and a specific example is applied in the method to explain the principle and the implementation mode of the application, and the description of the embodiment is only used for helping to understand the method and the core idea of the application; meanwhile, for a person skilled in the art, according to the idea of the present application, the specific embodiments and the application range may be changed. In view of the above, the description should not be taken as limiting the application.

Claims

1. A method for providing a problem advisory service, comprising:

receiving a target question to be consulted by a user in a target session;

2. The method of claim 1,

the target session comprises a session to which no human customer service resource has been allocated;

the judging whether the target question needs to be answered through intelligent customer service resources comprises the following steps:

3. The method of claim 1,

the target session comprises a session to which a target artificial customer service resource has been allocated;

4. The method of claim 3,

the judging whether the target artificial customer service resource cannot answer the target question comprises:

5. The method of claim 3,

6. The method of claim 3,

knowledge content in the knowledge model comprises: language habit information of the target artificial customer service resource;

and the answer content is generated by simulating the mode of answering the target question by the artificial customer service resource and the language habit of the target artificial customer service resource according to the knowledge model.

7. The method of claim 3,

knowledge content in the knowledge model comprises: voice characteristic information of the target artificial customer service resource;

if the target artificial customer service has a dialogue with the user in a voice mode in the historical dialogue data of the target conversation, the method further comprises the following steps:

generating voice content from the answer content for outputting the voice content in the target session, the voice content being generated by simulating the target human customer service resource dictating the answer content according to the knowledge model.

8. The method of claim 3,

the knowledge content in the knowledge model further comprises: a plurality of artificial customer service resources in the associated target organization are respectively adept at solving question category information;

the method further comprises the following steps:

if the answer content cannot be determined, determining other artificial customer service resources which are adept at solving the target question according to the question category information to which the target question belongs;

and accessing the other artificial customer service resources into the target session so as to answer the target question by the other artificial customer service resources.

9. The method of any one of claims 1 to 8, further comprising:

determining a question scene corresponding to the target question;

and the answer content is generated by simulating artificial customer service resources to answer the target question in the questioning scene according to a pre-established knowledge model.

10. The method according to any one of claims 1 to 8,

the knowledge model is established based on historical conversation data textual content generated by users associated with the target organization during use of the associated communication system.

11. The method of claim 10,

the historical session data includes: the artificial customer service resources of the target organization mechanism provide session data generated in the process of problem consultation service for the consultant user, and/or the session data generated in the process of daily communication of the staff users related to the target organization mechanism.

12. A method for providing a problem advisory service, comprising:

receiving a target question to be consulted by a user in a target session;

13. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 12.

14. An electronic device, comprising:

one or more processors; and

a memory associated with the one or more processors for storing program instructions that, when read and executed by the one or more processors, perform the steps of the method of any of claims 1 to 12.