WO2020151220A1

WO2020151220A1 - Auxiliary customer service device, method, system, robot, and storage medium

Info

Publication number: WO2020151220A1
Application number: PCT/CN2019/100309
Authority: WO
Inventors: 吴悦; 刘云峰; 汶林丁; 陈志武
Original assignee: 深圳追一科技有限公司
Priority date: 2019-01-22
Filing date: 2019-08-13
Publication date: 2020-07-30

Abstract

Provided are an auxiliary customer service device, a method, a system, a robot, a storage medium, and a computer apparatus, pertaining to the technology of artificial intelligence. The method comprises: receiving user query information, and locating, according to the user query information, a user question list associated with the user query information; searching a robot knowledge database according to the user question list, displaying response information matching the user question in a combined list to a customer service employee, and pushing corresponding response information to the customer service employee according to a selection of the customer service employee; performing a customized response operation on the user question list according to the response information; and performing a feedback operation according to a selection of the customer service employee. The invention enables the customer service employee to add a customized response for different questions, such that the customer service employee is capable of directly using the contents of previously added customized responses when handling similar questions in actual conversations, thereby improving operation efficiency of customer services.

Description

Customer service auxiliary device, method, system, robot and storage medium

Cross references to related applications

This application is required to be submitted to the Chinese Patent Office on January 22, 2019, the application number is 2019100609987, the application name is "customer service assistance robots, methods and systems" and the application 2019100609493, the application name is "a customer service assistance device and method" in China The priority of the patent application, the entire content of which is incorporated in this application by reference.

Technical field

This application belongs to the field of artificial intelligence technology, and specifically relates to a customer service auxiliary device, method, system, robot, storage medium, and computer equipment.

Background technique

With the development of Internet technology, virtual robots based on artificial intelligence technology are increasingly used in various service fields. The existing customer service assistant robots will directly reuse the existing customer service robot capabilities for external services, and find appropriate answers in the robot knowledge base based on the user's information in the manual phase, and provide them for reference.

When dealing with the same question, the customer service assistant robot often gives a large and comprehensive descriptive answer, and the human customer service will gradually communicate with the user according to the previous handling experience. If for different questions, the human customer service can create exclusive custom replies. When the next time the conversation is processed, the customer service assistant robot will directly display the previously saved custom reply words to the human customer service, which will be greatly shortened. The processing time of manual customer service improves the work efficiency of manual customer service.

Summary of the invention

According to various embodiments provided in this application, a customer service assistance device, method, system, robot, storage medium, and computer equipment are provided:

A customer service auxiliary device, including:

The calling module is used to receive user consultation information and send the user consultation information to the location search module;

The location search module is configured to locate the user question list associated with the user question information according to the user question information, and search the robot knowledge base for response information that matches the user question list according to the user question information;

The display module is used to display the response information combination list matching the user question list to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

The custom reply management module is used for manual customer service to perform custom reply operations on the user question list based on the reply information.

A customer service assistant robot, including a receiving module, a searching module, and a pushing module;

The receiving module is configured to receive an acquisition request sent by a customer service staff, and the acquisition request carries user consultation information;

The search module is configured to search for and display a list of questions related to the consultation information to the customer service staff from the robot knowledge base according to the acquisition request, so that the customer service staff can select the questions related to the consultation information from the question list content;

The push module is configured to push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.

A customer service assistance method, including:

Receive user consultation information and locate the user question list associated with the user consultation information based on the user consultation information;

According to the user question list, search and display the response information combination list matching the user's question list in the robot knowledge base to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

Customize the reply operation to the questions in the user's question list based on the reply information;

Perform feedback operations based on the selection of manual customer service.

A customer service assistance method, including:

Receiving user consultation information, and searching and displaying a list of questions related to the consultation information from the robot knowledge base according to the consultation information;

Show the list of questions to the customer service staff;

Push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.

A customer service system, including a customer service robot module, a manual customer service module, and an auxiliary robot module;

The customer service robot module is used to receive consultation information input by a user;

The manual customer service module is used to receive and generate an acquisition request based on the consultation information, and at the same time send the acquisition request to the auxiliary robot module, and is also used for the selection of the customer service staff in the question list, and send it to the auxiliary robot module Selection instruction, the acquisition request carries the consultation information;

The auxiliary robot module is configured to receive the acquisition request, search and display a list of questions related to the consultation information from the robot knowledge base according to the acquisition request; receive and send it to the manual customer service module according to the selection instruction Push corresponding reply information, where the reply information includes at least one process guidance node, and each process guidance node corresponds to a preset reply text.

One or more non-volatile computer-readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to perform the following steps:

Perform feedback operations based on the selection of manual customer service.

Show the list of questions to the customer service staff;

A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more Each processor performs the following steps:

Perform feedback operations based on the selection of manual customer service.

Show the list of questions to the customer service staff;

The details of one or more embodiments of the application are set forth in the following drawings and description. Other features, purposes and advantages of this application will become apparent from the description, drawings and claims.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

Figure 1 is a schematic diagram of a customer service assistance device in an embodiment;

Figure 2 is a schematic diagram of a custom reply management operation in an embodiment;

FIG. 3 is a schematic diagram of a result of a custom reply management operation in an embodiment;

FIG. 4 is a schematic diagram of the structure of a customer service assistant robot in an embodiment;

Fig. 5 is a schematic diagram of setting a scene in an embodiment;

FIG. 6 is a schematic diagram of setting a process node in an embodiment;

Figure 7 is a schematic diagram of a master node setup in an embodiment;

FIG. 8 is a schematic diagram of a master node setting in an embodiment;

FIG. 9 is a schematic structural diagram of a customer service assistant robot in an embodiment;

Figure 10 is an application environment diagram of a customer service assistance method in an embodiment;

Figure 11 is a schematic diagram of the internal structure of a computer device in an embodiment;

Figure 12 is a step diagram of a customer service assistance method in an embodiment;

Figure 13 is a step diagram of a customer service assistance method in another embodiment;

Fig. 14 is a step diagram of a customer service assistance method in another embodiment.

detailed description

In order to make the objectives, technical solutions, and advantages of this application clearer, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the application, and not to limit the application.

As shown in Figure 1, this application provides a customer service assistance device, which includes:

The calling module is used to receive user consultation information and send the user consultation information to the location search module.

Further, the user consultation information includes the last group of conversations received in the customer service robot stage, and the first two groups of conversations received in the queue stage and the manual stage.

Specifically, in the customer service workbench, when receiving the consulting information sent by the user, the customer service robot first handles the problem. If the processing capacity of the customer service robot cannot meet the user’s problem needs, the system transfers to manual customer service; After two rounds of conversations are completed in the customer service processing stage, the system calls the customer service assistant robot to assist the manual customer service for problem handling according to a specific strategy.

For example, at the customer service workbench, a user inquiry message "Hello, has my purchase been shipped?" At this time, the system arranges the customer service robot to handle the problem, and the response sent by the customer service robot based on the user's inquiry is:

"Hello, I am Xiao Yi, the robot, I am very happy to serve you

Do you want to ask:

1. When will the order be shipped?

2. What kind of express delivery does logistics use

3. Return process

4. What to do if there is a problem with the goods"

If the user adopts the answer list displayed by the customer service robot, the user clicks on the corresponding list according to his needs to continue the consultation; if the user does not adopt the answer list displayed by the customer service robot, but selects the manual customer service for consultation, the system will transfer to the manual customer service. deal with.

It should be noted that the above-mentioned specific strategy is to combine user consultation information sent by users at different stages ((bot service stage, the last text message of the user) + (manual customer service stage, the first two text messages of the user))) Call the customer service assistant robot.

The location search module is used to locate the user question list associated with the user question information according to the user question information, and search the robot knowledge base for the answer information that matches the user question list according to the user question list.

Further, the user question list includes at least one question topic associated with the user consultation information.

Specifically, after calling the customer service assistant robot, the customer service assistant robot locates the user question list associated with the user consultation information according to the received user consultation information, and displays the top1 or top3 matched questions to the human customer service through the display module.

For example, if a user inquiry message "Hello, has my purchase shipped?" is received at the customer service workbench, the customer service assistant robot locates the associated user question list based on the user inquiry information, as shown in Table 1:

Table 1

Further, the response information combination list includes at least one question topic and its corresponding matching rate.

Specifically, the customer service assistant robot searches the robot knowledge base for the answer information that matches each question in the user's question list according to the located question list, and displays it to the human customer service, as shown in Table 2:

Table 2

The display module is used to display the response information combination list matching the user's question list to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service.

Specifically, after the customer service assistant robot completes the search and matching process in Table 2 above, it displays the matched response information combination list to the human customer service through the display module. When the manual customer service clicks to select the corresponding question in the corresponding reply information combination list, the reply message text that matches the question is automatically copied to the input area, and the manual customer service can perform secondary editing based on the actual conversation, which shortens the manual customer service's problem processing time ,Improve work efficiency.

The custom reply management module is used for manual customer service to perform custom reply operations on the questions in the user's question list based on the reply information.

Further, custom response operations include: adopt and add; among them, adopt-used for manual customer service to select the reply information pushed by the customer service assistant robot; add-used for manual customer service to add the reply information pushed by the customer service assistant robot as custom reply content .

Specifically, according to the actual conversation, the human customer service clicks on a related question from the response information combination list displayed by the customer service assistant robot, and the question displays its configured response content. The user can click the "Edit custom response" configured under the question. In the pop-up window, add the reply message sent to the user in this conversation to the custom reply. When the next time the conversation is processed, if the customer service assistant robot matches the same question, it will also show the previously added custom reply to the human customer service.

For example, take the "deposit refund" in the XX bicycle APP as an example. After receiving the user inquiry information "I want to consult how to refund the deposit" at the customer service workbench, the customer service assistant robot locates the user problem "Refund of the deposit" related to "I want to consult how to refund the deposit" according to the user consultation information; manual customer service is based on actual conditions The conversation adopted the relevant reply information of "deposit refund", and made a custom reply operation under the topic of "deposit refund". Click "Edit Custom Reply" under the question, and when the edit window appears, there are two ways to add a custom reply, as shown in Figure 2. 1. Directly click on the message sent by the human customer service in this conversation on the far right , The new content is automatically generated to complete the addition of the custom reply; 2. Click "Add custom reply" on the left side of the page, and when the editing window appears, manually add the new content to complete the addition of the custom reply.

Further, the reply information is marked as annotated data after a custom reply operation and the number of uses is recorded.

First of all, after the manual customer service adopts the reply information displayed by the customer service assistant robot, whether it is to adopt the standard reply configured under the question or a custom reply, it is equivalent to generating a label data (the label data is: user consultation information and user consultation information Matching question), and record the number of times that the manual customer service adopts the custom reply configured under the question; secondly, the operation of adding a custom reply to the manual customer service is equivalent to generating an annotation data (the annotation data is: User consultation information and user consultation information matching issues). The annotation data generated in the above two cases can be used to return to the robot module for training, which can reduce the operating cost of the robot.

Based on the customer service assistance device provided by the foregoing embodiment, the embodiment of the present application provides another alternative implementation. As shown in FIG. 3, the device also includes a feedback module, which is used for manual customer service not responding to information If the response information matching the user's consultation information is found in the combination list, the feedback question is missing; at the same time, the feedback module is also used to use the feedback data generated by the manual customer service during the operation to train the customer service assistant robot.

Further, the aforementioned feedback data includes user consultation information and reply information.

Specifically, after the customer service assistant robot shows the list of questions found by its positioning to the manual customer service, if the manual customer service does not find a response that matches the user’s consulting information, the manual customer service can manually search for the problem, and the manual search has There are two ways: 1. Click the search icon and enter user consultation information in the search box that appears to search; 2. Hover the mouse over the user consultation information, click the search icon that appears, and enter the user consultation information in the search box that appears Perform a search. After manual search by manual customer service, if the answer that matches the user's inquiry information cannot be located, the administrator is notified that the question is missing.

As shown in FIG. 4, in one embodiment, a customer service assistance robot is provided, including a receiving module, a searching module, and a pushing module;

The receiving module is used to receive the acquisition request sent by the customer service staff, and the acquisition request carries the user's consultation information.

Specifically, in the customer service workbench, the system sends user consultation information texts sent by users at different stages according to specific strategies ((bot service stage, the user's last text message) + (manual customer service stage, the user's first two text messages)) Call the customer service assistant robot after the combination.

The search module is used to search and display a list of questions related to the consultation information from the robot knowledge base according to the acquisition request, so that the customer service staff can select content related to the consultation information from the question list.

Specifically, after the customer service assistant robot is successfully called, it searches the robot knowledge base for a list of questions related to the user consultation information according to the user consultation information received by the receiving module, and displays the question list to the customer service staff for reference after the search is successful.

For example, in an application scenario of the XX bicycle APP, the user asks the question "How do I return the deposit?" After the system successfully invokes the customer service assistant robot according to the user's question text message, the intelligent assistant function is turned on for the customer service staff, and the customer service assistant robot automatically In the robot knowledge base, the search location matches the question list of "How do I return the deposit". The search results are shown in Table 3:

table 3

Further, the above-mentioned search module is also used to search and display a list of words related to the consultation information from the quick response library according to the acquisition request and display to the customer service staff, so that the customer service staff can select content related to the consultation information from the speech list.

Also referring to the above example, in the application scenario of the XX bicycle APP, the user asks the question "How do I return the deposit?" After the system successfully invokes the customer service assistant robot according to the user's question text message, the intelligent assistant function is turned on for the customer service staff, and the customer service assistant robot The list of questions matching "I want to consult how to return the deposit" is automatically searched in the quick reply database. The search results are shown in Table 4:

Table 4

The content of the list of words in the quick response database in Table 4 is mostly used to soothe the emotions of the customers, so that the customer service staff can respond to the users in time when processing the user's inquiry information, shorten the waiting time of the users, and improve the user experience.

The push module is used to push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, wherein the reply information includes at least one process guidance node, and each process guidance node corresponds to a preset reply text.

Specifically, each reply message corresponds to a topic scene, and each topic scene corresponds to a specific question topic in the robot knowledge base.

It should be noted that, for the setting of scenes, the administrator creates process guidelines in the background system, and each scene corresponds to a question in the robot knowledge base. At least one process master node is set for each scene, and each process master node can be set with one or more first-level sub-nodes; each process master node or each first-level sub-node is associated with any one of the robot knowledge base or the quick reply library ; The robot knowledge base or quick reply library contains the verbal content corresponding to the topic of the question.

For example, the credit card loss reporting business scenario in the credit card business, as shown in Figure 5, first, the scenario name is set, the scenario problem and the scenario description, where the scenario name is "Credit Card Loss Reporting Scenario", and the scenario description is "Credit Card Loss Reporting is the discovery of the stolen credit card or What needs to be done when it is lost, many users may think of holding an ID card to report the loss. However, many sophisticated criminals can transfer a large amount of cash from the card in a very short time, especially credit cards. ". This scenario description helps administrators manage and distinguish the content of each scenario. Secondly, as shown in Figure 6, 14 process nodes can be set according to this scenario, as shown in Table 5:

table 5

Finally, as shown in Figure 7, after setting up each process node of the scene, each main node and first-level child node selects the associated robot knowledge base as the source of reply words; or, as shown in Figure 8, in the setting After completing each process node of the scene, each main node and first-level child node selects the associated quick reply library as the source of the reply words, and completes the setting of the flow guide content of the scene. It should be noted that, in the robot knowledge base mentioned in this embodiment, the verbal content in the quick reply library is pre-organized and can be directly used in association.

Based on the customer service assistance robot provided in the above embodiment, the administrator has carried out different process guidance configurations for different user intentions. When the customer service personnel actually deal with the problem, they have clear processing ideas and targeted speech content support. Standardize the uniformity of services and improve the work efficiency of customer service staff.

Based on the customer service assistance robot provided by the foregoing embodiment, the embodiment of the present application provides another alternative implementation. As shown in FIG. 9, it also includes a positioning module, which is used to obtain user consultation information carried in the request. Locate user intent questions and show the user intent questions to the customer service staff through the question list.

Further, the above question list includes at least one user intention question and its corresponding matching rate.

Specifically, taking the XX bicycle APP as an example, in an application scenario of the XX bicycle APP, the user asks the question "How do I return the deposit?" After the system activates the customer service assistant robot according to the user's question text message, it turns on intelligent assistance for the customer service staff Function, the customer service assistant robot automatically searches the robot knowledge base for words that match the "I want to consult how to return the deposit". The search results are shown in Table 6:

Table 6

After successfully calling the customer service assistant robot at the customer service workbench, first, the customer service assistant robot locates the user’s intention question as “deposit refund” or “how to refund balance/deposit” or “has been Operation deposit refund” also gives the corresponding matching degree. Secondly, the customer service assistant robot searches the robot knowledge base for preset reply texts related to the three user intent questions based on the three user intent questions located, and sends them to The customer displays the list of questions. Finally, after the customer service staff receives the list of questions in Table 6, they can make a selection based on the matching degree. If the customer service staff chooses the highly matched "deposit refund" topic related speech flow guide content, by clicking on the "refund deposit" master node, the master node has pre-configured speech content "Hello, the deposit refund method is as follows: worship Cycling APP-Personal Center in the upper left corner......" will be automatically copied to the input area, and the customer service staff can use this wording for secondary editing and use. Through this process, the operator can promptly communicate with the user and save the customer service staff. Answer retrieval and output time improve the work efficiency of customer service staff.

Fig. 10 is an application environment diagram of a method for assisting customer service in an embodiment. 10, the application environment includes a terminal 1010 and a server 1020, and the terminal 1010 can connect and communicate with the server 1020 through a network. Among them, the server 1020 receives the user consultation information, and locates the user question list associated with the user consultation information according to the user consultation information; the server 1020 searches the robot knowledge base according to the user question list and displays to the manual customer service information that matches the user question list The reply information combination list pushes corresponding reply information to the manual customer service according to the selection of the manual customer service; the server 1020 performs a custom reply operation to the questions in the user question list according to the reply information; the server 1020 performs the feedback operation according to the selection of the manual customer service. The terminal 1010 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server 420 may be implemented by an independent server or a server cluster composed of multiple servers.

Fig. 11 is a schematic diagram of the internal structure of a computer device in an embodiment. The computer device may specifically be a server 1020 or a user terminal 1010 as shown in FIG. 10. 11, the computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and computer-readable instructions. When the computer-readable instructions are executed, the processor can execute a customer service assistance method. Computer readable instructions may be stored in the internal memory of the computer device, and when the computer readable instructions are executed by the processor, the processor can make the processor execute a customer service assistance method. The network interface of the computer device is used to communicate via the network accordingly, such as receiving user consultation information.

FIG. 12 is a schematic flowchart of a customer service assistance method in an embodiment of the present application. This embodiment mainly uses the method applied to the server 1020 in FIG. 10 as an example. As shown in FIG. 12, the method includes:

S1201: Receive user consultation information, and locate a user question list associated with the user consultation information according to the user consultation information.

S1202: Search the robot knowledge base according to the user question list and display to the manual customer service a list of response information combinations that match the user's question list, and push corresponding response information to the manual customer service according to the selection of the manual customer service.

S1203: Perform a custom reply operation on the user question list according to the reply information.

S1204: Perform a feedback operation according to the selection of manual customer service.

In one embodiment, after step S1201, the method further includes the step

Show the list of questions to the customer service staff. According to the selection of the customer service personnel, the corresponding reply information is pushed to the customer service personnel. The reply information includes at least one process guidance node, and each process guidance node corresponds to a preset reply text.

Specifically, the user’s question text information is received through the human-computer interaction platform at the customer service workbench, and the system combines the user’s text messages at different stages according to a specific strategy, and then automatically calls the customer service assistance robot; on the customer service assistance robot platform, the robot The model locates in the robot knowledge base to find the process guidance content that matches the above-mentioned user problem (the text content in the robot knowledge base is preset), and then displays the found process to the customer service staff through the process guidance content management module Guide content, customer service personnel click on each node in the corresponding process guide through the process guide operation module, and the configured speech content of each node will be directly copied to the input area for customer service staff to refer to and use, and the customer service staff can perform this speech content twice edit.

In one embodiment, a customer service assistance method is provided. As shown in FIG. 13, the method includes:

Step 1301, receiving the user's consultation information, and searching and displaying a list of questions related to the consultation information from the robot knowledge base according to the consultation information;

Step 1302, show the question list to the customer service staff;

Step 1303: Push corresponding reply information to the customer service personnel according to the selection of the customer service personnel. The reply information includes at least one process guidance node, and each process guidance node corresponds to a preset reply text.

Based on the customer service assistance method provided by the foregoing embodiment, the embodiment of this application provides another alternative implementation, as shown in FIG. 14:

Step 1401, receiving user consultation information, and searching and displaying a list of words related to the consultation information from the quick response database according to the consultation information;

Step 1402, show a list of words to the customer service staff;

In step 1403, according to the selection of the customer service personnel, the verbal content related to the user consultation information is pushed to the customer service personnel.

Specifically, the user’s question text information is received through the human-computer interaction platform at the customer service workbench, and the system combines the user’s text messages at different stages according to a specific strategy, and then automatically calls the customer service assistance robot; on the customer service assistance robot platform, the robot The model locates in the quick reply database to find a list of words that match the above-mentioned user question (the text content in the quick reply database is preset), so that the customer service staff can select the words related to the user's consultation information from the word list For the content, the customer service staff clicks on the related question in the speech skill list, and the configured speech skill content will be directly copied to the input area for reference by the customer service staff, and the customer service staff can edit the speech skill content twice.

What needs to be added is that after the customer service assistant robot is successfully called on the customer service workbench, the customer service staff can manually search for user problems. There are two search methods: one is to click the search icon on the customer service conversation interface, and the search appears Enter keywords related to the user’s question in the box to search, and select the relevant scene as the process guide content according to the search result; one is to hover the mouse over the user message in the customer service conversation interface, click the search icon that appears, and search Enter keywords related to the user's question in the box to search, and select the relevant scene as the process guide content according to the search structure.

If after manual search by the customer service personnel, the scene theme related to the above user problem cannot be located, the customer service personnel clicks "Notify the administrator of missing problem" located at the bottom of the customer service personnel conversation interface to notify the administrator that the above user problem cannot be found , The administrator can create in the background scene guidance process content related to the above user issues based on the conversation records of the customer service staff. Through the process guidance management of different questions, the customer service staff can save the time to think about the related speech content after receiving the user’s question, and directly click on each node in the process guide to complete the response of the relevant speech content, which overcomes the problem of online manual services. The shortcomings of the slow speed of customer service staff to answer questions, improve the efficiency of customer service.

When dealing with user questions, the customer service assistant robots often give large and comprehensive descriptive answers, and the manual customer service will gradually communicate with users according to certain process steps. Therefore, the answers provided by existing customer service assistant robots are often untargeted and difficult to use directly.

Based on the customer service assistance method provided by the above-mentioned embodiment, based on the customer service assistance robot finding positioning problems, the custom response of manual customer service is used for recording and precipitation, so as to achieve targeted improvement of the work efficiency of manual customer service, and at the same time manual customer service The feedback data during use can be used to train the robot and reduce the operating cost of the robot.

The customer service assistance method provided in the embodiments of the present application can be implemented by the above-mentioned embodiments of the customer service assistance device and the customer service assistance robot, and will not be repeated here.

In order to realize the above technical solution, an embodiment of the present application also provides a customer service assistance system, including a customer service robot module, a manual customer service module, and an auxiliary robot module;

specific,

The customer service robot module is used to receive the consultation information input by the user;

The manual customer service module is used to receive and generate an acquisition request based on the consulting information, and at the same time send the acquisition request to the auxiliary robot module, and also used for the selection of the customer service staff in the question list, and send the selection instruction to the auxiliary robot module, where the acquisition request carries the consulting information;

The auxiliary robot module is used to receive the acquisition request, according to the acquisition request, find and display the list of questions related to the consulting information from the robot knowledge base; receive and push the corresponding reply information to the manual customer service module according to the selection instruction, and the reply information contains at least one process guide Node, each process guidance node corresponds to the preset reply text.

The functions of each module of the customer service assistance system provided in the embodiments of the present application can be implemented through the steps in the above customer service assistance method embodiments, and will not be repeated here.

A computer device including a memory and one or more processors. The memory stores computer readable instructions. When the computer readable instructions are executed by one or more processors, the one or more processors can implement any one of the present application. The steps of the customer service assistance method provided in the embodiment.

One or more non-volatile computer-readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors implement any one of the embodiments of the present application. Provide the steps of the customer service assistance method.

A customer service assistance device, method, robot, system, storage medium, and computer equipment provided by the embodiments of this application. Based on the customer service assistance robot positioning and finding user problems, the manual customer service can easily add custom replies. When the customer service processes the same conversation, the customer service assistant robot will display the previously added custom reply content. By accumulating the custom reply records of the manual customer service, the work efficiency of the manual customer service can be improved in a targeted manner, and the manual customer service is in the operation process The feedback data can also be used for robot model training, which reduces the operating cost of the robot and improves the efficiency of customer service. And on the basis of using the customer service assistant robot to find and locate the user's intention problem, firstly, for different user intention problems, the administrator pre-sets the processing flow steps, and different steps correspond to different words content; secondly, in the customer service workbench, use The customer service assistant robot searches for and displays a list of questions related to the user's consultation information in the robot knowledge base; finally, according to the customer service staff's choice, pushes the process guidance steps and words of the problem configuration to the customer service staff, and provides them to the customer service staff. For reference, if the customer service staff does not select the content related to the consultation information from the question list, the customer service staff will deal with the actual conversation. The above technical solutions enable customer service personnel to have clear processing ideas and targeted language support when actually handling problems, standardize the uniformity of services, and improve the efficiency of customer service.

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application. Should be covered within the scope of protection of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A customer service auxiliary device, characterized by comprising:

The calling module is used to receive user consultation information and send the user consultation information to the location search module;

The location search module is configured to locate the user question list associated with the user question information according to the user question information, and search the robot knowledge base for response information that matches the user question list according to the user question information;

The display module is used to display the response information combination list matching the user question list to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

The custom reply management module is used for manual customer service to perform custom reply operations on the user question list based on the reply information.
The device according to claim 1, further comprising a feedback module;

The feedback module is used for manual customer service to fail to find the reply information that matches the user's consultation information from the reply information combination list, then the feedback question is missing.
The device according to claim 2, wherein the feedback module is further used to use the feedback data generated by the manual customer service during the operation process to train the customer service assistant robot.
The device according to claim 3, wherein the feedback data includes the user consultation information and the reply information.
The device according to claim 1, wherein the user consultation information includes the last group of conversations received in the customer service robot phase, and the first two groups of conversations received in the queuing phase and the manual phase.
The device according to claim 1, wherein the custom reply operation includes: adopt and add;

The acceptance is used for manual customer service to select the reply information pushed by the customer service assistant robot;

The adding is used for manual customer service to add the reply information pushed by the customer service assistant robot as custom reply content.
8. The device according to claim 7, wherein the reply information is marked as annotated data after the custom reply operation and the number of uses is recorded.
The apparatus according to claim 1, wherein the user question list includes at least one question topic associated with the user consultation information.
8. The device according to claim 8, wherein the response information combination list contains at least one of the question topics and their corresponding matching rates.
A customer service assistant robot, characterized in that it includes a receiving module, a searching module and a pushing module;

The receiving module is configured to receive an acquisition request sent by a customer service staff, and the acquisition request carries user consultation information;

The search module is configured to search for and display a list of questions related to the consultation information to the customer service staff from the robot knowledge base according to the acquisition request, so that the customer service staff can select the questions related to the consultation information from the question list content;

The push module is configured to push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.
The customer service assistant robot according to claim 10, wherein the search module is further configured to search from a quick response database according to the acquisition request and display a list of words related to the consultation information to the customer service staff, so that the customer service The person selects content related to the consultation information from the speech list.
The customer service assistance robot according to claim 10, further comprising a positioning module;

The positioning module is configured to locate the user's intention problem according to the consulting information carried by the user in the acquisition request;

The user intention question is displayed to the customer service staff through the question list.
The customer service assistance robot according to claim 12, wherein the question list contains at least one of the user intention questions and the corresponding matching rate.
The customer service assistance robot according to claim 10, wherein each of the reply information corresponds to a topic scene, and each of the topic scenes corresponds to a specific question topic in the robot knowledge base.
The customer service assistance robot according to claim 14, wherein each of the theme scenes includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.
The customer service assistant robot according to claim 15, wherein each of the process guidance nodes includes a main node and a first-level child node;

Each main node may include one or more first-level sub-nodes, and each main node and/or each first-level sub-node corresponds to a preset reply text.
The customer service assistance robot according to any one of claims 10-16, wherein if the customer service personnel does not select content related to the consultation information from the question list or the speech list, the customer service personnel The actual conversation is processed.
A customer service assistance method, characterized in that it comprises:

Receive user consultation information and locate the user question list associated with the user consultation information based on the user consultation information;

According to the user question list, search and display the response information combination list matching the user's question list in the robot knowledge base to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

Customize the reply operation to the questions in the user's question list based on the reply information;

Perform feedback operations based on the selection of manual customer service.
A customer service assistance method, characterized in that, the method further includes:

Receiving user consultation information, and searching and displaying a list of questions related to the consultation information from the robot knowledge base according to the consultation information;

Show the list of questions to the customer service staff;

Push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.
A customer service system, characterized by comprising a customer service robot module, a manual customer service module and an auxiliary robot module;

The customer service robot module is used to receive consultation information input by a user;

The manual customer service module is used to receive and generate an acquisition request based on the consultation information, and at the same time send the acquisition request to the auxiliary robot module, and is also used for the selection of the customer service staff in the question list, and send it to the auxiliary robot module Selection instruction, the acquisition request carries the consultation information;

The auxiliary robot module is configured to receive the acquisition request, search and display a list of questions related to the consultation information from the robot knowledge base according to the acquisition request; receive and send it to the manual customer service module according to the selection instruction Push corresponding reply information, where the reply information includes at least one process guidance node, and each process guidance node corresponds to a preset reply text.
One or more non-volatile computer-readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to perform the following steps:

Receive user consultation information and locate the user question list associated with the user consultation information based on the user consultation information;

According to the list of user questions, search and display to the manual customer service a list of response information combinations that match the user's question list in the robot knowledge base, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

Customize the reply operation to the questions in the user's question list based on the reply information;

Perform feedback operations based on the selection of manual customer service.
One or more non-volatile computer-readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to perform the following steps:

Receiving user consultation information, and searching and displaying a list of questions related to the consultation information from the robot knowledge base according to the consultation information;

Show the list of questions to the customer service staff;

Push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.
A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more Each processor performs the following steps:

Receive user consultation information and locate the user question list associated with the user consultation information based on the user consultation information;

According to the user question list, search and display the response information combination list matching the user's question list in the robot knowledge base to the manual customer service, and push the corresponding response information to the manual customer service according to the selection of the manual customer service;

Customize the reply operation to the questions in the user's question list based on the reply information;

Perform feedback operations based on the selection of manual customer service.
A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more Each processor performs the following steps:

Receiving user consultation information, and searching and displaying a list of questions related to the consultation information from the robot knowledge base according to the consultation information;

Show the list of questions to the customer service staff;

Push corresponding reply information to the customer service personnel according to the selection of the customer service personnel, the reply information includes at least one process guidance node, and each of the process guidance nodes corresponds to a preset reply text.