CN114637828A - Customer service response method and device - Google Patents
Customer service response method and device Download PDFInfo
- Publication number
- CN114637828A CN114637828A CN202210151129.7A CN202210151129A CN114637828A CN 114637828 A CN114637828 A CN 114637828A CN 202210151129 A CN202210151129 A CN 202210151129A CN 114637828 A CN114637828 A CN 114637828A
- Authority
- CN
- China
- Prior art keywords
- user
- question
- answer
- intention
- customer service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
Abstract
The application provides a customer service response method, which comprises the following steps: receiving question information sent by a user in a sub-account of a primary account through the primary account; identifying a user question intention according to the question information and the sending interface of the user question information; matching answers corresponding to the question information for the user according to the question intention, and matching predicted questions for the user according to the question intention; and replying the answer and the predicted question to the user. By unifying a plurality of accounts of the user to the same account and acquiring the question information of the user in different applications through the unified account, a question operation process with simple operation is provided for the user, the answer of the question of the user can be replied, and the use experience of the user is improved. The application also provides a customer service response device.
Description
Technical Field
The application relates to an automatic customer service technology, in particular to a customer service response method. The application also provides a customer service response device.
Background
Currently, the mobile internet has been emerging to rapidly transfer service contents such as user life, work, etc. to the online, and users have obtained different types of convenient services through various different service terminals and enjoy benefits brought by the same. However, with the development of online service, due to data isolation between service terminals with different function types, a user needs to correspondingly open different applications when performing customer service consultation on each application, and in some scenarios, the user may need to consult a problem with an association relationship among multiple applications, and thus the existing application cannot provide a relevant reply.
In the prior art, a user has a problem of having a problem in multiple application terminals, and needs to consult and understand through multiple applications, so that the operation is complicated, and the user experience is not good because the user may not consult the associated problems in the multiple applications.
Content of application
The application provides a customer service response method, which aims to solve the problem of poor experience when detecting online consultation of a user. Simultaneously, this application still provides a customer service answering device.
The application provides a customer service response method, which comprises the following steps:
receiving question information sent by a user in a sub-account of a primary account through the primary account;
identifying a user question intention according to the question information and the sending interface of the user question information;
matching answers corresponding to the question information for the user according to the question intention, and matching predicted questions for the user according to the question intention;
and replying the answer and the predicted question to the user.
Optionally, the identifying the user question intention includes:
processing the question information through an NLP natural language to obtain a formatted language;
identifying the user intent from the formatting language.
Optionally, the matching the answer according to the formatting language includes:
calling a pre-trained answer matching model;
and inputting the user intention into the answer matching model to obtain the answer.
Optionally, the NPL natural language processing further includes:
and judging the emotion of the user according to the question information.
Optionally, the answer is stored in a knowledge base.
The present application further provides a customer service response device, including:
the receiving module is used for receiving question information sent by a user in a sub-account of a primary account through the primary account;
the identification module is used for identifying the user question intention according to the question information and the sending interface of the user question information;
the matching module is used for matching answers corresponding to the question information for the user according to the question intention and matching predicted questions for the user according to the question intention;
and the reply module is used for replying the answer and the predicted question to the user.
Optionally, the identification module further includes:
the conversion unit is used for processing the question information through an NLP natural language to obtain a formatted language;
identifying the user intent from the formatting language.
Optionally, the matching module further includes:
the calling unit is used for calling a pre-trained answer matching model;
and the processing unit is used for inputting the user intention into the answer matching model to obtain the answer.
Optionally, the NPL natural language processing further includes:
and judging the emotion of the user according to the question information.
Optionally, the answer is stored in a knowledge base.
Compared with the prior art, the application has the following advantages:
the application provides a customer service response method, which comprises the following steps: receiving question information sent by a user in a sub-account of a primary account through the primary account; identifying a user question intention according to the question information and the sending interface of the user question information; matching answers corresponding to the question information for the user according to the question intention, and matching predicted questions for the user according to the question intention; and replying the answer and the predicted question to the user. By unifying a plurality of accounts of the user to the same account and acquiring the question information of the user in different applications through the unified account, a question operation process with simple operation is provided for the user, the answer of the question of the user can be replied, and the use experience of the user is improved.
Drawings
Fig. 1 is a flowchart of a customer service response method in the present application.
Fig. 2 is a schematic diagram of a customer service answering device in the present application.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be implemented in many ways other than those described herein, and it will be apparent to those skilled in the art that the present application may be practiced without departing from the spirit of the present application, and therefore the present application is not limited to the specific implementations disclosed below.
For explaining the technical scheme of the application in detail, the conception method of the application is described below with reference to the accompanying drawings.
Fig. 1 is a flowchart of a customer service response method in the present application.
Referring to fig. 1, in S101, a primary account receives question information issued by a user in a sub-account of the primary account;
in this application, the primary account refers to a unified account created by a user according to authorization of the user when the user owns one or more applications, and the accounts of the applications include: and the business unit accounts comprise a house sub-account, a financial sub-account, a big data sub-account and/or a partner sub-account. The primary account is a primary account of the business unit account owned by the user, and each business unit account of the user is a sub-account of the primary account.
In the application, the primary account and the sub-account have the same user information, and when the user performs application login on the application login port, the user is prompted to establish the sub-account. When the user establishes the primary account, this may log into the application through the primary account instead of the business unit account.
In this application, the primary account may obtain information in each sub-account of the user through authority control, but the sub-accounts may not obtain information of other sub-accounts of the user through the primary account.
And receiving messages sent by the sub-accounts of the primary account at the primary account, wherein the messages comprise question messages sent by users.
S102, identifying a user question intention according to the question information and the sending interface of the user question information;
in the application, the question information of the user is sent out through a sub-account of the user, and the sub-account is a service system application corresponding to a specific service, so that an interface for sending out the question information by the user is obtained, the current browsing position of the user can be identified, and the question range of the question information of the user is obtained. The issuing interface is an interface which is operated by a user at an application end, and is sent to the primary account when the user issues a question message.
Obtaining a question range of a user question according to the sending interface, specifically comprising: obtaining a question range keyword according to the issue interface, wherein the question range keyword comprises: the name of the service operated by the user, the name of the current operation step and the like; and screening the question intention of the user according to the keywords of the question range.
Then, obtaining a question intention of the user according to the question message of the user, wherein the question intention refers to the purpose of the user to ask a question, and comprises the following steps: the user issues the question as to which operation to do, or the user issues what choice to do with the question, etc.
Acquiring the user intention, firstly translating the question message into a language which can be understood by a machine, namely NPL (neutral network language) natural language processing, and translating the question message to obtain a formatted language which can be identified by the machine; and then acquiring the user intention according to the formatting language.
In the present application, the user intentions may be distinguished by classification tags. Therefore, when the user intention needs to be obtained, the classification label needs to be extracted according to the question message for determination.
S103, matching answers corresponding to the question information for the user according to the question intentions, and matching predicted questions for the user according to the question intentions;
and when the question message is acquired and the question intention is acquired according to the question message and the corresponding issuing interface of the question message, determining the question intention of the user according to the question message and the issuing interface.
Extracting a pre-trained answer matching model according to the questioning intention; and inputting the question intention into the answer matching model to obtain the answer, wherein the answer is stored in a knowledge base. Specifically, the pre-trained answer matching model is obtained through the following steps:
acquiring a plurality of question intentions of a user, desensitizing the question intentions, and numbering in sequence;
inputting the serial numbers into a convolutional neural network in sequence to obtain an answer matching result, and simultaneously performing manual matching according to the question message to obtain a manual matching result;
comparing the matching result with the artificial matching result to obtain a result error, and correcting the parameter of the convolutional neural network according to the error;
when the error is smaller than a preset error threshold, the answer matching model can be used for answer matching.
Meanwhile, according to the question intention, other questions which the user wants to know can be further matched, for example, when the question intention has a keyword AABB, other predicted questions with the keyword AABB are obtained together, and the predicted questions refer to preset question information used by the user for asking questions.
And S104, replying the answer and the predicted question to the user.
After the answer and the predicted question corresponding to the question information asked by the user are obtained, the answer and the predicted question are respectively replied to a question interface of the user, the answer is used for answering the question information of the user, and the predicted question is used for providing a user with understanding directions for further understanding.
Furthermore, the emotion of the user can be judged according to the question information, and the tone of the answer can be replied according to the emotion planning of the user.
The present application further provides a customer service response device, including: a receiving module 101, an identifying module 102, a matching module 103 and a replying module 104.
Fig. 2 is a schematic diagram of a customer service answering device in the present application.
Referring to fig. 2, the receiving module 101 is configured to receive, through a primary account, question information issued by a user in a sub-account of the primary account;
in this application, the primary account refers to a unified account created by a user according to authorization of the user when the user owns one or more applications, and the accounts of the applications include: and the business unit accounts comprise a house sub-account, a financial sub-account, a big data sub-account and/or a partner sub-account. The primary account is a primary account of the business unit account owned by the user, and each business unit account of the user is a sub-account of the primary account.
In the application, the primary account and the sub-account have the same user information, and when the user performs application login on the application login port, the user is prompted to establish the sub-account. When the user establishes the primary account, this may log into the application through the primary account instead of the business unit account.
In this application, the primary account may obtain information in each sub-account of the user through authority control, but the sub-accounts may not obtain information of other sub-accounts of the user through the primary account.
And receiving messages sent by the sub-accounts of the primary account at the primary account, wherein the messages comprise question messages sent by users.
The identification module 102 is used for identifying the user question intention according to the question information and the sending interface of the user question information;
in the application, the question information of the user is sent out through a sub-account of the user, and the sub-account is a service system application corresponding to a specific service, so that an interface for sending out the question information by the user is obtained, the current browsing position of the user can be identified, and the question range of the question information of the user is obtained. The issuing interface is an interface which is operated by a user at an application end, and is sent to the primary account when the user issues a question message.
Obtaining a question range of a user question according to the sending interface, specifically comprising: obtaining a question range keyword according to the issue interface, wherein the question range keyword comprises: the name of the service operated by the user, the name of the current operation step and the like; and screening the question intention of the user according to the keywords of the question range.
Then, obtaining a question intention of the user according to the question message of the user, wherein the question intention refers to the purpose of the user to ask a question, and comprises the following steps: the user issues the question as to which operation to do, or the user issues what choice to do with the question, etc.
Acquiring the user intention, firstly translating the question message into a language which can be understood by a machine, namely NPL (neutral network language) natural language processing, and translating the question message to obtain a formatted language which can be identified by the machine; and then acquiring the user intention according to the formatting language.
In the present application, the user intentions may be distinguished by category labels. Therefore, when the user intention needs to be obtained, the classification label needs to be extracted according to the question message for determination.
The matching module 103 is configured to match answers corresponding to the quiz information for the user according to the quiz intention, and match predicted questions for the user according to the quiz intention;
and when the question message is acquired and the question intention is acquired according to the question message and the sending interface corresponding to the question message, determining the question intention of the user according to the question message and the sending interface.
Extracting a pre-trained answer matching model according to the questioning intention; and inputting the question intention into the answer matching model to obtain the answer, wherein the answer is stored in a knowledge base. Specifically, the pre-trained answer matching model is obtained through the following steps:
acquiring a plurality of question intentions of a user, desensitizing the question intentions, and numbering in sequence;
sequentially inputting the serial numbers into a convolutional neural network to obtain answer matching results, and simultaneously performing manual matching according to the question information to obtain manual matching results;
comparing the matching result with the artificial matching result to obtain a result error, and correcting the parameter of the convolutional neural network according to the error;
when the error is smaller than a preset error threshold, the answer matching model can be used for answer matching.
Meanwhile, according to the question intention, other questions which the user wants to know can be further matched, for example, when the question intention has a keyword AABB, other predicted questions with the keyword AABB are obtained together, and the predicted questions refer to preset question information used by the user for asking questions.
And a reply module 104 for replying the answer and the predicted question to the user.
After the answer and the predicted question corresponding to the question information asked by the user are obtained, the answer and the predicted question are respectively replied to a question interface of the user, the answer is used for answering the question information of the user, and the predicted question is used for providing a user with understanding directions for further understanding.
Furthermore, the emotion of the user can be judged according to the question information, and the tone of the answer can be replied according to the emotion planning of the user.
Claims (10)
1. A customer service response method, comprising:
receiving question information sent by a user in a sub-account of a primary account through the primary account;
identifying a user question intention according to the question information and the sending interface of the user question information;
matching answers corresponding to the question information for the user according to the question intention, and matching predicted questions for the user according to the question intention;
and replying the answer and the predicted question to the user.
2. The customer service response method according to claim 1, wherein the identifying the user's questioning intent comprises:
processing the question information through an NLP natural language to obtain a formatted language;
identifying the user intent from the formatting language.
3. The customer service response method of claim 2, wherein said matching said answer according to said formatted language comprises:
calling a pre-trained answer matching model;
and inputting the user intention into the answer matching model to obtain the answer.
4. The customer service response method according to claim 2, wherein the NPL natural language processing further comprises:
and judging the emotion of the user according to the question information.
5. The customer service response method of claim 1, wherein the answers are stored in a knowledge base.
6. A customer service response apparatus, comprising:
the receiving module is used for receiving question information sent by a user in a sub-account of a primary account through the primary account;
the identification module is used for identifying the user question intention according to the question information and the sending interface of the user question information;
the matching module is used for matching answers corresponding to the question information for the user according to the question intention and matching predicted questions for the user according to the question intention;
and the reply module is used for replying the answer and the predicted question to the user.
7. The customer service answering device according to claim 6, wherein the identification module further comprises:
the conversion unit is used for processing the question information through an NLP natural language to obtain a formatted language;
identifying the user intent from the formatting language.
8. The customer service response unit of claim 7, wherein the matching module further comprises:
the calling unit is used for calling a pre-trained answer matching model;
and the processing unit is used for inputting the user intention into the answer matching model to obtain the answer.
9. The customer service response device of claim 7 wherein the NPL natural language processing further comprises:
and judging the emotion of the user according to the question information.
10. The customer service response unit of claim 6 wherein the answers are stored in a knowledge base.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210151129.7A CN114637828A (en) | 2022-02-14 | 2022-02-14 | Customer service response method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210151129.7A CN114637828A (en) | 2022-02-14 | 2022-02-14 | Customer service response method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114637828A true CN114637828A (en) | 2022-06-17 |
Family
ID=81945999
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210151129.7A Pending CN114637828A (en) | 2022-02-14 | 2022-02-14 | Customer service response method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114637828A (en) |
-
2022
- 2022-02-14 CN CN202210151129.7A patent/CN114637828A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109672786B (en) | Incoming call answering method and device | |
CN111192060B (en) | Full-channel self-service response implementation method based on power IT service | |
CN110266899B (en) | Client intention identification method and customer service system | |
WO2016201815A1 (en) | Method and apparatus for providing online customer service | |
US7248678B2 (en) | Methods and apparatus for routing and accounting of revenue generating calls using natural language voice recognition | |
CN112417158A (en) | Training method, classification method, device and equipment of text data classification model | |
CN111246008A (en) | Method, system and device for realizing telephone assistant | |
CN113840040B (en) | Man-machine cooperation outbound method, device, equipment and storage medium | |
CN110321414B (en) | Artificial intelligence consultation service method and system based on deep learning | |
CN111611365A (en) | Flow control method, device, equipment and storage medium of dialog system | |
CN111291157A (en) | Response method, response device, terminal and storage medium | |
CN111563037B (en) | Test optimization method and device based on test questions, storage medium and terminal | |
CN110047473B (en) | Man-machine cooperative interaction method and system | |
CN114637828A (en) | Customer service response method and device | |
WO2023090380A1 (en) | Program, information processing system, and information processing method | |
CN110798576A (en) | Incoming call identity recognition method and device and related equipment | |
CN107832342B (en) | Robot chatting method and system | |
CN112860873B (en) | Intelligent response method, device and storage medium | |
US20190213554A1 (en) | User support system with automatic message categorization | |
CN113641814A (en) | Intelligent robot communication system, method and device based on 5G message | |
CN112633919A (en) | Method and system for realizing intelligent customer service | |
WO2020124444A1 (en) | Information processing method and related apparatus | |
TWM591212U (en) | Automatic customer service agent system | |
RU2722461C1 (en) | Voice robotic question-answer system and method of its automatic interaction with electronic device of user | |
CN111680141B (en) | One-to-many intelligent question-answering response method, device, equipment and readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |