CN109087175A - The method, apparatus and system of customer service session switching - Google Patents
The method, apparatus and system of customer service session switching Download PDFInfo
- Publication number
- CN109087175A CN109087175A CN201810927997.3A CN201810927997A CN109087175A CN 109087175 A CN109087175 A CN 109087175A CN 201810927997 A CN201810927997 A CN 201810927997A CN 109087175 A CN109087175 A CN 109087175A
- Authority
- CN
- China
- Prior art keywords
- urgent
- customer service
- emergency
- state
- training
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0613—Third-party assisted
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Manipulator (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention relates to the method, apparatus and system of a kind of customer service session switching, comprising: receives the solicited message that client sends;The urgent Early-warning Model that training obtains in advance is sent by solicited message;Obtain the state of emergency of the solicited message of urgent Early-warning Model output;Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.Based on this, the urgent expression of client can effectively being identified, targetedly providing service according to urgent expression, can the unpredictable ill effect generated because that can not identify the urgent expression of client also more preferably can be effectively avoided in the experience of user.
Description
Technical field
The present invention relates to customer service robotic technology fields, and in particular to a kind of method, apparatus of customer service session switching and is
System.
Background technique
With the development of service trade, client increasingly focuses on obtaining the convenience of service, and people just have developed service machine
The scheme that device people and client are linked up accelerates the response speed of the query proposed to client with this.
Currently, customer service robot is typically all after the problem of receiving client, first using order models to the normal of candidate
See that the problem of problem (FAQ, Frequently Asked Questions) is ranked up, judges client is related to each FAQ
Property, the confidence level of the corresponding answer of FAQ then to make number one further according to the judgement of confidence level model.If made number one
Confidence level is sufficiently high, then the corresponding answer of FAQ to make number one is directly returned to client, otherwise, just recommends N possibility
Higher FAQ is independently selected for client.
Although the scheme of above-mentioned customer service robot and customer interaction can efficiently solve the problems, such as the part high frequency of client,
But when client encounters serious emergency case, when in problem along with some urgent expression, current interaction schemes are not
This urgent expression can be identified, also asking for client cannot be solved using more appropriate scheme according to urgent expression
The problem of inscribing, thus client being allowed to feel oneself is not properly settled, it is possible to meeting impaired user experience, or even bring one
A little unpredictable ill effects.
Summary of the invention
In view of this, it is an object of the invention to overcome the deficiencies of the prior art and provide a kind of sides of customer service session switching
Method, device and its system.
According to the embodiment of the present application in a first aspect, providing a kind of method of customer service session switching, comprising:
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent
Vocabulary and the training of urgent labeled data obtain;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The basis
The state of emergency access robot customer service or artificial customer service, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described
Urgent Early-warning Model.
Optionally, the solicited message includes scene information.
According to the second aspect of the embodiment of the present application, a kind of device of customer service session switching is provided, comprising:
Receiving module, for receiving the solicited message of client's transmission;
Sending module, for sending the urgent Early-warning Model that training obtains in advance for the solicited message;It is described urgent
Early-warning Model is obtained by urgent vocabulary and the training of urgent labeled data;
First obtains module, the state of emergency of the solicited message for obtaining the urgent Early-warning Model output;
AM access module, for accessing robot customer service or artificial customer service according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The access
Module includes:
First access unit accesses robot customer service if being the not state of emergency for the state of emergency;
Second access unit judges whether artificial customer service is busy if being the slight state of emergency for the state of emergency: when
When manually customer service is busy, robot customer service is accessed;When human customer is not busy, artificial customer service is accessed;
Third access unit accesses artificial customer service if being the severe state of emergency for the state of emergency.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
The present apparatus further include:
Second obtains module, for obtaining artificial customer service log;
Extraction module, for extracting the urgent vocabulary and the urgent labeled data from the artificial customer service log;
First training module, for using the urgent vocabulary as training data, the training antistop list to enhance model;
Second training module, for training the urgent identification model using the urgent labeled data as training data;
Third training module, for training the classification layer using the urgent labeled data as training data;
Link block, for by the antistop list enhance model and the urgent identification model respectively with the classification layer
Connection obtains the urgent Early-warning Model.
Optionally, the solicited message includes scene information.
According to the third aspect of the application, a kind of customer service session switching system is provided, comprising:
Processor, and the memory being connected with the processor;
For storing computer program, the computer program is at least used to execute customer service as described below the memory
The method of session switching:
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent
Vocabulary and the training of urgent labeled data obtain;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The basis
The state of emergency access robot customer service or artificial customer service, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described
Urgent Early-warning Model.
Optionally, the solicited message includes scene information.
The processor is for calling and executing the computer program in the memory.
According to the fourth aspect of the application, a kind of objective storage medium is provided, the storage medium is stored with computer program,
When the computer program is executed by processor, each step in the method for customer service session switching as described below is realized:
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent
Vocabulary and the training of urgent labeled data obtain;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The basis
The state of emergency access robot customer service or artificial customer service, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described
Urgent Early-warning Model.
Optionally, the solicited message includes scene information.
The invention adopts the above technical scheme, sends urgent Early-warning Model for the information that the client received sends, obtains
The state of emergency of the information sent to client accesses robot customer service or artificial customer service according to the state of emergency.Based on this,
Effectively to identify the urgent expression of client, service is targetedly provided according to urgent expression, the experience of user can more
It is good, it also can be effectively avoided because that can not identify the unpredictable ill effect generated due to client's urgent expression.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of the method for customer service session switching that the embodiment of the present invention one provides.
Fig. 2 is a kind of flow diagram of the method for customer service session switching provided by Embodiment 2 of the present invention.
Fig. 3 is a kind of composition schematic diagram for urgent Early-warning Model that the embodiment of the present invention one provides.
Fig. 4 is a kind of flow diagram of the training method for urgent Early-warning Model that the embodiment of the present invention one provides.
Fig. 5 is a kind of structural schematic diagram of the device for customer service session switching that the embodiment of the present invention three provides.
Fig. 6 is a kind of structural schematic diagram for AM access module that the embodiment of the present invention three provides.
Fig. 7 is a kind of structural schematic diagram for urgent Early-warning Model training device that the embodiment of the present invention three provides.
Fig. 8 is a kind of structural schematic diagram for customer service session switching system that the embodiment of the present invention four provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, technical solution of the present invention will be carried out below
Detailed description.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are obtained all without making creative work
Other embodiment belongs to the range that the present invention is protected.
Fig. 1 is a kind of flow diagram of the method for customer service session switching that the embodiment of the present invention one provides.
As shown in Figure 1, the method for the present embodiment includes:
Step 11 receives the solicited message that client sends.
The solicited message is sent the urgent Early-warning Model that training obtains in advance by step 12;The urgent early warning mould
Type is obtained by urgent vocabulary and the training of urgent labeled data.
Step 13, the state of emergency for obtaining the solicited message that the urgent Early-warning Model exports.
Step 14 accesses robot customer service or artificial customer service according to the grade of the state of emergency.
Since the information for sending the client received is sent to urgent Early-warning Model, the tight of the information of client's transmission is obtained
Anxious state accesses robot customer service or artificial customer service according to the state of emergency.Based on this, it can effectively identify that client's is tight
Anxious expression targetedly provides service according to urgent expression, the experience of user can more preferably, also can be effectively avoided because
The unpredictable ill effect that can not be identified the urgent expression of client and generate.
Wherein, in step 11, client send solicited message can with but be not limited only to be text information, can be according to client
The equipment of solicited message is inputted to adjust the type of the solicited message of client's transmission.For example, the input equipment as client is phone
When, solicited message can be set by the key behavior of client.So this programme can not only be applied to the customer service of mobile terminal
In session context, it can be applied in the scenes such as phone customer service, video customer service, corresponding solicited message can be set to electricity
Key behavior, voice messaging or the gesture information of key behavior, visual interface are talked about, or even can also be expression information.
In step 12, urgent Early-warning Model can be obtained by urgent vocabulary and the training of urgent labeled data.
Wherein, the structure of urgent Early-warning Model is as shown in Figure 3.Urgent Early-warning Model include antistop list enhancing model 31,
Urgent identification model 32 and classification layer 33, when the solicited message that client sends is text information, urgent Early-warning Model is received
After above-mentioned text information, antistop list enhancing model and urgent identification model can all be analyzed text information, antistop list
Enhancing model obtain the result is that keyword feature, it is that urgent identification model obtains the result is that in text information each word it is urgent
Weight feature, two results, which can be all input in classification layer, to be carried out whole and obtain complete characterization, and layer of finally classifying is according to obtaining
Text information is divided into not urgent, slight urgent or severe promptly and exported by complete characterization.
It should be noted that the training method of urgent Early-warning Model can with but be not limited only to method as described below, such as Fig. 4
Shown, the training method of urgent Early-warning Model may include:
Step 41 obtains artificial customer service log.
Step 42 extracts the urgent vocabulary and the urgent labeled data from the artificial customer service log.
Step 43, using the urgent vocabulary as training data, the training antistop list enhances model.
Step 44, using the urgent mark secretary as training data, the training urgent identification model.
Step 45, using the urgent labeled data as training data, the training classification layer.
Antistop list enhancing model and the urgent identification model are connect with the classification layer by step 46 respectively
To the urgent Early-warning Model.
In step 42, the method that urgent vocabulary and urgent labeled data are extracted from artificial customer service log can have more
Kind, it can be the method manually extracted, the method for being also possible to extract using the extraction model of training in advance.
In step 13, the state of emergency can be divided into three grades, i.e., not urgent, slight urgent and urgent three shapes of severe
State, certainly, when it is implemented, being not limited to the present embodiment these three described grades.
In step 14, robot customer service or artificial customer service are accessed according to the grade of the state of emergency, specifically access step
The step of being discussed further below.
As shown in Fig. 2, after obtaining the state of emergency of the solicited message of the urgent Early-warning Model output, according to tight
The grade access robot customer service of anxious state or artificial customer service may include:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
It should be noted that may include the scene information of session in the solicited message that client sends, scene information can be with
Keep urgent Early-warning Model more accurate to the judgement of the state of emergency.
In addition, Fig. 5 is a kind of structural schematic diagram for customer service session switching device that embodiments herein three provides.
A kind of customer service session switching device provided in this embodiment may include:
Receiving module 51, for receiving the solicited message of client's transmission;
Sending module 52, for sending the urgent Early-warning Model that training obtains in advance for the solicited message;It is described tight
Anxious Early-warning Model is obtained by urgent vocabulary and the training of urgent labeled data;
First obtains module 53, the state of emergency of the solicited message for obtaining the urgent Early-warning Model output;
AM access module 54, for accessing robot customer service or artificial customer service according to the grade of the state of emergency.
Wherein, the state of emergency may include the not state of emergency, the slight state of emergency and the severe state of emergency.As shown in fig. 6,
AM access module may include the first access unit 541, the second access unit 542 and third access unit 543.
Specifically, the first access unit accesses robot customer service if being the not state of emergency for the state of emergency;
Second access unit judges whether artificial customer service is busy if being the slight state of emergency for the state of emergency: when
When manually customer service is busy, robot customer service is accessed;When human customer is not busy, artificial customer service is accessed;
Third access unit accesses artificial customer service if being the severe state of emergency for the state of emergency.
It should be noted that urgent Early-warning Model may include antistop list enhancing model, urgent identification model and classification
Layer.It is corresponding, as shown in fig. 7, the present apparatus can also include:
Second obtains module 61, for obtaining artificial customer service log;
Extraction module 62, for extracting the urgent vocabulary and the urgent mark number from the artificial customer service log
According to;
First training module 63, for using the urgent vocabulary as training data, the training antistop list to enhance mould
Type;
Second training module 64, for training the urgent identification mould using the urgent labeled data as training data
Type;
Third training module 65, for training the classification layer using the urgent labeled data as training data;
Link block 66, for by the antistop list enhance model and the urgent identification model respectively with the classification
Layer connection obtains the urgent Early-warning Model.
In addition, the solicited message that client sends can also include the scene information of session, scene information can make urgent pre-
Alert model is more accurate to the judgement of the state of emergency.
Fig. 8 is a kind of structural schematic diagram for customer service session switching system that embodiments herein four provides.
As shown in figure 8, this system includes processor 71 and the memory 72 that is attached thereto.
Wherein, processor, and the memory being connected with the processor;
For storing computer program, the computer program is at least used to execute customer service as described below the memory
The method of session switching:
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent
Vocabulary and the training of urgent labeled data obtain;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The basis
The state of emergency access robot customer service or artificial customer service, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described
Urgent Early-warning Model.
Optionally, the solicited message includes scene information.
The processor is for calling and executing the computer program in the memory.
The application also provides a kind of storage medium, and the storage medium is stored with computer program, the computer program
When being executed by processor, each step in the method for customer service session switching as described below is realized:
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent
Vocabulary and the training of urgent labeled data obtain;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
Optionally, the state of emergency includes the not state of emergency, the slight state of emergency and the severe state of emergency;The basis
The state of emergency access robot customer service or artificial customer service, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judges whether artificial customer service is busy: when artificial customer service is busy, connecing
Enter robot customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
Optionally, the urgent Early-warning Model includes antistop list enhancing model, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described
Urgent Early-warning Model.
Optionally, the solicited message includes scene information.
It is understood that same or similar part can mutually refer in the various embodiments described above, in some embodiments
Unspecified content may refer to the same or similar content in other embodiments.
It should be noted that in the description of the present invention, term " first ", " second " etc. are used for description purposes only, without
It can be interpreted as indication or suggestion relative importance.In addition, in the description of the present invention, unless otherwise indicated, the meaning of " multiple "
Refer at least two.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, to execute function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
It should be appreciated that each section of the invention can be realized with hardware, software, firmware or their combination.Above-mentioned
In embodiment, software that multiple steps or method can be executed in memory and by suitable instruction execution system with storage
Or firmware is realized.It, and in another embodiment, can be under well known in the art for example, if realized with hardware
Any one of column technology or their combination are realized: having a logic gates for realizing logic function to data-signal
Discrete logic, with suitable combinational logic gate circuit specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
It, can also be in addition, each functional unit in each embodiment of the present invention can integrate in a processing module
It is that each unit physically exists alone, can also be integrated in two or more units in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized and when sold or used as an independent product in the form of software function module, also can store in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of method of customer service session switching characterized by comprising
Receive the solicited message that client sends;
The urgent Early-warning Model that training obtains in advance is sent by the solicited message;The urgent Early-warning Model is by urgent vocabulary
It is obtained with the training of urgent labeled data;
Obtain the state of emergency of the solicited message of the urgent Early-warning Model output;
Robot customer service or artificial customer service are accessed according to the grade of the state of emergency.
2. the method according to claim 1, wherein the state of emergency includes the not state of emergency, slight urgent
State and the severe state of emergency;It is described that robot customer service or artificial customer service are accessed according to the state of emergency, comprising:
If the state of emergency is the not state of emergency, robot customer service is accessed;
If the state of emergency is the slight state of emergency, judge whether artificial customer service is busy: when artificial customer service is busy, accessing machine
Device people's customer service;When human customer is not busy, artificial customer service is accessed;
If the state of emergency is the severe state of emergency, artificial customer service is accessed.
3. the method according to claim 1, wherein the urgent Early-warning Model includes antistop list enhancing mould
Type, urgent identification model and classification layer;
This method further include:
Obtain artificial customer service log;
The urgent vocabulary and the urgent labeled data are extracted from the artificial customer service log;
Using the urgent vocabulary as training data, the training antistop list enhances model;
Using the urgent labeled data as training data, the training urgent identification model;
Using the urgent labeled data as training data, the training classification layer;
Antistop list enhancing model and the urgent identification model are connect to obtain respectively with the classification layer described urgent
Early-warning Model.
4. the method according to claim 1, wherein the solicited message includes scene information.
5. a kind of device of customer service session switching characterized by comprising
Receiving module, for receiving the solicited message of client's transmission;
Sending module, for sending the urgent Early-warning Model that training obtains in advance for the solicited message;The urgent early warning
Model is obtained by urgent vocabulary and the training of urgent labeled data;
First obtains module, the state of emergency of the solicited message for obtaining the urgent Early-warning Model output;
AM access module, for accessing robot customer service or artificial customer service according to the grade of the state of emergency.
6. device according to claim 5, which is characterized in that the state of emergency includes the not state of emergency, slight urgent
State and the severe state of emergency;The AM access module includes:
First access unit accesses robot customer service if being the not state of emergency for the state of emergency;
Second access unit judges whether artificial customer service is busy: when artificial if being the slight state of emergency for the state of emergency
When customer service is busy, robot customer service is accessed;When human customer is not busy, artificial customer service is accessed;
Third access unit accesses artificial customer service if being the severe state of emergency for the state of emergency.
7. device according to claim 5, which is characterized in that the urgent Early-warning Model includes antistop list enhancing mould
Type, urgent identification model and classification layer;
The present apparatus further include:
Second obtains module, for obtaining artificial customer service log;
Extraction module, for extracting the urgent vocabulary and the urgent labeled data from the artificial customer service log;
First training module, for using the urgent vocabulary as training data, the training antistop list to enhance model;
Second training module, for training the urgent identification model using the urgent labeled data as training data;
Third training module, for training the classification layer using the urgent labeled data as training data;
Link block, for antistop list enhancing model and the urgent identification model to be connect with the classification layer respectively
Obtain the urgent Early-warning Model.
8. device according to claim 5, which is characterized in that the solicited message includes scene information.
9. a kind of system of customer service session switching characterized by comprising
Processor, and the memory being connected with the processor;
The memory is at least used for perform claim and requires any one of 1-4 for storing computer program, the computer program
The method of the customer service session switching;
The processor is for calling and executing the computer program in the memory.
10. a kind of storage medium, which is characterized in that the storage medium is stored with computer program, the computer program quilt
When processor executes, each step in the method for customer service session switching according to any one of claims 1-4 is realized.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810927997.3A CN109087175A (en) | 2018-08-15 | 2018-08-15 | The method, apparatus and system of customer service session switching |
PCT/CN2019/100288 WO2020034928A1 (en) | 2018-08-15 | 2019-08-13 | Method and system for switching customer service session, and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810927997.3A CN109087175A (en) | 2018-08-15 | 2018-08-15 | The method, apparatus and system of customer service session switching |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109087175A true CN109087175A (en) | 2018-12-25 |
Family
ID=64793538
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810927997.3A Pending CN109087175A (en) | 2018-08-15 | 2018-08-15 | The method, apparatus and system of customer service session switching |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109087175A (en) |
WO (1) | WO2020034928A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020034928A1 (en) * | 2018-08-15 | 2020-02-20 | 深圳追一科技有限公司 | Method and system for switching customer service session, and storage medium |
CN111526253A (en) * | 2020-03-09 | 2020-08-11 | 深圳追一科技有限公司 | Call control method, device, computer equipment and storage medium |
CN111565151A (en) * | 2020-04-27 | 2020-08-21 | 中国银行股份有限公司 | Customer service line routing method and device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105592237A (en) * | 2014-10-24 | 2016-05-18 | ***通信集团公司 | Method and apparatus for session switching, and intelligent customer service robot |
CN106844750A (en) * | 2017-02-16 | 2017-06-13 | 深圳追科技有限公司 | Emotion is pacified in a kind of robot based on customer service man-machine interaction method and system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160269883A1 (en) * | 2015-03-13 | 2016-09-15 | Kapali Eswaran | Automated Service Systems and Methods |
CN107590159A (en) * | 2016-07-08 | 2018-01-16 | 阿里巴巴集团控股有限公司 | The method and apparatus that robot customer service turns artificial customer service |
CN107979704A (en) * | 2017-12-01 | 2018-05-01 | 中国联合网络通信集团有限公司 | Queuing strategy, queuing system |
CN109087175A (en) * | 2018-08-15 | 2018-12-25 | 深圳追科技有限公司 | The method, apparatus and system of customer service session switching |
-
2018
- 2018-08-15 CN CN201810927997.3A patent/CN109087175A/en active Pending
-
2019
- 2019-08-13 WO PCT/CN2019/100288 patent/WO2020034928A1/en active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105592237A (en) * | 2014-10-24 | 2016-05-18 | ***通信集团公司 | Method and apparatus for session switching, and intelligent customer service robot |
CN106844750A (en) * | 2017-02-16 | 2017-06-13 | 深圳追科技有限公司 | Emotion is pacified in a kind of robot based on customer service man-machine interaction method and system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020034928A1 (en) * | 2018-08-15 | 2020-02-20 | 深圳追一科技有限公司 | Method and system for switching customer service session, and storage medium |
CN111526253A (en) * | 2020-03-09 | 2020-08-11 | 深圳追一科技有限公司 | Call control method, device, computer equipment and storage medium |
CN111526253B (en) * | 2020-03-09 | 2022-02-08 | 深圳追一科技有限公司 | Call control method, device, computer equipment and storage medium |
CN111565151A (en) * | 2020-04-27 | 2020-08-21 | 中国银行股份有限公司 | Customer service line routing method and device |
Also Published As
Publication number | Publication date |
---|---|
WO2020034928A1 (en) | 2020-02-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200301954A1 (en) | Reply information obtaining method and apparatus | |
WO2018201964A1 (en) | Processing method for session information, server, and computer readable storage medium | |
WO2019174072A1 (en) | Intelligent robot based training method and apparatus, computer device and storage medium | |
JP6604836B2 (en) | Dialog text summarization apparatus and method | |
US10803850B2 (en) | Voice generation with predetermined emotion type | |
US20080160487A1 (en) | Modularized computer-aided language learning method and system | |
CN104572852B (en) | The recommended method and device of resource | |
CN110020059A (en) | System and method for inclusiveness CAPTCHA | |
CN109947911B (en) | Man-machine interaction method and device, computing equipment and computer storage medium | |
CN109087175A (en) | The method, apparatus and system of customer service session switching | |
CN105912692A (en) | Intelligent voice dialogue method and apparatus | |
CN109493866A (en) | Intelligent sound box and its operating method | |
CN114757176A (en) | Method for obtaining target intention recognition model and intention recognition method | |
CN114328852B (en) | Text processing method, related device and equipment | |
JP7436077B2 (en) | Skill voice wake-up method and device | |
CN109815482A (en) | A kind of method, apparatus, equipment and the computer storage medium of news interaction | |
CN112231554A (en) | Search recommendation word generation method and device, storage medium and computer equipment | |
CN117349515A (en) | Search processing method, electronic device and storage medium | |
CN116775815B (en) | Dialogue data processing method and device, electronic equipment and storage medium | |
US20220207066A1 (en) | System and method for self-generated entity-specific bot | |
KR20210009885A (en) | Method, device and computer readable storage medium for automatically generating content regarding offline object | |
CN117319340A (en) | Voice message playing method, device, terminal and storage medium | |
WO2021240673A1 (en) | Conversation program, device, and method | |
CN115862794A (en) | Medical record text generation method and device, computer equipment and storage medium | |
CN109145084A (en) | Data processing method, data processing equipment and server |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181225 |
|
RJ01 | Rejection of invention patent application after publication |