CN110334194A - A kind of more wheel session marketing methods based on speech robot people - Google Patents
A kind of more wheel session marketing methods based on speech robot people Download PDFInfo
- Publication number
- CN110334194A CN110334194A CN201910492328.2A CN201910492328A CN110334194A CN 110334194 A CN110334194 A CN 110334194A CN 201910492328 A CN201910492328 A CN 201910492328A CN 110334194 A CN110334194 A CN 110334194A
- Authority
- CN
- China
- Prior art keywords
- session
- wheel
- speech
- data object
- robot people
- 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
- 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
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/16—Real estate
Abstract
The invention discloses a kind of more wheel session marketing methods based on speech robot people, including speech robot people take turns session more, and speech robot people issues the user with n wheel session request；Obtain the user speech information of n wheel；N customer requirement model is established by the user speech information that n takes turns；Each round session, which obtains objective number of labels and presets, obtains objective threshold value progress difference comparsion, and when the n-th wheel session, which obtains objective number of labels, meets condition, speech robot people completes marketing task, and speech robot people takes turns conversation end more.The present invention solves existing manual telephone system marketing and obtains that visitor is at high cost, obtains the technical issues of objective low efficiency.
Description
Technical field
The present invention relates to artificial intelligence field more particularly to a kind of more wheel session marketing methods based on speech robot people.
Background technique
Currently, computer technology has evolved to a highly developed level, artificial intelligence is the one of computer science
A branch, it attempts to understand essence of intelligence, and produces and a kind of new can make a response in such a way that human intelligence is similar
Intelligence machine, the research in the field include robot, language identification, image recognition, natural language processing and expert system etc..
Artificial speech industry is typical manpower-intensive type industry, and amount of flow is especially big, is that artificial intelligence is most possibly run
The job category for covering and replacing.There are a large amount of repetitive operations for black phone sales industry, and as demographic dividend disappears,
Human cost it is higher and higher, cut payroll to improve efficiency and also become the core demand of enterprise, with the fast development of domestic all trades and professions,
Competition between enterprise is also more fierce.
The major way that enterprise markets at present is to obtain visitor by telemarketing or network publicity, wherein with manpower
Cost increases, industry competition is more fierce, is also constantly increased by the objective cost that obtains that manual telephone system is sold, and compared to growth
Cost of labor obtains objective efficiency and does not get a promotion.
Summary of the invention
Obtain that visitor is at high cost, obtains the technical issues of objective low efficiency to solve existing manual telephone system marketing, the present invention devises one
More wheel session marketing methods of the kind based on speech robot people.
The present invention adopts the following technical scheme:
A kind of more wheel session marketing methods based on speech robot people, including speech robot people take turns session more, speech robot people to
User issues n and takes turns session request；Obtain the user speech information of n wheel；Establishing n user by the user speech information that n takes turns needs
Modulus type；Each round session, which obtains objective number of labels and presets, obtains objective threshold value progress difference comparsion, until the n-th wheel session obtains objective label
When quantity meets condition, speech robot people completes marketing task, and speech robot people takes turns conversation end more.
Preferably, the first round session request that speech robot human hair goes out is made friends for opening remarks and is proposed under preset themes
Questionnaire, obtain the user demand information of the first round, the user of the first round established by the user demand information of the first round
Demand model, speech robot people export farewell art and hang up.
Preferably, the step of establishing nth user's demand model is as follows: the number in the n-th wheel user speech information of hitting
According to object attribute values, data object attribute value is the characteristic information with mathematics numerical value in the n-th wheel user speech information；Extract number
According to data object, the data object keyword relevant to data object before object attribute values, after extracting data object attribute value
Data object numerical value unit；It is crucial according to data object and data object attribute value, data object numerical value unit, data object
Relevance forms the data object model of the n-th wheel between word；N-th is formed by relevance between the data object model of n-th wheel
A customer requirement model.
Preferably, the statistical method for obtaining objective number of labels is as follows: speech robot people hits the number in user speech information
According to object key word, there are certainty word, then the data object keyword is denoted as and obtains objective label after data object keyword, this
Wheel session obtains objective number of labels and adds 1, described to obtain objective threshold value and obtain objective precision setting for market.
Preferably, each round session obtains objective number of labels and the default method for obtaining objective threshold value progress difference comparsion is as follows:
It obtains objective number of labels-and obtains objective threshold value >=0, speech robot people completes marketing task, and speech robot people takes turns conversation end more；Obtain objective mark
Label quantity-obtains objective threshold value < 0, and speech robot people does not complete marketing task, and speech robot people continues to initiate a session request until n-th
Objective number of labels-, which is obtained, in wheel session obtains objective threshold value >=0.
It matches to obtain by (n-1)th customer requirement model preferably, speech robot people n-th takes turns words art.
Preferably, the questionnaire includes the initial option of user information under preset themes, the user demand letter
Breath is generated according to user investigation questionnaire result.
Preferably, the initiation time of speech robot people's each round session is according to last round of session voice customer requirement model
It determines.
Preferably, speech robot people delivers user information to true man business personnel after the completion of speech robot people's marketing task
Follow-up.
Preferably, n >=2.
The beneficial effects of the present invention are: 1, speech robot people covers a city inhabitant with quasi- true manization liaison mode, lead to
It crosses intelligent sound and citizen carries out interaction, the first round session phone that speech robot human hair goes out is to make friends and propose to preset
Questionnaire under theme avoids harassing；2, the preset themes mostly in wheel session are arranged according to demand, such as house property, shopping, religion
Educate, theme can be set in the contents such as medical treatment, establish different user demands model according to different preset themes, realize the essence of marketing
Standard is launched；3, speech robot people has intention citizen by mostly wheel session delineation, and worksheet processing implements its demand to true man business personnel；4,
True man business personnel by artificial intelligence management go precisely it is high-end implement Citizen needs, feedback result continues to manage to speech robot people.
Detailed description of the invention
Fig. 1 is the flow chart that speech robot people of the present invention takes turns session more；
Fig. 2 is the flow chart that the present invention establishes nth user's demand model；
Fig. 3 is the flow chart that speech robot people statistics of the present invention obtains objective number of labels.
Specific embodiment
Below by specific embodiment, and in conjunction with attached drawing, technical scheme of the present invention is further described:
Embodiment: in conjunction with shown in attached drawing 1, a kind of more wheel session marketing methods based on speech robot people, including speech robot people
More wheel sessions, speech robot people issue the user with n wheel session request；Obtain the user speech information of n wheel；The user taken turns by n
Voice messaging establishes n customer requirement model；Each round session, which obtains objective number of labels and presets, obtains objective threshold value progress difference comparsion,
When the n-th wheel session, which obtains objective number of labels, meets condition, speech robot people completes marketing task, and speech robot people takes turns session more
Terminate.
The first round session request that speech robot human hair goes out makes friends and proposes that the investigation under preset themes is asked for opening remarks
Volume, obtains the user speech information of the first round, the customer requirement model of the first round is established by the user speech information of the first round,
Speech robot people exports farewell art and hangs up, and it is art if greeting, saying hello and introduce myself, phonetic machine that opening remarks, which are made friends,
Device people is greeted with quasi- true manization voice and user, and first round words art is to introduce myself and based on user makes friends, such as
The such greeting mode of " you are good ", " today, weather was very good " and the self-introduction about company, the first round talk about the purpose of art
It is to introduce myself and make friends to user.
First round opening remarks are made friends, and words are postoperative, and speech robot people proposes that the questionnaire under preset themes obtains the first round
User information, preset themes are pre-set designated key, are illustrated by house property of preset themes, in first round meeting
In words request, speech robot people self-introduction content includes house property content, and the questionnaire that speech robot people proposes includes user
Initial option to house property information, including the rigid condition such as position, price, area, mating, position is either what some was determined
Cell, location are also possible to certain block region in map, and mating house property includes educational conditions, medical condition, sport condition, purchase
Object condition, transportation condition etc., area, price are price range, and the house property demand information of user is asked according to user's House property inquiry
It is generated when clove hitch fruit.
The flow chart for establishing nth user's demand model for the present invention as shown in Fig. 2 passes through the n-th wheel user speech letter
The step of breath establishes nth user's demand model is as follows:
Data object attribute value in S1, the n-th wheel user speech information of hitting, data object attribute value are the n-th wheel user user
With the characteristic information of mathematics numerical value in voice messaging；
S2, the data object before data object attribute value, data object keyword relevant to data object are extracted, described in extraction
Data object numerical value unit after data object attribute value；
S3, according to relevance between data object and data object attribute value, data object numerical value unit, data object keyword
Form the house property data object model of the n-th wheel；
Pass through relevance between S4, the n-th data object model taken turns and forms nth user's demand model.
It establishes customer requirement model as house property using preset themes to be illustrated, in each round session, speech robot people is cut
Data object attribute value in middle user speech information, data object attribute value are to have in current round session subscriber voice messaging
There is the characteristic information of mathematics numerical value, presetting data object in house property theme can be some specific place, such as user speech
Information includes " within X kilometers of WanDa Plaza ", and speech robot people hits data object attribute value " X ", while extracting the number
According to object " WanDa Plaza ", the numerical value unit " kilometer " of the data object, the data object keyword " distance ", data object
It may be some specific price, such as user speech information includes " it is desirable that house price is in Y member or so ", speech robot people
Data object attribute value of hitting " Y " extracts the numerical value unit " member " of the data object " house price ", data object, the data
The keyword of object " house price " is that " price " name of data object occurs in the present invention before data object attribute value
Claim and keyword relevant to data object, the quantization unit of data object occurs behind data object attribute value, to house property valence
For the data objects such as lattice, area, position, the data object attribute value of house property is generally some specific value, such as " member ",
" Wan Yuan ", " km ", " rice ", " square " etc., the title of data object also include that there are many expression ways, such as the area of house property
It can be expressed as " size " or " living space " etc., the height and position of house property also can be described as " in several buildings " or " floor is more
It is few " or " high level " or " low layer " etc., according to the data object of current round session and data object attribute value, data object number
Relevance forms the data object model of current round session between value unit, data object keyword, data object model
Quantity is usually a plurality of, forms customer requirement model by relevance between the data object model of current round session.
The flow chart for obtaining objective number of labels for speech robot people of the present invention statistics as shown in Fig. 3, obtains objective number of labels
Statistical method is as follows: speech robot people hits the data object keyword in user speech information, deposits after data object keyword
In certainty word, then the data object keyword is denoted as and obtains objective label, and epicycle session obtains objective number of labels and adds 1, described to obtain objective threshold
Value, which is used to market, obtains the precision setting of visitor, and each round session obtains objective number of labels and the default side for obtaining objective threshold value and carrying out difference comparsion
Method is as follows: obtaining objective number of labels-and obtains objective threshold value >=0, speech robot people completes marketing task, and speech robot people takes turns session knot more
Beam；It obtains objective number of labels-and obtains objective threshold value < 0, speech robot people does not complete marketing task, and speech robot people, which continues initiation session, to be asked
It asks and obtains objective threshold value >=0 until obtaining objective number of labels-in the n-th wheel session.
In the embodiment of the present invention, speech robot people matches suitable house property according to last round of customer requirement model and pushes away to user
It recommends, obtains objective label record epicycle session and hit the number of last round of customer requirement model, by obtaining objective number of labels and obtaining objective threshold
Whether the house property that the difference comparsion of value judges that this is recommended meets user demand, when epicycle session subscriber obtains objective number of labels >=obtain
The data object keyword of objective threshold value, i.e. epicycle session hit reaches pre-set when obtaining objective threshold value, judges user to this
The house property of recommendation is satisfied, and speech robot people takes turns conversation end more, if user obtains objective number of labels < and obtains objective threshold value, i.e. epicycle session
The data object keyword of hit be not up to it is pre-set obtain objective threshold value, it is full to judge that user fails this house property recommended
Meaning, speech robot people wait next round session to user according to epicycle session establishment customer requirement model and the suitable house property of reselection
To recommend, data object keyword is keyword relevant to data object in user speech information, by taking preset themes house property as an example,
Data object keyword root is extracted according to data object, data object keyword can for " position, floor, cell, price,
A series of house properties such as commercial circle, school district, greening, property, security " relevant key words exist " very after data object keyword
It is good ", " suitable ", then the data object keyword is denoted as and obtains objective label for the certainties such as " good " or the property appreciated word, statistics is each
Objective number of labels is obtained in wheel session, obtaining objective number of labels and presetting for each round session obtains objective threshold value progress difference comparsion, when
When obtaining objective number of labels >=obtain objective threshold value of a certain wheel session, speech robot people completes marketing task, obtains objective threshold value according to presetting
Theme obtains objective precision and is independently arranged, when obtain objective threshold value it is higher when, need it is matched to obtain objective number of labels higher, obtain at this time
Objective precision is also higher, and speech robot conference round can also be increase accordingly；When obtain objective threshold value it is lower when, need matched to obtain objective mark
It is lower to sign quantity, obtains that objective precision is lower, the session round that speech robot people initiates also can be reduced accordingly.
Speech robot people n-th takes turns words art and matches to obtain by (n-1)th customer requirement model, for applying in house property,
If the data object in (n-1) wheel customer requirement model includes " room rate ", " low layer ", " green coverage ", " subway ", " ten thousand
Up to square ", the n-th wheel words art stresses that the low layer house property close to WanDa Plaza subway is selected to market, and green coverage, price
The requirement in customer requirement model should be met.
The initiation time of speech robot people's each round session is determining according to last round of session voice customer requirement model, voice
After the completion of robot marketing task, speech robot people delivers user information and follows up to true man business personnel, in the present embodiment, works as language
The house property that sound robot is recommended obtains user's approval, i.e. when obtaining objective label >=obtain objective threshold value of the round, judges speech robot people
Estate sales task is completed, and user information is handover to true man business personnel and carries out depth follow-up.
Above-mentioned embodiment is only a preferred solution of the present invention, not the present invention is made in any form
Limitation, there are also other variations and modifications on the premise of not exceeding the technical scheme recorded in the claims.
Claims (10)
1. a kind of more wheel session marketing methods based on speech robot people, including speech robot people take turns session more, characterized in that language
Sound robot issues the user with n wheel session request；Obtain the user speech information of n wheel；It is established by the user speech information of n wheel
N customer requirement model；Each round session, which obtains objective number of labels and presets, obtains objective threshold value progress difference comparsion, until the n-th wheel meeting
When words obtain objective number of labels and meet condition, speech robot people completes marketing task, and speech robot people takes turns conversation end more.
2. a kind of more wheel session marketing methods based on speech robot people according to claim 1, characterized in that phonetic machine
The first round session request that device human hair goes out makes friends and proposes the questionnaire under preset themes for opening remarks, obtains the first round
User demand information establishes the customer requirement model of the first round, speech robot people output by the user demand information of the first round
Farewell art is simultaneously hung up.
3. a kind of more wheel session marketing methods based on speech robot people according to claim 1, characterized in that establish
The step of n customer requirement model, is as follows: the data object attribute value in the n-th wheel user speech information of hitting, data object category
Property value be n-th wheel user speech information in mathematics numerical value characteristic information；Data pair before extracting data object attribute value
Data object numerical value unit as, data object keyword relevant to data object, after extracting data object attribute value；According to
Data object and data object attribute value, data object numerical value unit, relevance forms the n-th wheel between data object keyword
Data object model；Nth user's demand model is formed by relevance between the data object model of n-th wheel.
4. a kind of more wheel session marketing methods based on speech robot people according to claim 3, characterized in that obtain objective mark
The statistical method for signing quantity is as follows: speech robot people hits the data object keyword in user speech information, and data object closes
There are certainty word, then the data object keyword is denoted as and obtains objective label after keyword, and epicycle session obtains objective number of labels and adds 1, institute
It states and obtains objective threshold value for the precision setting for obtaining visitor of marketing.
5. a kind of more wheel session marketing methods based on speech robot people according to claim 4, characterized in that each round
Session obtain objective number of labels and it is default obtain objective threshold value carry out difference comparsion method it is as follows: obtain objective number of labels-obtain objective threshold value >=
0, speech robot people completes marketing task, and speech robot people takes turns conversation end more；It obtains objective number of labels-and obtains objective threshold value < 0, voice
Robot does not complete marketing task, and speech robot people continues to initiate a session request to be obtained until obtaining objective number of labels-in the n-th wheel session
Objective threshold value >=0.
6. a kind of more wheel session marketing methods based on speech robot people according to claim 3, characterized in that phonetic machine
Device people n-th takes turns words art and matches to obtain by (n-1)th customer requirement model.
7. a kind of more wheel session marketing methods based on speech robot people according to claim 6, characterized in that the tune
The initial option that volume includes user information under preset themes is interrogated, the user demand information is raw according to user investigation questionnaire result
At.
8. a kind of more wheel session marketing methods carried out based on speech robot people according to claim 1, characterized in that language
The initiation time of sound robot each round session determines according to last round of session voice customer requirement model.
9. a kind of more wheel session marketing methods carried out based on speech robot people according to claim 1, characterized in that language
After the completion of sound robot marketing task, speech robot people delivers user information to true man business personnel follow-up.
10. a kind of more wheel session marketing methods based on speech robot people according to claim 1, characterized in that the n
≥2。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910492328.2A CN110334194A (en) | 2019-06-06 | 2019-06-06 | A kind of more wheel session marketing methods based on speech robot people |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910492328.2A CN110334194A (en) | 2019-06-06 | 2019-06-06 | A kind of more wheel session marketing methods based on speech robot people |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334194A true CN110334194A (en) | 2019-10-15 |
Family
ID=68140813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910492328.2A Pending CN110334194A (en) | 2019-06-06 | 2019-06-06 | A kind of more wheel session marketing methods based on speech robot people |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334194A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111159501A (en) * | 2019-11-22 | 2020-05-15 | 杭州蛋壳商务信息技术有限公司 | Method for establishing passenger judging model based on multilayer neural network and passenger judging method |
| CN112613727A (en) * | 2020-12-18 | 2021-04-06 | 深圳市思为软件技术有限公司 | Customer service expansion method and related equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760128A (en) * | 2011-04-26 | 2012-10-31 | 华东师范大学 | Telecommunication field package recommending method based on intelligent customer service robot interaction |
| CN107566670A (en) * | 2017-09-05 | 2018-01-09 | 南京硅基智能科技有限公司 | The method that autonomous dialing, Intelligent dialogue and information analysis are realized based on speech recognition |
| CN107609058A (en) * | 2017-08-25 | 2018-01-19 | 维沃移动通信有限公司 | A kind of marketing method, robot customer side, artificial customer side and user terminal |
| CN107783977A (en) * | 2016-08-24 | 2018-03-09 | 阿里巴巴集团控股有限公司 | Resource object information recommendation method, client and system |
| CN108521525A (en) * | 2018-04-03 | 2018-09-11 | 南京甄视智能科技有限公司 | Intelligent robot customer service marketing method and system based on user tag system |
-
2019
- 2019-06-06 CN CN201910492328.2A patent/CN110334194A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760128A (en) * | 2011-04-26 | 2012-10-31 | 华东师范大学 | Telecommunication field package recommending method based on intelligent customer service robot interaction |
| CN107783977A (en) * | 2016-08-24 | 2018-03-09 | 阿里巴巴集团控股有限公司 | Resource object information recommendation method, client and system |
| CN107609058A (en) * | 2017-08-25 | 2018-01-19 | 维沃移动通信有限公司 | A kind of marketing method, robot customer side, artificial customer side and user terminal |
| CN107566670A (en) * | 2017-09-05 | 2018-01-09 | 南京硅基智能科技有限公司 | The method that autonomous dialing, Intelligent dialogue and information analysis are realized based on speech recognition |
| CN108521525A (en) * | 2018-04-03 | 2018-09-11 | 南京甄视智能科技有限公司 | Intelligent robot customer service marketing method and system based on user tag system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111159501A (en) * | 2019-11-22 | 2020-05-15 | 杭州蛋壳商务信息技术有限公司 | Method for establishing passenger judging model based on multilayer neural network and passenger judging method |
| CN111159501B (en) * | 2019-11-22 | 2023-09-22 | 杭州蛋壳商务信息技术有限公司 | Method for establishing passenger judgment model based on multilayer neural network and passenger judgment method |
| CN112613727A (en) * | 2020-12-18 | 2021-04-06 | 深圳市思为软件技术有限公司 | Customer service expansion method and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108521525A (en) | Intelligent robot customer service marketing method and system based on user tag system | |
| CN109658928A (en) | A kind of home-services robot cloud multi-modal dialog method, apparatus and system | |
| CN110347839A (en) | A kind of file classification method based on production multi-task learning model | |
| CN106683678A (en) | Artificial telephone customer service auxiliary system and method | |
| CN105872185A (en) | Information prompting method, device and system | |
| CN112184497B (en) | Customer visit track tracking and passenger flow analysis system and method | |
| CN109919641A (en) | A kind of advertisement placement method and platform | |
| CN105427855A (en) | Voice broadcast system and voice broadcast method of intelligent software | |
| CN110334194A (en) | A kind of more wheel session marketing methods based on speech robot people | |
| CN106682137A (en) | Intelligent stock investment adviser questioning-answering interaction method and intelligent stock investment adviser questioning-answering interaction system | |
| CN107092692A (en) | The update method and intelligent customer service system of knowledge base | |
| CN110287299A (en) | Loquacity term sentence intelligent switch method in a kind of call | |
| CN110415086A (en) | Intelligence financing recommended method based on user's Continuous behavior sequence signature | |
| CN106991577A (en) | A kind of method and device for determining targeted customer | |
| CN109189902A (en) | The method and device of automatic-answering back device consulting | |
| CN114398512A (en) | Big data-based voice portrait analysis method for communication operator business customer | |
| CN109544248A (en) | A method of sales lead excavation is carried out based on internet information | |
| CN112235470A (en) | Incoming call client follow-up method, device and equipment based on voice recognition | |
| CN106874451A (en) | A kind of method of the personal exclusive corpus of automatic foundation | |
| CN113486166A (en) | Method, device and equipment for constructing intelligent customer service robot and storage medium | |
| CN116485450B (en) | Automatic demand identification method | |
| CN106844735A (en) | A kind of method of the personal exclusive corpus of automatic foundation | |
| CN116303978B (en) | Potential user mining method based on voice analysis | |
| CN107678545A (en) | A kind of information interactive terminal and method | |
| CN106844732A (en) | The method that automatic acquisition is carried out for the session context label that cannot directly gather |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191015 |