CN110457679A - Construction method, device, computer equipment and the storage medium of user's portrait - Google Patents
Construction method, device, computer equipment and the storage medium of user's portrait Download PDFInfo
- Publication number
- CN110457679A CN110457679A CN201910583155.5A CN201910583155A CN110457679A CN 110457679 A CN110457679 A CN 110457679A CN 201910583155 A CN201910583155 A CN 201910583155A CN 110457679 A CN110457679 A CN 110457679A
- Authority
- CN
- China
- Prior art keywords
- target
- service
- text data
- user
- entity
- 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.)
- Granted
Links
- 238000010276 construction Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000004590 computer program Methods 0.000 claims description 27
- 241001269238 Data Species 0.000 claims description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
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
- G06Q30/02—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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Telephonic Communication Services (AREA)
Abstract
This application involves construction method, device, computer equipment and the storage mediums of a kind of user portrait.Method includes: to obtain service text data and the corresponding target user's mark of the service text data and target service type;Keyword identification is carried out to the service text data according to the target service type;If unidentified in the service text data arrive preset keyword, semantics recognition and Entity recognition are carried out to the service text data according to the target service type, obtain target semantic label and target text entity;According to target user mark, the target semantic label and target text entity building user's portrait.Through the embodiment of the present invention, the dimension of building user portrait is wider, and it is the accurate recommended products of client that user's portrait of building is more accurate, and then attendant can preferably be helped to carry out two time selling.
Description
Technical field
This application involves technical field of information processing, construction method, device, calculating more particularly to a kind of user portrait
Machine equipment and storage medium.
Background technique
User's portrait is a kind of effective tool delineated target user, contact user's demand and design direction, user's portrait
It is widely used in every field.For example, containing a large amount of valuable information in the dialogue of telemarketing, according to sale dialogue building
User's portrait can help sales force to be drawn a portrait according to user and carry out two time selling, be the accurate recommended products of target user.
Currently, the mode of building user's portrait is usually to use keyword identification technology.For example, client show to want to buy it is electronic
Vehicle, recognizing keyword therein is " electric vehicle ", then constructs user's portrait according to keyword " electric vehicle ".
But keyword is only identified from sale dialogue, the information of building user's portrait is less, is unfavorable for sales force's root
It is the accurate recommended products of target user according to user's portrait.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide one kind and be capable of providing more users portrait information, preferably
Help construction method, device, computer equipment and the storage medium of user's portrait of sales force's recommended products.
In a first aspect, the embodiment of the invention provides a kind of construction methods of user portrait, this method comprises:
Obtain service text data and the corresponding target user's mark of service text data and target service type;
Keyword identification is carried out to service text data according to target service type;
If unidentified in service text data arrive preset keyword, according to target service type to service textual data
According to semantics recognition and Entity recognition is carried out, target semantic label and target text entity are obtained;
According to target user's mark, target semantic label and target text entity building user's portrait.
Service text data includes multiple sub-services text datas in one of the embodiments,;
It is above-mentioned that semantics recognition and Entity recognition are carried out to service text data according to target service type, obtain target semanteme
Label and target text entity, comprising:
Semantics recognition and Entity recognition are carried out to each sub-services text data respectively, obtain the mesh of each sub-services text data
Mark semantic label and target text entity;
It is above-mentioned to be drawn a portrait according to target user's mark, target semantic label and target text entity building user, comprising:
The incidence relation between the target semantic label and target text entity of each sub-services text data is established respectively;
By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
It is above-mentioned in one of the embodiments, that semantics recognition and reality are carried out to service text data according to target service type
Body identification, comprising:
According to the corresponding relationship between preset type of service and identification model, mesh is chosen from multiple semantics recognition models
The corresponding target semanteme identification model of type of service is marked, and chooses target service type from multiple entity recognition models and corresponds to
Target entity identification model;
Semantics recognition is carried out to service text data using target semanteme identification model, obtains target semantic label;Wherein,
Semantic label is used to indicate the classification of concern information;
Entity recognition is carried out to service text data using target entity identification model, obtains target text entity;Wherein,
Text entities are used to indicate concern information.
Obtaining service text data and the corresponding target user's mark of service text data in one of the embodiments,
Before knowledge and target service type, this method further include:
Obtain the voice data in service process;
Convert voice data into service text data.
The corresponding target user's mark of above-mentioned acquisition service text data in one of the embodiments, comprising:
Receive target user's mark of input;Or,
Identify that target user identifies from voice data and/or service text data.
The corresponding target service type of above-mentioned acquisition service text data in one of the embodiments, comprising:
The selection instruction for obtaining type of service, choosing instruction includes that type of service identifies;
It is identified according to type of service and determines target service type.
This method in one of the embodiments, further include:
If identifying preset keyword in service text data, according to preset keyword and keyword label it
Between corresponding relationship, determine the corresponding keyword label of service text data;
By keyword label storage corresponding with target user's mark, building user's portrait.
Second aspect, the embodiment of the invention provides a kind of construction device of user portrait, which includes:
Text data obtains module, for obtaining service text data and the corresponding target user's mark of service text data
Know and target service type;
Keyword identification module, for carrying out keyword identification to service text data according to target service type;
Semantic and Entity recognition module, if arriving preset keyword, basis for unidentified in service text data
Target service type carries out semantics recognition and Entity recognition to service text data, obtains target semantic label and target text is real
Body;
First user, which draws a portrait, constructs module, for according to target user's mark, target semantic label and target text entity
Construct user's portrait.
Service text data includes multiple sub-services text datas in one of the embodiments,;
Semantic and Entity recognition module is specifically used for carrying out semantics recognition to each sub-services text data respectively and entity is known
Not, the target semantic label and target text entity of each sub-services text data are obtained;
First user draw a portrait building module, specifically for establish respectively each sub-services text data target semantic label and
Incidence relation between target text entity;By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
Semantic and Entity recognition module includes: in one of the embodiments,
Model chooses submodule, for according to the corresponding relationship between preset type of service and identification model, from multiple
The corresponding target semanteme identification model of target service type is chosen in semantics recognition model, and from multiple entity recognition models
Choose the corresponding target entity identification model of target service type;
Semantics recognition submodule is obtained for carrying out semantics recognition to service text data using target semanteme identification model
To target semantic label;Wherein, semantic label is used to indicate the classification of concern information;
Entity recognition submodule is obtained for carrying out Entity recognition to service text data using target entity identification model
To target text entity;Wherein, text entities are used to indicate concern information.
The device in one of the embodiments, further include:
Voice data obtains module, for obtaining the voice data in service process;
Text conversion module, for converting voice data into service text data.
Above-mentioned text data acquisition module includes: in one of the embodiments,
Target user identifies receiving submodule, target user's mark for receiving input;
Target user identifies identification submodule, for identifying that target is used from voice data and/or service text data
Family mark.
Above-mentioned text data acquisition module includes: in one of the embodiments,
Instruction acquisition submodule is chosen, the selection for obtaining type of service instructs, and choosing instruction includes type of service mark
Know;
Target service type determination module determines target service type for identifying according to type of service.
The device in one of the embodiments, further include:
Keyword label determining module, if for identifying preset keyword in service text data, according to pre-
If keyword and keyword label between corresponding relationship, determine the corresponding keyword label of service text data;
Second user portrait building module, for constructing user for keyword label storage corresponding with target user's mark
Portrait.
The third aspect, the embodiment of the invention provides a kind of computer equipments, including memory and processor, memory to deposit
Computer program is contained, processor is realized when executing the computer program such as the step of the above method.
Fourth aspect, the embodiment of the invention provides a kind of computer readable storage mediums, are stored thereon with computer journey
Sequence is realized when the computer program is executed by processor such as the step of the above method.
Construction method, device, computer equipment and the storage medium of above-mentioned user portrait, obtain service text data and
Service the corresponding target user's mark of text data and target service type;According to target service type to service text data into
The identification of row keyword;If unidentified in service text data arrive preset keyword, according to target service type to service
Text data carries out semantics recognition and Entity recognition, obtains target semantic label and target text entity;It is marked according to target user
Know, target semantic label and target text entity building user are drawn a portrait.Through the embodiment of the present invention, when constructing user's portrait,
Other than carrying out keyword identification, semantics recognition and Entity recognition are also carried out, to obtain target semantic label and target text
This entity keeps the dimension for constructing user's portrait wider, structure according to target semantic label and target text entity building user's portrait
The user's portrait built is more accurate, and then attendant can preferably be helped to carry out two time selling, accurately recommends to produce for client
Product.
Detailed description of the invention
Fig. 1 is the applied environment figure of the construction method of user's portrait in one embodiment;
Fig. 2 is the flow diagram of the construction method of user's portrait in one embodiment;
Fig. 3 is the flow diagram of semantics recognition and Entity recognition step in one embodiment;
Fig. 4 is the flow diagram of the construction method of user's portrait in another embodiment;
Fig. 5 is the structural block diagram of the construction device of user's portrait in one embodiment;
Fig. 6 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
The construction method of user's portrait provided by the present application, can be applied in application environment as shown in Figure 1.It is servicing
In the process, attendant is talked with by attendant by terminal 102.At the end of service, attendant with serviced
The chat script of personnel is stored into server 104, and server 104 is according to attendant and by the chat script structure of attendant
It builds by user's portrait of attendant.Wherein, terminal 102 is communicated with server 104 by network by network.Terminal
102 can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and portable wearable set
It is standby;Server 104 can be realized with the server cluster of the either multiple server compositions of independent server.
In one embodiment, as shown in Fig. 2, providing a kind of construction method of user's portrait, it is applied in this way
It is illustrated for server in Fig. 1, comprising the following steps:
Step 201, service text data and the corresponding target user's mark of service text data and target service are obtained
Type.
In the present embodiment, in service process, attendant with text chat can be used by attendant.It is tied in service
Shu Hou, attendant store chat script to server.When constructing user's portrait, server obtains service text data,
And obtain the corresponding target user's mark of service text data and target service type.Wherein, target user mark can be by
Title, code of attendant etc..For example, target user is identified as " Mr. Wang ".Target service type is used to indicate service type.
For example, target service type is purchase vehicle or target service type is house-purchase.The embodiment of the present invention does not make target service type
It limits, can be configured according to the actual situation in detail.
Step 202, keyword identification is carried out to service text data according to target service type.
In the present embodiment, server, can be according to target industry after getting service text data and target service type
Service type carries out keyword identification to service text data.Specifically, different passes is set for different types of service in advance
Keyword determines in service text data whether include the corresponding keyword of target service type after determining target service type.
Do not include the corresponding keyword of target type in text data if serviced, thens follow the steps 203.
Determine in service text data whether include that preset keyword can use Keywords matching mode.Specifically,
Subordinate sentence processing is carried out to service text data and word segmentation processing obtains multiple service words, calculate separately each service word and is preset
Keyword similarity, be greater than default similarity threshold in the similarity of at least one service word and preset keyword
When, determine that service text data includes preset keyword.If the similarity of multiple service words and preset keyword is equal
Less than default similarity threshold, it is determined that service text data does not include preset keyword.When calculating similarity, Ke Yiji
The Euclidean distance of word vectors is calculated, other modes can also be used, the embodiment of the present invention does not limit this in detail, can basis
Actual conditions are configured.
Step 203, if unidentified in service text data arrive preset keyword, according to target service type to clothes
Business text data carries out semantics recognition and Entity recognition, obtains target semantic label and target text entity.
In the present embodiment, if unidentified in service text data arrive preset keyword, to service text data
Semantics recognition and Entity recognition are carried out, target semantic label and target text entity are obtained.For example, including in service text data
" price of car be within 150,000 my acceptable " carries out the available mesh of semantics recognition to the service text data
It marks semantic label " price ", meanwhile, the available target text entity " 15 of Entity recognition is carried out to the service text data
Ten thousand ".In another example including " this has subway, a plurality of public bus network around the house " in service text data, to the service textual data
According to the progress available target semantic label " traffic " of semantics recognition, meanwhile, carrying out Entity recognition to the service text data can
To obtain target text entity " subway ", " public transport ".The embodiment of the present invention does not make target semantic label and target text entity
It limits, can be configured according to the actual situation in detail.
Step 204, according to target user's mark, target semantic label and target text entity building user's portrait.
In the present embodiment, after determining target user's mark, target semantic label and target text entity, target use is established
Corresponding relationship between family mark, target semantic label and target text entity, to construct user's portrait.For example, establishing mesh
Mark user identifier " Mr. Wang ", target semantic label " price ", the corresponding relationship between target text entity " 150,000 ", then it can be with
User's portrait of " Mr. Wang " is constructed, i.e. Mr. Wang's acceptable price in terms of purchasing vehicle is 150,000.In this way, just facilitating service people
Member carries out two time selling to by attendant according to user portrait, thus more accurate recommended products.
In the construction method of above-mentioned user's portrait, obtains service text data and the corresponding target of service text data is used
Family mark and target service type;Keyword identification is carried out to service text data according to target service type;If in service text
It is unidentified in notebook data to arrive preset keyword, then semantics recognition and reality are carried out to service text data according to target service type
Body identification, obtains target semantic label and target text entity;According to target user's mark, target semantic label and target text
Entity constructs user's portrait.Through the embodiment of the present invention, when constructing user's portrait, other than carrying out keyword identification, also
Carry out semantics recognition and Entity recognition, to obtain target semantic label and target text entity, according to target semantic label and
Target text entity constructs user's portrait, keeps the dimension for constructing user's portrait wider, and user's portrait of building is more accurate, in turn
It can preferably help attendant to carry out two time selling, be the accurate recommended products of client.
In another embodiment, what is involved is a kind of optional processes of building user's portrait for the present embodiment.Above-mentioned
On the basis of embodiment illustrated in fig. 2, service text data includes multiple sub-services text datas;
Above-mentioned steps 203 can specifically include: semantics recognition and Entity recognition are carried out to each sub-services text data respectively,
Obtain the target semantic label and target text entity of each sub-services text data.
In the present embodiment, it is generally the case that attendant and by the talk between attendant not only one take turns, wherein each
Wheel talk can obtain corresponding sub-services text data.When carrying out semantics recognition and Entity recognition, for each sub-services
Text data carries out semantics recognition and Entity recognition, and target semantic label and the target text for obtaining each sub-services text data are real
Body.
As shown in Figure 3, it carries out semantics recognition and Entity recognition can specifically include following steps:
Step 301, according to the corresponding relationship between preset type of service and identification model, from multiple semantics recognition models
The corresponding target semanteme identification model of middle selection target service type, and target service is chosen from multiple entity recognition models
The corresponding target entity identification model of type.
In the present embodiment, multiple semantics recognition models, the semantic mark that each semantics recognition model can identify are trained in advance
Label are different.Meanwhile multiple entity recognition models are preset, the text entities that each entity recognition model can identify are different.
Then, the correspondence between the semantic label according to needed for type of service and text entities setting type of service and semantics recognition model
Corresponding relationship between relationship and type of service and entity recognition model.
For example, train 5 semantics recognition models in advance, and 5 entity recognition models are provided with, according to business demand,
Corresponding relationship between type of service " purchase vehicle " and semantics recognition model 1 and entity recognition model 3, and setting service class are set
Corresponding relationship between type " house-purchase " and semantics recognition model 3 and entity recognition model 2.
After determining target service type, it is semantic that the corresponding target of target service type is chosen from multiple semantics recognition models
Identification model chooses the corresponding target entity identification model of target service type from multiple entity recognition models.For example, if
Target service type is purchase vehicle, then semantics recognition model 1 is chosen from multiple semantics recognition models, from multiple entity recognition models
Middle selection entity recognition model 3;If target service type is house-purchase, semantics recognition is chosen from multiple semantics recognition models
Model 2 chooses entity recognition model 2 from multiple entity recognition models.
Step 302, semantics recognition is carried out to service text data using target semanteme identification model, obtains target semanteme mark
Label;Wherein, semantic label is used to indicate the classification of concern information.
In the present embodiment, after determining the corresponding target semanteme identification model of target service type, using target semantics recognition
Model carries out semantics recognition to service text data, obtains target semantic label.Wherein, the class of semantic label instruction concern information
Not.
For example, service text data is " price of car be my acceptable " within 150,000, known using semanteme
Other model 1 carries out semantics recognition to the service text data, which is concerned with " 150,000 ", belonging to class
Not Wei " price ", therefore obtain target semantic label be " price ".In another example service text data is that " this has around the house
Subway, a plurality of public bus network " carries out semantics recognition, the service textual data to the service text data using semantics recognition model 2
What it is according to concern includes " subway " and " public transport ", belonging to classification be " traffic ", therefore obtain target semantic label as " traffic ".
Step 303, Entity recognition is carried out to service text data using target entity identification model, obtains target text reality
Body;Wherein, text entities are used to indicate concern information.
In the present embodiment, after determining the corresponding target entity identification model of target service type, known using target language entity
Other model carries out Entity recognition to service text data, obtains target text entity.Wherein, text entities are to pay close attention to information.
For example, service text data is " price of car be my acceptable " within 150,000, known using entity
Other model 3 carries out Entity recognition to the service text data, which is concerned with " 150,000 ", therefore obtains mesh
Marking text entities is " 150,000 ".In another example service text data is " this has subway, a plurality of public bus network around the house ",
Entity recognition is carried out to the service text data using entity recognition model 2, service text data concern includes " subway "
" public transport ", therefore obtaining target text entity is " subway " and " public transport ".
Semantics recognition and reality are carried out to service text data using target semanteme identification model and target entity identification model
Body identification, available more user informations provide letter so as to abundant user's portrait for the two time selling of attendant
Breath basis.
Above-mentioned steps 204 can specifically include: establish the target semantic label and target of each sub-services text data respectively
Incidence relation between text entities;By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
In the present embodiment, after identifying the target semantic label and target text entity of each sub-services text data, build
Found the incidence relation between the target semantic label and target text entity of each sub-services text data.For example, identifying sub- clothes
The target semantic label of business text data 1 is " price ", and target text entity is " 150,000 ";Identify sub-services text data
2 target semantic label is " exterior trim ", and target text entity is " skylight ";Then establish the association between " 150,000 " and " price "
Incidence relation between relationship and " exterior trim " and " skylight ".
It establishes after incidence relation, by the storage corresponding with target user's mark of multiple incidence relations, is then built into user's picture
Picture.For example, by between the incidence relation and " exterior trim " and " skylight " between " 150,000 " and " price " incidence relation and " king
Certain " storage is corresponded to, it is built into user's portrait of " Mr. Wang ".In this way, attendant can draw a portrait according to user carries out secondary pin
It sells, is the accurate recommended products of customer service, to improve sales achievement.
During above-mentioned building user portrait, service text data is split as multiple sub-services text datas, to each
Sub-services text data is identified, the target semantic label and target text entity of each sub-services text data are obtained, and is established
The incidence relation between the target semantic label and target text entity of each sub-services text data is respectively established respectively;By multiple passes
The storage corresponding with target user's mark of connection relationship, building user's portrait.Through the embodiment of the present invention, the incidence relation obtained is got over
More, the information of user's portrait is abundanter, thus user's portrait is just more accurate, and then can preferably help attendant to carry out
Two time selling.
In another embodiment, as shown in figure 4, the present embodiment what is involved is building user portrait a kind of optional mistake
Journey.On the basis of above-mentioned embodiment illustrated in fig. 2, it can specifically include following steps:
Step 401, the voice data in service process is obtained;Convert voice data into service text data.
In the present embodiment, in service process, attendant with conversation voice can be used by attendant.It is tied in service
Shu Hou, attendant store the voice data in service process to server.Then server can use ASR
(Automatic Speech Recognition, automatic speech recognition) technology, converts voice data into service text data,
In case subsequent builds user draws a portrait.
And it is possible to the processing that voice data is converted to service text data be carried out in a server, in another clothes
Business device carries out the building of user's portrait, and the embodiment of the present invention does not limit this in detail, can be configured according to the actual situation.
Step 402, target user's mark of input is received;Or, being identified from voice data and/or service text data
Target user's mark.
In the present embodiment, target user's mark of attendant's input can receive, for example, attendant is in storaged voice
When data or service text data, input customer name " Mr. Wang ", server receives target user and identifies " Mr. Wang ".It can also be from
It identifies that target user identifies in voice data, for example, the name that client introduces oneself is " Mr. Wang " during talk, then may be used
To identify that target user is identified as " Mr. Wang ".It can also identify that target user identifies from service text data, for example, right
It services text data and carries out Entity recognition, identify name " Mr. Wang " from service text data.How really the embodiment of the present invention to
The user identifier that sets the goal does not limit in detail, can be configured according to the actual situation.
Step 403, the selection instruction for obtaining type of service, choosing instruction includes that type of service identifies;According to type of service
It identifies and determines target service type.
In the present embodiment, server can show multiple types of service, and then attendant selects from multiple types of service
Target service type is taken, then server gets the selection instruction of type of service, according to the type of service for including in selection instruction
It identifies and determines target service type.
For example, multiple types of service, the attendant such as server display purchase vehicle, house-purchase, purchase insurance choose purchase vehicle, then take
Business device gets the selection instruction for choosing purchase vehicle, determines target service type for purchase vehicle according to instruction is chosen.
Step 404, keyword identification is carried out to service text data according to target service type.
Step 405, if preset keyword is identified in service text data, according to preset keyword and key
Corresponding relationship between word label determines the corresponding keyword label of service text data;By keyword label and target user
The corresponding storage of mark, building user's portrait.
In the present embodiment, the corresponding relationship between keyword and keyword label is preset, if in service textual data
Preset keyword is identified in, then the corresponding keyword label of service text data is determined according to corresponding relationship.
For example, service text data includes " wanting to buy electric vehicle ", preset key is identified in the service text data
Word " electric vehicle ", it is corresponding keyword label " new energy " due to presetting keyword " electric vehicle ", then it can determine service text
The corresponding keyword label of notebook data " wanting to buy electric vehicle " is " new energy ".The embodiment of the present invention is to preset keyword and key
Word label does not limit in detail, can be configured according to the actual situation.
Step 406, if unidentified in service text data arrive preset keyword, according to target service type to clothes
Business text data carries out semantics recognition and Entity recognition, obtains target semantic label and target text entity.
Step 407, according to target user's mark, target semantic label and target text entity building user's portrait.
During above-mentioned building user portrait, the voice data in service process is obtained, and convert voice data into
Service text data;Obtain target user's mark;Obtain target service type;According to target service type to service text data
Carry out keyword identification;If preset keyword is identified in service text data, according to preset keyword and key
Corresponding relationship between word label determines the corresponding keyword label of service text data;By keyword label and target user
The corresponding storage of mark, building user's portrait;If unidentified in service text data arrive preset keyword, according to target industry
Service type carries out semantics recognition and Entity recognition to service text data, obtains target semantic label and target text entity.It is logical
The embodiment of the present invention is crossed, user's portrait can be constructed according to keyword, it can also be real according to target semantic label and target text
Body constructs user's portrait, and the dimension of building user's portrait is wider, and the information for obtaining user's portrait is more, to make user's portrait more
To be accurate, and then attendant can preferably be helped to carry out two time selling.
It should be understood that although each step in the flow chart of Fig. 2-4 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-4
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 5, providing a kind of construction device of user's portrait, comprising:
Text data obtains module 501, uses for obtaining service text data and the corresponding target of service text data
Family mark and target service type;
Keyword identification module 502, for carrying out keyword identification to service text data according to target service type;
Semantic and Entity recognition module 503, if arriving preset keyword, root for unidentified in service text data
Semantics recognition and Entity recognition are carried out to service text data according to target service type, obtain target semantic label and target text
Entity;
First user, which draws a portrait, constructs module 504, for real according to target user's mark, target semantic label and target text
Body constructs user's portrait.
Service text data includes multiple sub-services text datas in one of the embodiments,;
Semantic and Entity recognition module is specifically used for carrying out semantics recognition to each sub-services text data respectively and entity is known
Not, the target semantic label and target text entity of each sub-services text data are obtained;
First user draw a portrait building module, specifically for establish respectively each sub-services text data target semantic label and
Incidence relation between target text entity;By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
Semantic and Entity recognition module includes: in one of the embodiments,
Model chooses submodule, for according to the corresponding relationship between preset type of service and identification model, from multiple
The corresponding target semanteme identification model of target service type is chosen in semantics recognition model, and from multiple entity recognition models
Choose the corresponding target entity identification model of target service type;
Semantics recognition submodule is obtained for carrying out semantics recognition to service text data using target semanteme identification model
To target semantic label;Wherein, semantic label is used to indicate the classification of concern information;
Entity recognition submodule is obtained for carrying out Entity recognition to service text data using target entity identification model
To target text entity;Wherein, text entities are used to indicate concern information.
The device in one of the embodiments, further include:
Voice data obtains module, for obtaining the voice data in service process;
Text conversion module, for converting voice data into service text data.
Above-mentioned text data acquisition module includes: in one of the embodiments,
Target user identifies receiving submodule, target user's mark for receiving input;
Target user identifies identification submodule, for identifying that target is used from voice data and/or service text data
Family mark.
Above-mentioned text data acquisition module includes: in one of the embodiments,
Instruction acquisition submodule is chosen, the selection for obtaining type of service instructs, and choosing instruction includes type of service mark
Know;
Target service type determination module determines target service type for identifying according to type of service.
The device in one of the embodiments, further include:
Keyword label determining module, if for identifying preset keyword in service text data, according to pre-
If keyword and keyword label between corresponding relationship, determine the corresponding keyword label of service text data;
Second user portrait building module, for constructing user for keyword label storage corresponding with target user's mark
Portrait.
The specific of construction device about user's portrait limits the construction method that may refer to draw a portrait above for user
Restriction, details are not described herein.Modules in the construction device of above-mentioned user portrait can be fully or partially through software, hard
Part and combinations thereof is realized.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment,
It can also be stored in a software form in the memory in computer equipment, execute the above modules in order to which processor calls
Corresponding operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is used to store the building data of user's portrait.The network interface of the computer equipment is used for and external end
End passes through network connection communication.A kind of construction method of user's portrait is realized when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 6, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, is stored in memory
Computer program, the processor perform the steps of when executing computer program
Obtain service text data and the corresponding target user's mark of service text data and target service type;
Keyword identification is carried out to service text data according to target service type;
If unidentified in service text data arrive preset keyword, according to target service type to service textual data
According to semantics recognition and Entity recognition is carried out, target semantic label and target text entity are obtained;
According to target user's mark, target semantic label and target text entity building user's portrait.
Service text data includes multiple sub-services text datas in one of the embodiments,;Processor executes calculating
It is also performed the steps of when machine program
Semantics recognition and Entity recognition are carried out to each sub-services text data respectively, obtain the mesh of each sub-services text data
Mark semantic label and target text entity;
The incidence relation between the target semantic label and target text entity of each sub-services text data is established respectively;
By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
In one embodiment, it is also performed the steps of when processor executes computer program
According to the corresponding relationship between preset type of service and identification model, mesh is chosen from multiple semantics recognition models
The corresponding target semanteme identification model of type of service is marked, and chooses target service type from multiple entity recognition models and corresponds to
Target entity identification model;
Semantics recognition is carried out to service text data using target semanteme identification model, obtains target semantic label;Wherein,
Semantic label is used to indicate the classification of concern information;
Entity recognition is carried out to service text data using target entity identification model, obtains target text entity;Wherein,
Text entities are used to indicate concern information.
In one embodiment, it is also performed the steps of when processor executes computer program
Obtain the voice data in service process;
Convert voice data into service text data.
In one embodiment, it is also performed the steps of when processor executes computer program
Receive target user's mark of input;Or,
Identify that target user identifies from voice data and/or service text data.
In one embodiment, it is also performed the steps of when processor executes computer program
The selection instruction for obtaining type of service, choosing instruction includes that type of service identifies;
It is identified according to type of service and determines target service type.
In one embodiment, it is also performed the steps of when processor executes computer program
If identifying preset keyword in service text data, according to preset keyword and keyword label it
Between corresponding relationship, determine the corresponding keyword label of service text data;
By keyword label storage corresponding with target user's mark, building user's portrait.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Obtain service text data and the corresponding target user's mark of service text data and target service type;
Keyword identification is carried out to service text data according to target service type;
If unidentified in service text data arrive preset keyword, according to target service type to service textual data
According to semantics recognition and Entity recognition is carried out, target semantic label and target text entity are obtained;
According to target user's mark, target semantic label and target text entity building user's portrait.
Service text data includes multiple sub-services text datas in one of the embodiments,;Computer program is located
Reason device also performs the steps of when executing
Semantics recognition and Entity recognition are carried out to each sub-services text data respectively, obtain the mesh of each sub-services text data
Mark semantic label and target text entity;
The incidence relation between the target semantic label and target text entity of each sub-services text data is established respectively;
By the storage corresponding with target user's mark of multiple incidence relations, building user's portrait.
In one embodiment, it is also performed the steps of when computer program is executed by processor
According to the corresponding relationship between preset type of service and identification model, mesh is chosen from multiple semantics recognition models
The corresponding target semanteme identification model of type of service is marked, and chooses target service type from multiple entity recognition models and corresponds to
Target entity identification model;
Semantics recognition is carried out to service text data using target semanteme identification model, obtains target semantic label;Wherein,
Semantic label is used to indicate the classification of concern information;
Entity recognition is carried out to service text data using target entity identification model, obtains target text entity;Wherein,
Text entities are used to indicate concern information.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Obtain the voice data in service process;
Convert voice data into service text data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Receive target user's mark of input;Or,
Identify that target user identifies from voice data and/or service text data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The selection instruction for obtaining type of service, choosing instruction includes that type of service identifies;
It is identified according to type of service and determines target service type.
In one embodiment, it is also performed the steps of when computer program is executed by processor
If identifying preset keyword in service text data, according to preset keyword and keyword label it
Between corresponding relationship, determine the corresponding keyword label of service text data;
By keyword label storage corresponding with target user's mark, building user's portrait.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Instruct relevant hardware to complete by computer program, computer program to can be stored in a non-volatile computer readable
It takes in storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, this Shen
Please provided by any reference used in each embodiment to memory, storage, database or other media, may each comprise
Non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
Above embodiments only express the several embodiments of the application, and the description thereof is more specific and detailed, but can not
Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art, In
Under the premise of not departing from the application design, various modifications and improvements can be made, these belong to the protection scope of the application.
Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of construction method of user's portrait, which is characterized in that the described method includes:
Obtain service text data and the corresponding target user's mark of the service text data and target service type;
Keyword identification is carried out to the service text data according to the target service type;
If unidentified in the service text data arrive preset keyword, according to the target service type to the clothes
Business text data carries out semantics recognition and Entity recognition, obtains target semantic label and target text entity;
According to target user mark, the target semantic label and target text entity building user's portrait.
2. the method according to claim 1, wherein the service text data includes multiple sub-services textual datas
According to;
It is described that semantics recognition and Entity recognition are carried out to the service text data according to the target service type, obtain target
Semantic label and target text entity, comprising:
Semantics recognition and Entity recognition are carried out to each sub-services text data respectively, obtain each sub-services text data
Target semantic label and target text entity;
It is described to be drawn a portrait according to target user mark, the target semantic label and target text entity building user,
Include:
The incidence relation between the target semantic label and target text entity of each sub-services text data is established respectively;
By the storage corresponding with target user mark of multiple incidence relations, user's portrait is constructed.
3. the method according to claim 1, wherein described literary to the service according to the target service type
Notebook data carries out semantics recognition and Entity recognition, comprising:
According to the corresponding relationship between preset type of service and identification model, the mesh is chosen from multiple semantics recognition models
The corresponding target semanteme identification model of type of service is marked, and chooses the target service type from multiple entity recognition models
Corresponding target entity identification model;
Semantics recognition is carried out to the service text data using the target semanteme identification model, obtains the target semanteme mark
Label;Wherein, semantic label is used to indicate the classification of concern information;
Entity recognition is carried out to the service text data using the target entity identification model, it is real to obtain the target text
Body;Wherein, text entities are used to indicate the concern information.
4. the method according to claim 1, wherein servicing text data and service text in the acquisition
Before the corresponding target user's mark of notebook data and target service type, the method also includes:
Obtain the voice data in service process;
The voice data is converted into the service text data.
5. according to the method described in claim 4, it is characterized in that, the corresponding target of the service text data that obtains is used
Family mark, comprising:
Receive the target user mark of input;Or,
Target user's mark is identified from the voice data and/or the service text data.
6. the method according to claim 1, wherein described obtain the corresponding target industry of the service text data
Service type, comprising:
The selection instruction of type of service is obtained, the selection instruction includes that type of service identifies;
The target service type is determined according to type of service mark.
7. the method according to claim 1, wherein the method also includes:
If identifying preset keyword in the service text data, according to preset keyword and keyword label it
Between corresponding relationship, determine the corresponding keyword label of the service text data;
By keyword label storage corresponding with target user mark, user's portrait is constructed.
8. a kind of construction device of user portrait, which is characterized in that described device includes:
Text data obtains module, for obtaining service text data and the corresponding target user's mark of the service text data
Know and target service type;
Keyword identification module, for carrying out keyword identification to the service text data according to the target service type;
Semantic and Entity recognition module, if arriving preset keyword, basis for unidentified in the service text data
The target service type carries out semantics recognition and Entity recognition to the service text data, obtains target semantic label and mesh
Mark text entities;
First user, which draws a portrait, constructs module, for according to target user mark, the target semantic label and the target
Text entities construct user's portrait.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 7 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910583155.5A CN110457679B (en) | 2019-07-01 | 2019-07-01 | User portrait construction method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910583155.5A CN110457679B (en) | 2019-07-01 | 2019-07-01 | User portrait construction method, device, computer equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110457679A true CN110457679A (en) | 2019-11-15 |
CN110457679B CN110457679B (en) | 2023-11-28 |
Family
ID=68481972
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910583155.5A Active CN110457679B (en) | 2019-07-01 | 2019-07-01 | User portrait construction method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110457679B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111444341A (en) * | 2020-03-16 | 2020-07-24 | 中国平安人寿保险股份有限公司 | User portrait construction method, device and equipment and readable storage medium |
CN112307337A (en) * | 2020-10-30 | 2021-02-02 | 中国平安人寿保险股份有限公司 | Association recommendation method and device based on label knowledge graph and computer equipment |
CN113379481A (en) * | 2021-05-25 | 2021-09-10 | 北京大米科技有限公司 | Data processing method and device |
CN113449202A (en) * | 2021-06-30 | 2021-09-28 | 未鲲(上海)科技服务有限公司 | Information matching method, device, equipment and medium |
CN113779372A (en) * | 2020-11-17 | 2021-12-10 | 北京京东尚科信息技术有限公司 | User group portrait establishing method and device |
CN113836379A (en) * | 2021-09-26 | 2021-12-24 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on customer image |
CN114048283A (en) * | 2022-01-11 | 2022-02-15 | 北京仁科互动网络技术有限公司 | User portrait generation method and device, electronic equipment and storage medium |
CN114153716A (en) * | 2022-02-08 | 2022-03-08 | 中国电子科技集团公司第五十四研究所 | Real-time portrait generation method for people and nobody objects under semantic information exchange network |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000250944A (en) * | 1998-12-28 | 2000-09-14 | Toshiba Corp | Information providing method and device, information receiving device and information describing method |
US20070106497A1 (en) * | 2005-11-09 | 2007-05-10 | Microsoft Corporation | Natural language interface for driving adaptive scenarios |
JP2007249584A (en) * | 2006-03-15 | 2007-09-27 | Softec:Kk | Client database creation method, data retrieval method, data retrieval system, data retrieval filtering system, client database creation program, data retreival program, data retreival filtering program, and computer-readable recording medium storing program or equipment recording program |
CN106940705A (en) * | 2016-12-20 | 2017-07-11 | 上海掌门科技有限公司 | A kind of method and apparatus for being used to build user's portrait |
CN108521439A (en) * | 2018-03-08 | 2018-09-11 | 北京三快在线科技有限公司 | A kind of method and apparatus of message push |
CN108898429A (en) * | 2018-06-19 | 2018-11-27 | 平安科技(深圳)有限公司 | Electronic device, preference tendency prediction technique and computer readable storage medium |
-
2019
- 2019-07-01 CN CN201910583155.5A patent/CN110457679B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000250944A (en) * | 1998-12-28 | 2000-09-14 | Toshiba Corp | Information providing method and device, information receiving device and information describing method |
US20070106497A1 (en) * | 2005-11-09 | 2007-05-10 | Microsoft Corporation | Natural language interface for driving adaptive scenarios |
JP2007249584A (en) * | 2006-03-15 | 2007-09-27 | Softec:Kk | Client database creation method, data retrieval method, data retrieval system, data retrieval filtering system, client database creation program, data retreival program, data retreival filtering program, and computer-readable recording medium storing program or equipment recording program |
CN106940705A (en) * | 2016-12-20 | 2017-07-11 | 上海掌门科技有限公司 | A kind of method and apparatus for being used to build user's portrait |
CN108521439A (en) * | 2018-03-08 | 2018-09-11 | 北京三快在线科技有限公司 | A kind of method and apparatus of message push |
CN108898429A (en) * | 2018-06-19 | 2018-11-27 | 平安科技(深圳)有限公司 | Electronic device, preference tendency prediction technique and computer readable storage medium |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111444341A (en) * | 2020-03-16 | 2020-07-24 | 中国平安人寿保险股份有限公司 | User portrait construction method, device and equipment and readable storage medium |
CN111444341B (en) * | 2020-03-16 | 2024-04-12 | 中国平安人寿保险股份有限公司 | User portrait construction method, device, equipment and readable storage medium |
CN112307337A (en) * | 2020-10-30 | 2021-02-02 | 中国平安人寿保险股份有限公司 | Association recommendation method and device based on label knowledge graph and computer equipment |
CN112307337B (en) * | 2020-10-30 | 2024-04-12 | 中国平安人寿保险股份有限公司 | Associated recommendation method and device based on tag knowledge graph and computer equipment |
CN113779372A (en) * | 2020-11-17 | 2021-12-10 | 北京京东尚科信息技术有限公司 | User group portrait establishing method and device |
CN113379481A (en) * | 2021-05-25 | 2021-09-10 | 北京大米科技有限公司 | Data processing method and device |
CN113449202A (en) * | 2021-06-30 | 2021-09-28 | 未鲲(上海)科技服务有限公司 | Information matching method, device, equipment and medium |
CN113836379A (en) * | 2021-09-26 | 2021-12-24 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on customer image |
CN113836379B (en) * | 2021-09-26 | 2023-08-25 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on client image |
CN114048283A (en) * | 2022-01-11 | 2022-02-15 | 北京仁科互动网络技术有限公司 | User portrait generation method and device, electronic equipment and storage medium |
CN114153716A (en) * | 2022-02-08 | 2022-03-08 | 中国电子科技集团公司第五十四研究所 | Real-time portrait generation method for people and nobody objects under semantic information exchange network |
Also Published As
Publication number | Publication date |
---|---|
CN110457679B (en) | 2023-11-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110457679A (en) | Construction method, device, computer equipment and the storage medium of user's portrait | |
CN110532361A (en) | Recognition methods, device, computer equipment and the storage medium that user is intended to | |
CN110472224B (en) | Quality of service detection method, apparatus, computer device and storage medium | |
CN110890088B (en) | Voice information feedback method and device, computer equipment and storage medium | |
CN112733042B (en) | Recommendation information generation method, related device and computer program product | |
CN110046806B (en) | Method and device for customer service order and computing equipment | |
CN112732911A (en) | Semantic recognition-based conversational recommendation method, device, equipment and storage medium | |
CN111274372A (en) | Method, electronic device, and computer-readable storage medium for human-computer interaction | |
CN110399473B (en) | Method and device for determining answers to user questions | |
CN107291775B (en) | Method and device for generating repairing linguistic data of error sample | |
CN107181879B (en) | Identification incoming call is intended to | |
CN113240510B (en) | Abnormal user prediction method, device, equipment and storage medium | |
CN111221949A (en) | Intelligent return visit method, device and equipment based on reinforcement learning and storage medium | |
CN111814487A (en) | Semantic understanding method, device, equipment and storage medium | |
CN111291172A (en) | Method and device for processing text | |
CN109033427B (en) | Stock screening method and device, computer equipment and readable storage medium | |
CN112235470A (en) | Incoming call client follow-up method, device and equipment based on voice recognition | |
CN114780701B (en) | Automatic question-answer matching method, device, computer equipment and storage medium | |
CN111738791B (en) | Text processing method, device, equipment and storage medium | |
CN110704599B (en) | Method and device for generating samples for prediction model and method and device for training prediction model | |
CN114974253A (en) | Natural language interpretation method and device based on character image and storage medium | |
CN114817507A (en) | Reply recommendation method, device, equipment and storage medium based on intention recognition | |
US11715469B2 (en) | Methods and apparatus for improving search retrieval using inter-utterance context | |
CN111782792A (en) | Method and apparatus for information processing | |
CN112669850A (en) | Voice quality detection method and device, computer equipment and 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |