CN110413758A - Dialog box framework construction method and device based on machine learning - Google Patents
Dialog box framework construction method and device based on machine learning Download PDFInfo
- Publication number
- CN110413758A CN110413758A CN201910694635.9A CN201910694635A CN110413758A CN 110413758 A CN110413758 A CN 110413758A CN 201910694635 A CN201910694635 A CN 201910694635A CN 110413758 A CN110413758 A CN 110413758A
- Authority
- CN
- China
- Prior art keywords
- model
- state
- session
- node
- flow nodes
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Machine Translation (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present invention provides a kind of dialog box framework construction method and device based on machine learning, this method comprises: selecting to the mark that client sends the flow nodes of multiple pre-configurations for user configuration；Obtain the session process of client feedback；Corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；The finite state machine model table of the session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition, as session frame, wherein, by pushing the alternate node of multiple pre-configurations to user, session frame can be quickly generated, in addition, it is being directed to different sessions, it can be multiplexed the alternate node of pre-configuration, greatly saved the development time and exploited natural resources, development rate is improved.
Description
Technical field
The present invention relates to field of artificial intelligence more particularly to a kind of dialog box framework construction methods based on machine learning
And device.
Background technique
With the development of technology, more and more equipment will have networked capabilities, and how these equipment interact with people
A challenge will be become.
Natural language becomes the novel interactive mode for adapting to the trend, and dialogue robot is expected to replace past website, such as
Modern APP, occupies human-computer interaction air port of new generation.Dialogue robot is inherently one and is used to simulate human conversation or chat
Computer program.In technological layer, talking with robot can be divided into three classes: chat robot, question and answer robot, task equipment
People.
With the development of artificial intelligence and natural language processing technique, talk with robot in financial service, home life, a
It will gradually be applied in people assistant.
Currently to the market demand diversification of robot, need to create corresponding dialogue machine for different application demands
For device people to realize personalized function, still, the building of the dialogue robot of different function is online from planning, being implemented into, and is related to
Link is complicated, and time-consuming for iteration optimization, and creation threshold is high, and Project Scheduling low efficiency limits the extensive use of dialogue robot.
Summary of the invention
For the problems of the prior art, the present invention provides a kind of dialog box framework construction method based on machine learning, dress
It sets, electronic equipment and computer readable storage medium, problems of the prior art can at least be partially solved.
To achieve the goals above, the present invention adopts the following technical scheme:
In a first aspect, providing a kind of dialog box framework construction method based on machine learning, comprising:
The mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, the process of multiple pre-configurations
Node executes multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple states, each state
It include: existing state, condition, movement and next state；
The session process of client feedback is obtained, which includes multiple flow nodes marks and node flow direction；
Corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；
The limited of the session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
State machine model table, as session frame.
Further, which is the state that dialogue is presently in；A movement is triggered when the condition is fulfilled or is held
State transition of row, the next state are the new state to be moved to after condition meets；
This obtains having for the session process according to the flow nodes information that the node in the session process flows to tissue acquisition
State machine model table is limited, as session frame, comprising:
The flow nodes information of acquisition is ranked up according to the node flow direction in the session process；
The next state of each state in flow nodes information after configuration sequence, to be flowed to according to the node in session process
Node is carried out to jump.
Further, which includes: address field, is used to indicate the new state to be moved to after condition meets；
The next state of each state in flow nodes information after configuration sequence, comprising:
It will be in the address field in the next state of the jump address insertion present node most end state of latter node.
It further, include: that rule template identifies and/or handle model identification in the movement,
The dialog box framework construction method further include:
Corresponding rule are called from database according to the rule template mark for respectively acting included in limit state machine model table
Then template；
According to respectively acted in limit state machine model table included processing model identification from called in database it is corresponding from
Manage model.
Further, which includes: participle model, textual classification model or elements recognition model.
Further, further includes:
Reason model identification calls the model training process of the model from model training engine according to this；
Obtain the model training data through marking；
The processing model is trained according to the model training data through marking and the model training process.
Further, further includes:
Obtain the model measurement data through marking；
The processing model is tested using the model measurement data through marking.
Second aspect provides a kind of session framework establishment device based on machine learning, comprising:
Alternate node display module, the mark for sending the flow nodes of multiple pre-configurations to client are selected for user configuration
It selects, the flow nodes of multiple pre-configurations execute multiple and different operations respectively, and the flow nodes information of each pre-configuration includes
Multiple states, each state includes: existing state, condition, movement and next state；
Session process receiving module obtains the session process of client feedback, which includes multiple flow nodes
Mark and node flow direction；
Flow nodes data obtaining module obtains corresponding process according to multiple flow nodes mark in the session process
Nodal information；
Dialog box framework models block, is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
The finite state machine model table of the session process, as session frame.
Further, which is the state that dialogue is presently in；A movement is triggered when the condition is fulfilled or is held
State transition of row, the next state are the new state to be moved to after condition meets；
The dialog box framework models block
Framework organizational unit is ranked up the flow nodes information of acquisition according to the node flow direction in the session process；
Configuration unit, the next state of each state in flow nodes information after configuration sequence, so as to according in session process
Node flow direction carry out node jump.
Further, which includes: address field, is used to indicate the new state to be moved to after condition meets；
The configuration unit includes:
Configuration subelement is jumped, by the address in the next state of the jump address insertion present node most end state of latter node
In column.
It further, include: that rule template identifies and/or handle model identification in the movement,
The session framework establishment device further include:
Rule template calling module is identified according to included rule template is respectively acted in limit state machine model table from data
Corresponding rule template is called in library；
Model calling module is handled, respectively acts included processing model identification from data according in limit state machine model table
Corresponding processing model is called in library.
Further, which includes: participle model, textual classification model or elements recognition model.
Further, further includes:
Model training process obtains module, manages the mould that model identification calls the model from model training engine according to this
Type training process；
Training sample obtains module, obtains the model training data through marking；
Model training module, according to the model training data through marking and the model training process to the processing model into
Row training.
Further, further includes:
Test sample obtains module, obtains the model measurement data through marking；
Model measurement module tests the processing model using the model measurement data through marking.
The third aspect, provides a kind of electronic equipment, including memory, processor and storage on a memory and can handled
The computer program run on device, the processor realize the above-mentioned session framework establishment based on machine learning when executing the program
The step of method.
Fourth aspect provides a kind of computer readable storage medium, is stored thereon with computer program, the computer program
The step of above-mentioned dialog box framework construction method based on machine learning is realized when being executed by processor.
Dialog box framework construction method, device, electronic equipment and computer provided by the invention based on machine learning can
Storage medium is read, this method comprises: selected to the mark that client sends the flow nodes of multiple pre-configurations for user configuration, it is more
The flow nodes of a pre-configuration execute multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple shapes
State, each state includes: existing state, condition, movement and next state；Obtain the session process of client feedback, the session process
Including multiple flow nodes mark and node flow direction；It is obtained according to multiple flow nodes mark in the session process corresponding
Flow nodes information；The session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
Finite state machine model table, as session frame, wherein, can be fast by pushing the alternate node of multiple pre-configurations to user
Fast-growing is at session frame, in addition, can be multiplexed the alternate node of pre-configuration being directed to different sessions, greatly save and opened
Hair and is exploited natural resources at the time, improves development rate, and user only need to drag dotted line connection and will trained or test data by dilatory
It is circulated into server, can create and train the dialogue robot of meet demand, can be online directly in engineering, and allow machine
People becomes smarter more intelligent as the dialogue with people constantly learns, and realizes flexible customization and efficiently scheduling, can be fast
Fast simple batch " production " talks with robot.
For above and other objects, features and advantages of the invention can be clearer and more comprehensible, preferred embodiment is cited below particularly,
And cooperate institute's accompanying drawings, it is described in detail below.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the application
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.In the accompanying drawings:
Fig. 1 is the server S 1 in the embodiment of the present invention and the configuration diagram between client device B1；
Fig. 2 is the framework between server S 1, client device B1 and database server S2 in the embodiment of the present invention
Schematic diagram；
Fig. 3 is the flow diagram one of the dialog box framework construction method based on machine learning in the embodiment of the present invention；
Fig. 4 shows what the use dialog box framework construction method provided in an embodiment of the present invention based on machine learning constructed
Session frame；
Fig. 5 shows the specific steps of step S400 in Fig. 3；
Fig. 6 is the structural block diagram of the session framework establishment device based on machine learning in the embodiment of the present invention；
Fig. 7 is the structure chart of electronic equipment of the embodiment of the present invention.
Specific embodiment
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection
It encloses.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
It should be noted that term " includes " and " tool in the description and claims of this application and above-mentioned attached drawing
Have " and their any deformation, it is intended that cover it is non-exclusive include, for example, containing a series of steps or units
Process, method, system, product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include without clear
Other step or units listing to Chu or intrinsic for these process, methods, product or equipment.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Currently to the market demand diversification of robot, need to create corresponding dialogue machine for different application demands
For device people to realize personalized function, still, the building of the dialogue robot of different function is online from planning, being implemented into, and is related to
Link is complicated, and time-consuming for iteration optimization, and creation threshold is high, and Project Scheduling low efficiency limits the extensive use of dialogue robot.
At least partly to solve the above-mentioned technical problems in the prior art, the embodiment of the present invention provides a kind of based on machine
The dialog box framework construction method of device study can quickly generate session by pushing the alternate node of multiple pre-configurations to user
Frame, in addition, the alternate node of pre-configuration can be multiplexed being directed to different sessions, greatly saved the development time and opened
Resource is sent out, improves development rate, user only need to drag dotted line connection and trained or test data is circulated into service by dilatory
In device, can create and train the dialogue robot of meet demand, can be online directly in engineering, and allow robot with people
Dialogue constantly learn and become smarter more intelligent, realize flexible customization and efficiently scheduling, can quickly and easily batch
" production " talks with robot.
In view of this, this application provides a kind of session framework establishment device based on machine learning, the device can be
A kind of server S 1, referring to Fig. 1, which can communicate to connect at least one client device B1, the server S 1
The mark for sending the flow nodes of multiple pre-configurations to client device B1 is selected for user configuration, the process section of multiple pre-configurations
Point executes multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple states, each state
It include: existing state, condition, movement and next state；Selected session process can be sent to the clothes by the client device B1
Be engaged in device S1, and the session process includes that multiple flow nodes marks and node flow direction, the server S 1 can receive online
The session process obtains corresponding flow nodes information according to multiple flow nodes mark in the session process；According to
The flow nodes information that node in the session process flows to tissue acquisition obtains the finite state machine mould of the session process
Type table, as session frame.
In addition, referring to fig. 2, the server S 1 can also be communicated to connect at least one database server S2, described
Database server S2 is for storage rule template, the training process of processing model and each model, for the server S 1
It calls, the server S 1 can receive the model training data through marking online, then according to the model training number through marking
Model training is carried out according to required processing model.
Based on above content, the server S 1 can receive the model training data through marking online, then utilize institute
The model measurement data through marking are stated to test the processing model.
It is understood that the client device B1 may include smart phone, Flat electronic equipment, portable computing
Machine, desktop computer, personal digital assistant (PDA) etc..
In practical applications, the conversate part of framework establishment can be held in 1 side of server S as described in above content
Row, that is, framework as shown in Figure 1, operation that can also be all is all completed in the client device B1, and the client
End equipment B1 can be directly communicatively coupled with database server S2.It specifically can be according to the place of the client device B1
Reason ability and the limitation of user's usage scenario etc. select.The application is not construed as limiting this.If all operations are all in institute
It states and is completed in client device B1, the client device B1 can also include processor, for the framework establishment that conversates
Specific processing.
Any suitable network protocol can be used between the server and the client device to be communicated, including
In the network protocol that the application submitting day is not yet developed.The network protocol for example may include ICP/IP protocol, UDP/IP
Agreement, http protocol, HTTPS agreement etc..Certainly, the network protocol for example can also include using on above-mentioned agreement
RPC agreement (Remote Procedure Call Protocol, remote procedure call protocol), REST agreement
(Representational State Transfer, declarative state transfer protocol) etc..
Fig. 3 is the flow diagram one of the dialog box framework construction method based on machine learning in the embodiment of the present invention.Such as
Shown in Fig. 3, being somebody's turn to do the dialog box framework construction method based on machine learning may include the following contents:
Step S100: the mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, multiple pre-
The flow nodes of configuration execute multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple states,
Each state includes: existing state, condition, movement and next state；
Specifically, each flow nodes are pre-configured with by user, are needed in movement for executing a Xiang Zuoye, state
Model or algorithm of calling etc. are also pre-configured by user.
In addition, the flow nodes not only include the node for executing logical process, further includes replying words art node, reply words art
The specific words art replied in node is configured when creating session process according to specific task by user, is intended to according to user
Return to art if configuring.
Step S200: obtaining the session process of client feedback, the session process include multiple flow nodes marks with
And node flow direction.
Wherein, node flow direction shows the direction of process.
Specifically, user can drag node realization session stream needed for dotted line connects by drawing on the display of client
Journey configuration.
It is worth noting that the session process can be understood as realizing the flow chart of a task, referring to fig. 4.
As shown in figure 4, the session process is the session frame of the task of the realization adjustment credit card amount of user configuration.
Session process needed for the session frame are as follows:
Authentication is carried out first；When authentication failure, the Voice Navigation of a node is jumped to；Work as authentication
After success, jump volume is judged whether it is；If so, obtaining credit card amount, if meeting condition, words art " confirmation tune volume " is replied, if
It is unsatisfactory for condition, is intended to according to specific client, Voice Navigation node is jumped to or turns artificial or terminates session after replying.
The realization session process of node needed for dotted line connects according to the above process is dragged by drawing on the display of client
Configuration, and configure verbal message in reply art needed for art node is talked about in each answer.
Step S300: corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；
Specifically, during flow nodes are fed back to client so as to client's configuration flow, to reduce data transmission,
Only transmission flow node identification, the realization program of flow nodes are not required to be transmitted to client, on the one hand can be improved transmission speed,
On the other hand, node procedure can be prevented to be stolen duplication.
It include multiple flow nodes marks and node flow direction in the session process of user configuration based on this.Need basis
Flow nodes mark goes in database that the flow nodes is called to identify corresponding flow nodes information.
Wherein, the nodal community of user configuration may include: nodename, max-session round, whether edit words art,
With word slot, whether can interrupt.
Step S400: the meeting is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
The finite state machine model table for talking about process, as session frame.
It will be appreciated by persons skilled in the art that after obtaining each flow nodes information, because each flow nodes are believed
Breath is independent prefabricated, therefore, is not associated between each other, at this time, it may be necessary to flow to each process of tissue acquisition according to node
The flow nodes information of node, with the session frame of accomplished particular task, which can use finite state machine mould
Type table (referring to Tables 1 and 2) indicates, is judged to execute which state turns according to system current state and the event occurred
Exchange the letters number realizes the automatic variation of system mode with conditional branching technology.
Table 1
Table 2
Wherein, finite state machine model table includes existing state, condition, movement, next state.Existing state and condition are because acting and secondary
State is fruit.Wherein, in session frame:
Existing state: talk with the state being presently in.
Condition: when a condition is satisfied and (recognizes some intention), it will one movement of triggering executes a shape
The migration of state.
Movement: the movement that condition executes after meeting can move to new state after movement is finished, can also be still
It is old to maintain the original state.In talking with robot, if robot executes reply difference according to the different intentions (condition) recognized
Art (movement).
Next state: the new state to be moved to after condition meets.For existing state, next state is once activated for next state, just turns
Become new existing state.
Through the above technical solution it is known that the session framework establishment provided in an embodiment of the present invention based on machine learning
Method can quickly generate session frame by pushing the alternate node of multiple pre-configurations to user, in addition, for difference
Session, the alternate node of pre-configuration can be multiplexed, greatly saved and the development time and exploited natural resources, improve exploitation speed
Degree, user only need to be dragged dotted line connection and trained or test data are circulated into server, can created and train by dilatory
The dialogue robot of meet demand, can be online directly in engineering, and allow robot as the dialogue with people constantly learns and
Become smarter more intelligent, realize flexible customization and efficiently scheduling, quickly and easily can talk with robot by batch " production ".
Fig. 5 shows the specific steps of step S400 in Fig. 3；As shown in figure 5, step S400 may include in following
Hold:
Step S410: the flow nodes information of acquisition is ranked up according to the node flow direction in the session process；
It is worth noting that node flow direction may include that straight line flow direction and multi-fork formula flow direction, straight line flow to for example: node
1 → node, 2 → node, 3 → node 4 does not have bifurcated at each node.Multi-fork formula flows to referring to fig. 4, can be with bifurcated at node
With not bifurcated.
Specifically, the corresponding flow nodes information of each node of tissue is flowed to according to node.
Step S420: the next state of each state in flow nodes information after configuration sequence, so as to according in session process
Node flow direction carry out node jump.
Wherein, the existing state is the state that dialogue is presently in；When the condition is satisfied, triggering one acts or holds
State transition of row, the next state are the new state to be moved to after condition meets.
It is worth noting that the process of configuration is mainly that the relationship that jumps of each flow nodes is carried out group according to node flow direction
It knits, therefore, mainly configures the next state of each state in the corresponding flow nodes information of each node.
Specifically, next state may include: address field, be used to indicate the new state to be moved to after condition meets；
The next state of each state in flow nodes information after configuration sequence is are as follows: is inserted into the jump address of latter node
In address field in the next state of present node most end state.
By using above-mentioned technical proposal, each node can be made to match after executing the operation of this node according to user
The process set jumps to next node.
It in an alternative embodiment, include: rule template mark and/or processing model in the corresponding movement of each state
Mark, for executing certain movement.
The dialog box framework construction method further include:
Corresponding rule are called from database according to the rule template mark for respectively acting included in limit state machine model table
Then template；
According to respectively acted in limit state machine model table included processing model identification from called in database it is corresponding from
Manage model.
Wherein, which includes: common in the sessions such as participle model, textual classification model or elements recognition model
Computer model.
It will be appreciated by persons skilled in the art that rule template and processing model are typically complex, occupied space is big,
Therefore, corresponding rule template or processing model are not stored in each node, only by user configuration good rule template mark and/
Or processing model identification, therefore, it is necessary to according to the mark go in database call to rule template and processing model.
It is worth noting that the mark mentioned in the application can be title or number etc..
In an alternative embodiment, being somebody's turn to do the dialog box framework construction method based on machine learning can also include:
The model training process of the model is called from model training engine according to the processing model identification；
Obtain the model training data through marking；
The processing model is trained according to the model training data through marking and the model training process.
Wherein, there is corresponding model training process for different processing models, model training process is pre-stored in model instruction
Practice in engine, in practical application, call the model training process of the model from model training engine according to processing model identification,
For being trained to model.
By using above-mentioned technical proposal, machine learning method can be used and fast implement modeling training, it can quick structure
Dialogue robot and hoisting machine people's level of intelligence are built, the online cost of engineering is saved, prevents overlapping development, saved exploitation manpower
And resource.
In an alternative embodiment, being somebody's turn to do the dialog box framework construction method based on machine learning can also include:
Obtain the model measurement data through marking；
The processing model is tested using the model measurement data through marking.
Wherein, after model training, by the model after training the input of the model measurement data through marking, by mould
The output of type compares the label that test result marks in advance with model measurement data as test result, judges mould with this
Whether type precision meets the requirements, if so, the model after the training is distributed in node, if it is not, then replacing training data pair
Model carries out retraining.
In an alternative embodiment, it is somebody's turn to do the dialog box framework construction method based on machine learning and is getting user's input
Training and test data after, the data are named and are arranged with browse right.
Wherein, numerical nomenclature is that corresponding sample data set is selected in model training in order to subsequent.
Data permission setting be when multiple machine scenarios are researched and developed jointly, guarantee only have certain scene related personnel may have access to or
Modify data.
In an alternative embodiment, being somebody's turn to do the dialog box framework construction method based on machine learning can also include: setting
Data display parameters.
Specifically, display parameters include: preview data show line number, check the mode that data are shown, for example, data come
Source, number of data lines and upload date etc., to check data for user or developer after data import, it is ensured that data import
It is errorless.
In an alternative embodiment, being somebody's turn to do the dialog box framework construction method based on machine learning can also include:
The session frame of acquisition is numbered.
Specifically, the situation that multiple robots or a robot may be implemented with a variety of sessions, passes through generation
Session frame carries out label to distinguish different session frames.
In an alternative embodiment, being somebody's turn to do the dialog box framework construction method based on machine learning can also include: increment
Data annotation step.
Specifically, session data newly-increased daily is clustered by clustering algorithm, it is preliminary to identify that user is intended to, and by
It is manually modified, to carry out supplementary training to processing model using revised data.
In an alternative embodiment, machine learning model carries out publication control by publication control and realease agent.
Unified publication control PublishMaster mating operation station request.Isomery realease agent Publish Agent is the logic of publication
Control node receives the information of PublishMaster, is based on k8s Gateway, realizes the publication or upgrading of isomery model.
Machine learning model service management is managed and monitors to the service estimated on line, and service management passes through calling
K8S interface realizes the upper offline of service and management, and the model version in service is subscribed to and release status monitoring passes through registration
Zookeeper is realized.It is unified existing based on PAAS mysorethorn to estimate service.
It is worth noting that prestoring a variety of models in database, such as participle model (jieba etc.), textual classification model
(Fasttext, RNN, BERT etc.), elements recognition model (CRF, LSTM etc.).By calling above-mentioned model, robot is allowed to identify
User is intended to and extracts key element.For example user says: I will be given to 100 yuan of Xiao Ming.It is identified and is intended to by model are as follows: turned
Account.Identification key element: Xiao Ming, 100 pieces.
Model training engine utilizes big data scheduling of resource YARN, packaging model training engine, and H2O training of such as increasing income is drawn
It holds up, Python model training engine etc..
Based on the same inventive concept, the embodiment of the present application also provides a kind of, and the session framework establishment based on machine learning fills
It sets, can be used to implement method described in above-described embodiment, as described in the following examples.Due to the meeting based on machine learning
The principle that words framework establishment device solves the problems, such as is similar to the above method, therefore the session framework establishment device based on machine learning
Implementation may refer to the implementation of the above method, overlaps will not be repeated.It is used below, term " unit " or " mould
The combination of the software and/or hardware of predetermined function may be implemented in block ".Although device described in following embodiment is preferably with soft
Part is realized, but the realization of the combination of hardware or software and hardware is also that may and be contemplated.
Fig. 6 is the structural block diagram one of the session framework establishment device based on machine learning in the embodiment of the present invention.Such as Fig. 6
Shown, be somebody's turn to do the session framework establishment device based on machine learning and specifically include: alternate node display module 10, session process receive
Module 20, flow nodes data obtaining module 30 and dialog box framework model block 40.
Alternate node display module 10 is selected to the mark for the flow nodes that client sends multiple pre-configurations for user configuration
It selects, the flow nodes of multiple pre-configurations execute multiple and different operations respectively, and the flow nodes information of each pre-configuration includes
Multiple states, each state include: existing state, condition, movement and next state；
Specifically, each flow nodes are pre-configured with by user, are needed in movement for executing a Xiang Zuoye, state
Model or algorithm of calling etc. are also pre-configured by user.
In addition, the flow nodes not only include the node for executing logical process, further includes replying words art node, reply words art
The specific words art replied in node is configured when creating session process according to specific task by user, is intended to according to user
Return to art if configuring.
Session process receiving module 20 obtains the session process of client feedback, and the session process includes multiple process sections
Point identification and node flow direction；
Wherein, node flow direction shows the direction of process.
Specifically, user can drag node realization session stream needed for dotted line connects by drawing on the display of client
Journey configuration, and configure verbal message in reply art needed for art node is talked about in each answer.
It is worth noting that the session process can be understood as realizing the flow chart of a task
Flow nodes data obtaining module 30 obtains corresponding according to multiple flow nodes mark in the session process
Flow nodes information；
Specifically, during flow nodes are fed back to client so as to client's configuration flow, to reduce data transmission,
Only transmission flow node identification, the realization program of flow nodes are not required to be transmitted to client, on the one hand can be improved transmission speed,
On the other hand, node procedure can be prevented to be stolen duplication.
It include multiple flow nodes marks and node flow direction in the session process of user configuration based on this.Need basis
Flow nodes mark goes in database that the flow nodes is called to identify corresponding flow nodes information.
Wherein, the nodal community of user configuration may include: nodename, max-session round, whether edit words art,
With word slot, whether can interrupt.
Dialog box framework models the flow nodes information that block 40 flows to tissue acquisition according to the node in the session process
The finite state machine model table of the session process is obtained, as session frame.
It will be appreciated by persons skilled in the art that after obtaining each flow nodes information, because each flow nodes are believed
Breath is independent prefabricated, therefore, is not associated between each other, at this time, it may be necessary to flow to each process of tissue acquisition according to node
The flow nodes information of node, with the session frame of accomplished particular task, which can use finite state machine mould
Type table indicates, judges which state transition function executed according to system current state and the event occurred, with item
The automatic variation of part branch technique realization system mode.
Wherein, finite state machine model table includes existing state, condition, movement, next state.Existing state and condition are because acting and secondary
State is fruit.Wherein, in session frame:
Existing state: talk with the state being presently in.
Condition: when a condition is satisfied and (recognizes some intention), it will one movement of triggering executes a shape
The migration of state.
Movement: the movement that condition executes after meeting can move to new state after movement is finished, can also be still
It is old to maintain the original state.In talking with robot, if robot executes reply difference according to the different intentions (condition) recognized
Art (movement).
Next state: the new state to be moved to after condition meets.For existing state, next state is once activated for next state, just turns
Become new existing state.
Through the above technical solution it is known that the session framework establishment provided in an embodiment of the present invention based on machine learning
Device can quickly generate session frame by pushing the alternate node of multiple pre-configurations to user, in addition, for difference
Session, the alternate node of pre-configuration can be multiplexed, greatly saved and the development time and exploited natural resources, improve exploitation speed
Degree, user only need to be dragged dotted line connection and trained or test data are circulated into server, can created and train by dilatory
The dialogue robot of meet demand, can be online directly in engineering, and allow robot as the dialogue with people constantly learns and
Become smarter more intelligent, realize flexible customization and efficiently scheduling, quickly and easily can talk with robot by batch " production ".
In an alternative embodiment, the existing state is the state that dialogue is presently in；When the condition is satisfied
It triggers a movement or executes a state transition, the next state is the new state to be moved to after condition meets；
The dialog box framework modeling block 40 includes: framework organizational unit and configuration unit.
Framework organizational unit is ranked up the flow nodes information of acquisition according to the node flow direction in the session process；
It is worth noting that node flow direction may include that straight line flow direction and multi-fork formula flow direction, straight line flow to for example: node
1 → node, 2 → node, 3 → node 4 does not have bifurcated at each node.Multi-fork formula flows to referring to fig. 4, can be with bifurcated at node
With not bifurcated.
Specifically, the corresponding flow nodes information of each node of tissue is flowed to according to node.
The next state of each state in flow nodes information after configuration of described dispensing unit sequence, so as to according in session process
Node flow direction carries out node and jumps.
Wherein, the existing state is the state that dialogue is presently in；When the condition is satisfied, triggering one acts or holds
State transition of row, the next state are the new state to be moved to after condition meets.
It is worth noting that the process of configuration is mainly that the relationship that jumps of each flow nodes is carried out group according to node flow direction
It knits, therefore, mainly configures the next state of each state in the corresponding flow nodes information of each node.
In an alternative embodiment, the next state includes: address field, be used to indicate condition meet after to be moved to it is new
State；
The configuration unit includes: to jump configuration subelement, and the jump address of latter node is inserted into present node most end
In address field in the next state of state.
By using above-mentioned technical proposal, each node can be made to match after executing the operation of this node according to user
The process set jumps to next node.
It in an alternative embodiment, include: that rule template identifies and/or handle model identification in the movement,
The session framework establishment device further include: rule template calling module, processing model calling module.
Rule template of the rule template calling module included by respectively acting in limit state machine model table is identified from data
Corresponding rule template is called in library；
Processing model calling module respectively acts included processing model identification from data according in limit state machine model table
Corresponding processing model is called in library.
Wherein, which includes: common in the sessions such as participle model, textual classification model or elements recognition model
Computer model.
It will be appreciated by persons skilled in the art that rule template and processing model are typically complex, occupied space is big,
Therefore, corresponding rule template or processing model are not stored in each node, only by user configuration good rule template mark and/
Or processing model identification, therefore, it is necessary to according to the mark go in database call to rule template and processing model.
It is worth noting that the mark mentioned in the application can be title or number etc..
In an alternative embodiment, the session framework establishment device based on machine learning can also include: model instruction
Practice process and obtains module, training sample acquisition module and model training module.
Model training process obtains module and calls the model from model training engine according to the processing model identification
Model training process；
Training sample obtains module and obtains the model training data through marking；
Model training module is according to the model training data through marking and the model training process to the processing
Model is trained.
Wherein, there is corresponding model training process for different processing models, model training process is pre-stored in model instruction
Practice in engine, in practical application, call the model training process of the model from model training engine according to processing model identification,
For being trained to model.
By using above-mentioned technical proposal, machine learning method can be used and fast implement modeling training, it can quick structure
Dialogue robot and hoisting machine people's level of intelligence are built, the online cost of engineering is saved, prevents overlapping development, saved exploitation manpower
And resource.
In an alternative embodiment, the session framework establishment device based on machine learning can also include: test specimens
This acquisition module and model measurement module.
Test sample obtains module and obtains the model measurement data through marking；
Model measurement module tests the processing model using the model measurement data through marking.
Wherein, after model training, by the model after training the input of the model measurement data through marking, by mould
The output of type compares the label that test result marks in advance with model measurement data as test result, judges mould with this
Whether type precision meets the requirements, if so, the model after the training is distributed in node, if it is not, then replacing training data pair
Model carries out retraining.
In an alternative embodiment, being somebody's turn to do the session framework establishment device based on machine learning can also include name power
Setup module is limited, for being named and being arranged clear to the data after the training and test data for getting user's input
Look at permission.
Wherein, numerical nomenclature is that corresponding sample data set is selected in model training in order to subsequent.
Data permission setting be when multiple machine scenarios are researched and developed jointly, guarantee only have certain scene related personnel may have access to or
Modify data.
In an alternative embodiment, being somebody's turn to do the session framework establishment device based on machine learning can also include: display
Parameter setting module, for data display parameters to be arranged.
Specifically, display parameters include: preview data show line number, check the mode that data are shown, for example, data come
Source, number of data lines and upload date etc., to check data for user or developer after data import, it is ensured that data import
It is errorless.
In an alternative embodiment, being somebody's turn to do the session framework establishment device based on machine learning can also include:
The session frame of acquisition is numbered in session frame number module.
Specifically, the situation that multiple robots or a robot may be implemented with a variety of sessions, passes through generation
Session frame carries out label to distinguish different session frames.
In an alternative embodiment, being somebody's turn to do the session framework establishment device based on machine learning can also include: increment
Data labeling module clusters session data newly-increased daily by clustering algorithm, preliminary to identify that user is intended to.
It after identification user is intended to, is manually modified, to be supplemented using revised data processing model
Training.
In an alternative embodiment, machine learning model carries out publication control by publication control and realease agent.
Unified publication control PublishMaster mating operation station request.Isomery realease agent Publish Agent is the logic of publication
Control node receives the information of PublishMaster, is based on k8s Gateway, realizes the publication or upgrading of isomery model.
Machine learning model service management is managed and monitors to the service estimated on line, and service management passes through calling
K8S interface realizes the upper offline of service and management, and the model version in service is subscribed to and release status monitoring passes through registration
Zookeeper is realized.It is unified existing based on PAAS mysorethorn to estimate service.
It is worth noting that prestoring a variety of models in database, such as participle model (jieba etc.), textual classification model
(Fasttext, RNN, BERT etc.), elements recognition model (CRF, LSTM etc.).By calling above-mentioned model, robot is allowed to identify
User is intended to and extracts key element.For example user says: I will be given to 100 yuan of Xiao Ming.It is identified and is intended to by model are as follows: turned
Account.Identification key element: Xiao Ming, 100 pieces.
Model training engine utilizes big data scheduling of resource YARN, packaging model training engine, and H2O training of such as increasing income is drawn
It holds up, Python model training engine etc..
Device, module or the unit that above-described embodiment illustrates can specifically be realized, Huo Zheyou by computer chip or entity
Product with certain function is realized.It is a kind of typical to realize that equipment is electronic equipment, specifically, electronic equipment for example can be with
For personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media player,
Any in navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or these equipment sets
Standby combination.
Electronic equipment specifically includes memory, processor and storage on a memory and can in a typical example
The computer program run on a processor, the processor realize following step when executing described program:
The mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, the process of multiple pre-configurations
Node executes multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple states, each shape
State includes: existing state, condition, movement and next state；
The session process of client feedback is obtained, the session process includes multiple flow nodes marks and node-flow
To；
Corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；
The session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
Finite state machine model table, as session frame.
As can be seen from the above description, electronic equipment provided in an embodiment of the present invention, can be used for constructing session frame, by with
Family pushes the alternate node of multiple pre-configurations, can quickly generate session frame, in addition, being directed to different sessions, Neng Goufu
With the alternate node of pre-configuration, the development time is greatly saved and has exploited natural resources, improved development rate, user need to only pass through
It is dilatory to drag dotted line connection and trained or test data is circulated into server, it can create and train the dialogue machine of meet demand
Device people, can be online directly in engineering, and robot is allowed to become smarter more intelligent as the dialogue with people constantly learns,
Flexible customization and efficiently scheduling are realized, quickly and easily can talk with robot by batch " production ".
Below with reference to Fig. 7, it illustrates the structural representations for the electronic equipment 600 for being suitable for being used to realize the embodiment of the present application
Figure.
As shown in fig. 7, electronic equipment 600 includes central processing unit (CPU) 601, it can be according to being stored in read-only deposit
Program in reservoir (ROM) 602 is loaded into random access storage device (RAM) from storage section 608) program in 603 and
Execute various work appropriate and processing.In RAM603, also it is stored with system 600 and operates required various programs and data.
CPU601, ROM602 and RAM603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to bus
604。
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.；
And including such as LAN card, the communications portion 609 of the network interface card of modem etc..Communications portion 609 via such as because
The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon
Computer program be mounted as needed such as storage section 608.
Particularly, according to an embodiment of the invention, may be implemented as computer above with reference to the process of flow chart description
Software program.For example, the embodiment of the present invention includes a kind of computer readable storage medium, it is stored thereon with computer program,
The computer program realizes following step when being executed by processor:
The mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, the process of multiple pre-configurations
Node executes multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple states, each shape
State includes: existing state, condition, movement and next state；
The session process of client feedback is obtained, the session process includes multiple flow nodes marks and node-flow
To；
Corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；
The session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
Finite state machine model table, as session frame.
As can be seen from the above description, computer readable storage medium provided in an embodiment of the present invention, can be used for constructing dialog box
Frame can quickly generate session frame by pushing the alternate node of multiple pre-configurations to user, in addition, for different
Session, can be multiplexed the alternate node of pre-configuration, greatly saved the development time and exploited natural resources, improved development rate,
User only need to be dragged dotted line connection and trained or test data is circulated into server by dilatory, can create and train satisfaction
The dialogue robot of demand, can be online directly in engineering, and robot is allowed to become as the dialogue with people constantly learns
It is smarter more intelligent, flexible customization and efficiently scheduling are realized, quickly and easily can talk with robot by batch " production ".
In such embodiments, which can be downloaded and installed from network by communications portion 609,
And/or it is mounted from detachable media 611.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when application.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The application can describe in the general context of computer-executable instructions executed by a computer, such as program
Module.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, group
Part, data structure etc..The application can also be practiced in a distributed computing environment, in these distributed computing environments, by
Task is executed by the connected remote processing devices of communication network.In a distributed computing environment, program module can be with
In the local and remote computer storage media including storage equipment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (16)
1. a kind of dialog box framework construction method based on machine learning characterized by comprising
The mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, the flow nodes of multiple pre-configurations
Multiple and different operations is executed respectively, and the flow nodes information of each pre-configuration includes multiple states, and each state is equal
It include: existing state, condition, movement and next state；
The session process of client feedback is obtained, the session process includes multiple flow nodes marks and node flow direction；
Corresponding flow nodes information is obtained according to multiple flow nodes mark in the session process；
The limited of the session process is obtained according to the flow nodes information that the node in the session process flows to tissue acquisition
State machine model table, as session frame.
2. the dialog box framework construction method according to claim 1 based on machine learning, which is characterized in that the existing state is
Talk with the state being presently in；One movement of triggering or state transition of execution, the next state when the condition is satisfied
The new state to be moved to after meeting for condition；
The flow nodes information that the node according in the session process flows to tissue acquisition obtains the session process
Finite state machine model table, as session frame, comprising:
The flow nodes information of acquisition is ranked up according to the node flow direction in the session process；
The next state of each state in flow nodes information after configuration sequence, carries out to flow to according to the node in session process
Node jumps.
3. the dialog box framework construction method according to claim 2 based on machine learning, which is characterized in that the next state packet
Include: address field is used to indicate the new state to be moved to after condition meets；
The next state of each state in flow nodes information after the configuration sequence, comprising:
It will be in the address field in the next state of the jump address insertion present node most end state of latter node.
4. the dialog box framework construction method according to claim 1 based on machine learning, which is characterized in that in the movement
It include: that rule template identifies and/or handle model identification,
The dialog box framework construction method further include:
Corresponding regular mould is called from database according to included rule template mark is respectively acted in limit state machine model table
Plate；
Corresponding processing mould is called from database according to included processing model identification is respectively acted in limit state machine model table
Type.
5. the dialog box framework construction method according to claim 4 based on machine learning, which is characterized in that the processing mould
Type includes: participle model, textual classification model or elements recognition model.
6. the dialog box framework construction method according to claim 4 based on machine learning, which is characterized in that further include:
The model training process of the model is called from model training engine according to the processing model identification；
Obtain the model training data through marking；
The processing model is trained according to the model training data through marking and the model training process.
7. the dialog box framework construction method according to claim 6 based on machine learning, which is characterized in that further include:
Obtain the model measurement data through marking；
The processing model is tested using the model measurement data through marking.
8. a kind of session framework establishment device based on machine learning characterized by comprising
Alternate node display module, the mark for sending the flow nodes of multiple pre-configurations to client is selected for user configuration, more
The flow nodes of a pre-configuration execute multiple and different operations respectively, and the flow nodes information of each pre-configuration includes multiple shapes
State, each state include: existing state, condition, movement and next state；
Session process receiving module, obtains the session process of client feedback, and the session process includes multiple flow nodes marks
Know and node flows to；
Flow nodes data obtaining module obtains corresponding process section according to multiple flow nodes mark in the session process
Point information；
Dialog box framework models block, obtains institute according to the flow nodes information that the node in the session process flows to tissue acquisition
The finite state machine model table for stating session process, as session frame.
9. the session framework establishment device according to claim 8 based on machine learning, which is characterized in that the existing state is
Talk with the state being presently in；One movement of triggering or state transition of execution, the next state when the condition is satisfied
The new state to be moved to after meeting for condition；
The dialog box framework models block
Framework organizational unit is ranked up the flow nodes information of acquisition according to the node flow direction in the session process；
Configuration unit, the next state of each state in flow nodes information after configuration sequence, so as to according to the section in session process
Point flow direction carries out node and jumps.
10. the session framework establishment device according to claim 9 based on machine learning, which is characterized in that the next state
Include: address field, is used to indicate the new state to be moved to after condition meets；
The configuration unit includes:
Configuration subelement is jumped, by the address field in the next state of the jump address insertion present node most end state of latter node
In.
11. the session framework establishment device according to claim 8 based on machine learning, which is characterized in that the movement
In include: rule template mark and/or processing model identification,
The session framework establishment device further include:
Rule template calling module is identified from database according to included rule template is respectively acted in limit state machine model table
Call corresponding rule template；
Model calling module is handled, respectively acts included processing model identification from database according in limit state machine model table
Call corresponding processing model.
12. the session framework establishment device according to claim 11 based on machine learning, which is characterized in that the processing
Model includes: participle model, textual classification model or elements recognition model.
13. the session framework establishment device according to claim 11 based on machine learning, which is characterized in that further include:
Model training process obtains module, calls the model of the model from model training engine according to the processing model identification
Training process；
Training sample obtains module, obtains the model training data through marking；
Model training module, according to the model training data through marking and the model training process to the processing model
It is trained.
14. the session framework establishment device according to claim 13 based on machine learning, which is characterized in that further include:
Test sample obtains module, obtains the model measurement data through marking；
Model measurement module tests the processing model using the model measurement data through marking.
15. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes that claim 1 to 7 is described in any item and is based on machine when executing described program
The step of dialog box framework construction method of device study.
16. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program quilt
The step of claim 1 to 7 described in any item dialog box framework construction methods based on machine learning are realized when processor executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910694635.9A CN110413758B (en) | 2019-07-30 | 2019-07-30 | Session framework construction method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910694635.9A CN110413758B (en) | 2019-07-30 | 2019-07-30 | Session framework construction method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110413758A true CN110413758A (en) | 2019-11-05 |
| CN110413758B CN110413758B (en) | 2022-03-25 |
Family
ID=68364196
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910694635.9A Active CN110413758B (en) | 2019-07-30 | 2019-07-30 | Session framework construction method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110413758B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111241257A (en) * | 2020-01-03 | 2020-06-05 | 联想(北京)有限公司 | Information processing method and electronic device |
| CN111309603A (en) * | 2020-02-06 | 2020-06-19 | 北京奇艺世纪科技有限公司 | Task flow information storage method and device, electronic equipment and storage medium |
| CN111563151A (en) * | 2020-05-07 | 2020-08-21 | 腾讯科技（深圳）有限公司 | Information acquisition method, session configuration device and storage medium |
| CN111638871A (en) * | 2020-05-28 | 2020-09-08 | 深圳市元征科技股份有限公司 | Equipment information processing method, device, equipment and medium |
| CN112231027A (en) * | 2020-09-27 | 2021-01-15 | 中国建设银行股份有限公司 | Task type session configuration method and system |
| CN112363869A (en) * | 2020-11-19 | 2021-02-12 | 中国核动力研究设计院 | Nuclear DCS (distributed control System) backtracking method and system based on three-state variable finite state machine |
| CN112767930A (en) * | 2019-11-06 | 2021-05-07 | 苏州思必驰信息科技有限公司 | Voice skill dialog manager configuration method, man-machine dialog method and system |
| CN113379590A (en) * | 2021-06-07 | 2021-09-10 | 腾讯科技（上海）有限公司 | Animation data processing method, animation data processing device, computer equipment and storage medium |
| CN113609831A (en) * | 2021-08-05 | 2021-11-05 | 北京金堤科技有限公司 | Method and device for generating product delivery flow template |
| CN114221991A (en) * | 2021-11-08 | 2022-03-22 | 梅瑞生 | Big data-based session recommendation feedback processing method and deep learning service system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109514586A (en) * | 2019-01-30 | 2019-03-26 | 第四范式（北京）技术有限公司 | Realize the method and system of intelligent customer service robot |
| CN109743346A (en) * | 2018-11-19 | 2019-05-10 | 众安信息技术服务有限公司 | Session process configuration method, device and computer readable storage medium |
| US20190220774A1 (en) * | 2015-01-23 | 2019-07-18 | Conversica, Inc. | Systems and methods for configuring message exchanges in machine learning conversations |
-
2019
- 2019-07-30 CN CN201910694635.9A patent/CN110413758B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190220774A1 (en) * | 2015-01-23 | 2019-07-18 | Conversica, Inc. | Systems and methods for configuring message exchanges in machine learning conversations |
| CN109743346A (en) * | 2018-11-19 | 2019-05-10 | 众安信息技术服务有限公司 | Session process configuration method, device and computer readable storage medium |
| CN109514586A (en) * | 2019-01-30 | 2019-03-26 | 第四范式（北京）技术有限公司 | Realize the method and system of intelligent customer service robot |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112767930A (en) * | 2019-11-06 | 2021-05-07 | 苏州思必驰信息科技有限公司 | Voice skill dialog manager configuration method, man-machine dialog method and system |
| CN111241257A (en) * | 2020-01-03 | 2020-06-05 | 联想(北京)有限公司 | Information processing method and electronic device |
| CN111241257B (en) * | 2020-01-03 | 2023-07-21 | 联想(北京)有限公司 | Information processing method and electronic device |
| CN111309603A (en) * | 2020-02-06 | 2020-06-19 | 北京奇艺世纪科技有限公司 | Task flow information storage method and device, electronic equipment and storage medium |
| CN111563151A (en) * | 2020-05-07 | 2020-08-21 | 腾讯科技（深圳）有限公司 | Information acquisition method, session configuration device and storage medium |
| CN111563151B (en) * | 2020-05-07 | 2024-02-02 | 腾讯科技（深圳）有限公司 | Information acquisition method, session configuration method, device and storage medium |
| CN111638871A (en) * | 2020-05-28 | 2020-09-08 | 深圳市元征科技股份有限公司 | Equipment information processing method, device, equipment and medium |
| CN112231027A (en) * | 2020-09-27 | 2021-01-15 | 中国建设银行股份有限公司 | Task type session configuration method and system |
| CN112363869B (en) * | 2020-11-19 | 2022-03-01 | 中国核动力研究设计院 | Nuclear DCS (distributed control System) backtracking method and system based on three-state variable finite state machine |
| CN112363869A (en) * | 2020-11-19 | 2021-02-12 | 中国核动力研究设计院 | Nuclear DCS (distributed control System) backtracking method and system based on three-state variable finite state machine |
| CN113379590A (en) * | 2021-06-07 | 2021-09-10 | 腾讯科技（上海）有限公司 | Animation data processing method, animation data processing device, computer equipment and storage medium |
| CN113379590B (en) * | 2021-06-07 | 2023-06-30 | 腾讯科技（上海）有限公司 | Animation data processing method, device, computer equipment and storage medium |
| CN113609831A (en) * | 2021-08-05 | 2021-11-05 | 北京金堤科技有限公司 | Method and device for generating product delivery flow template |
| CN114221991A (en) * | 2021-11-08 | 2022-03-22 | 梅瑞生 | Big data-based session recommendation feedback processing method and deep learning service system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110413758B (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110413758A (en) | Dialog box framework construction method and device based on machine learning | |
| US20180025726A1 (en) | Creating coordinated multi-chatbots using natural dialogues by means of knowledge base | |
| Puttonen et al. | Semantics-based composition of factory automation processes encapsulated by web services | |
| CN108804643A (en) | Flow dispensing unit, method and the intelligent robot interactive system of more wheel sessions | |
| US11888941B2 (en) | Function-as-a-service for two-way communication systems | |
| CN110377720A (en) | The more wheel exchange methods of intelligence and system | |
| CN108306804A (en) | A kind of Ethercat main station controllers and its communication means and system | |
| Hoffmann | Smart agents for the Industry 4.0 | |
| Cui et al. | Analysis of Service‐Oriented Architecture and Scrum Software Development Approach for IIoT | |
| Kozma et al. | Achieving flexible digital production with the arrowhead workflow choreographer | |
| CN109118151B (en) | Work order transaction processing method and work order transaction processing system | |
| Van den Heuvel et al. | Software service engineering: Tenets and challenges | |
| Wolf | Succeedings of the second international software architecture workshop (isaw-2) | |
| CN113468303B (en) | Dialogue interaction processing method and computer-readable storage medium | |
| Hoffmann et al. | Web Service Composition. | |
| Ribeiro et al. | Multiagent mechatronic systems with simulation on the loop | |
| Mostafa et al. | Intent profiling and translation through emergent communication | |
| AU2023241384B2 (en) | Function-as-a-service cloud chatbot for two-way communication systems | |
| McGinnis et al. | Realizing agent dialogues with distributed protocols | |
| US11516091B2 (en) | Cloud infrastructure planning assistant via multi-agent AI | |
| KR102635363B1 (en) | Code generation system based on interactive chatbot | |
| Kazanskaia et al. | Knowledge-driven adaptive production management based on real-time user feedback and ontology updates | |
| Huang et al. | Implementation of Seamless Assistance with Google Assistant Leveraging Cloud Computing | |
| Visuwasam et al. | Innovative Turned and Collaborative Technology using Simulated IoT Applications | |
| Gharbi et al. | Functional safety of adaptive embedded control systems: new solutions |
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 |