CN110347817A - Intelligent response method and device, storage medium, electronic equipment - Google Patents
Intelligent response method and device, storage medium, electronic equipment Download PDFInfo
- Publication number
- CN110347817A CN110347817A CN201910637337.6A CN201910637337A CN110347817A CN 110347817 A CN110347817 A CN 110347817A CN 201910637337 A CN201910637337 A CN 201910637337A CN 110347817 A CN110347817 A CN 110347817A
- Authority
- CN
- China
- Prior art keywords
- information
- user
- behavior
- attribute
- behavioural
- 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
- 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
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present invention is about a kind of intelligent response method and device, storage medium, electronic equipment, belong to field of artificial intelligence, this method comprises: receiving the dialogue interactive information of the first user input, and obtain historical session information corresponding with first user;And the corresponding behavioural information of current time information of the dialogue interactive information generation and attribute information corresponding with the behavioural information is obtained from preset time behavior table;The dialogue interactive information, historical session information, behavioural information corresponding with the current time information and attribute information are input in behavior perception dialog model, response message is generated.The embodiment of the present invention improves the generation quantity of response message while improving the accuracy rate of response message.
Description
Technical field
The present embodiments relate to field of artificial intelligence, answer in particular to a kind of intelligent response method, intelligence
Answer device, computer readable storage medium and electronic equipment.
Background technique
After the proposition of seq2seq frame, that is, it is widely used in dialogue and generates.Seq2seq is that the input to user carries out
Coding, is then passed to decoding end for the intermediate representation of coding and carries out generation output.But seq2seq frame is easy to generate and lead to
And inconsistent reply.Wherein, versatility generally refer to reply content do not have information content, such as " ha ha ha " with
And " I is also " etc;Inconsistency refers to that, for identical problem, robot may reply conflicting answer twice,
Such as " how old are you ", the reply of different time may export two kinds of " I 16 " and " I 18 ".Therefore, how to be promoted
The diversity generated and consistency are replied into problem to be solved.
During studying diversity and consistency, everybody gradually wishes that it is personalized for talking with robot, that is,
Everybody can set the personality of robot oneself, age, hobby, occupation etc., and the reply generated is to meet robot individual character
's.In the existing modeling for customized information generates, most of is the mode based on template.For example, generally given machine
The individual character of device people is set, and artificial finishes writing some question and answer templates, is then carried out template matching every time to the input of user, is reached one
Determine threshold value and is returned to the template answer finished writing.
But on the one hand aforesaid way has following defects that, carry out question and answer by then passing through template, therefore can reply
Answer negligible amounts, and can cause matched accuracy rate lower due to artificial and then lead to the accuracy rate of session
It is lower;On the other hand, with the raising of user individual degree, the above method can consume a large amount of human resources, and more
It is time-consuming.
Accordingly, it is desirable to provide a kind of new intelligent response method and device.
It should be noted that the information in the invention of above-mentioned background technology part is only used for reinforcing the reason to background of the invention
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
The purpose of the present invention is to provide a kind of intelligent response methods, intelligent response device, computer readable storage medium
And electronic equipment, and then the standard of session caused by the limitation and defect due to the relevant technologies is overcome at least to a certain extent
The lower problem of true rate.
According to one aspect of the disclosure, a kind of intelligent response method is provided, comprising:
The dialogue interactive information of the first user input is received, and obtains historical session letter corresponding with first user
Breath;
Behavior corresponding with the current time information that the dialogue interactive information generates is obtained from preset time behavior table
Information and attribute information corresponding with the behavioural information;
By the dialogue interactive information, historical session information, behavioural information corresponding with the current time information and
Attribute information is input in behavior perception dialog model, generates response message.
In a kind of exemplary embodiment of the disclosure, obtained and the dialogue interactive information from preset time behavior table
Before the corresponding behavioural information of the current time information of generation and attribute information corresponding with the behavioural information, the intelligence
Answer method further include:
Classify to the daily behavior data of first user, and adds packet for the daily behavior data of each classification
Include the goal behavior data of status attribute;
According to the first preset time period belonging to the goal behavior data, the preset time behavior table is constructed.
In a kind of exemplary embodiment of the disclosure, the status attribute include keyword and with the keyword pair
The attribute value answered.
In a kind of exemplary embodiment of the disclosure, according to the first preset time belonging to the goal behavior data
Section, constructing the preset time behavior table includes:
The second preset time period is divided according to preset rules, obtains multiple first preset time periods;
According to the first preset time period belonging to the goal behavior data, the goal behavior data are filled to default
In period.
In a kind of exemplary embodiment of the disclosure, letter is interacted with the dialogue obtaining from preset time behavior table
After ceasing the corresponding behavioural information of current time information generated and attribute information corresponding with the behavioural information, the intelligence
Energy answer method further include:
It is sampled to obtain the first sampling results to the status attribute of the goal behavior data of first user, and according to
First sampling results determine the keyword and attribute value corresponding with the keyword of first user;
According to the keyword of first user and attribute value corresponding with the keyword, first user is determined
Personalized behavioural information set.
In a kind of exemplary embodiment of the disclosure, will talk with the dialogue interactive information, the historical session information,
Behavioural information corresponding with the current time information and attribute information are input in behavior perception dialog model, generate response
Information includes:
The dialogue interactive information, the historical session information, behavior corresponding with the current time information will be talked with
Information, attribute information and the personalized behavioural information set of first user are input in behavior perception dialog model, raw
At response message.
In a kind of exemplary embodiment of the disclosure, the intelligent response method further include:
It is carried out in the first preset time period belonging to the goal behavior data being sampled again and/or multiple sampling obtains
One or more second sampling results;
According to second sampling results determine corresponding with first user second user keyword and with institute
State the corresponding attribute value of keyword;
According to the keyword of the second user and attribute value corresponding with the keyword, the second user is determined
Personalized behavioural information set.
In a kind of exemplary embodiment of the disclosure, will talk with the dialogue interactive information, the historical session information,
Behavioural information corresponding with the current time information and attribute information are input in behavior perception dialog model, generate response
Information includes:
The dialogue interactive information, the historical session information, behavior corresponding with the current time information will be talked with
Information, attribute information, the personalized behavioural information of the personalized behavioural information of first user and the second user are defeated
Enter into behavior perception dialog model, generates the response message between first user and the second user.
According to one aspect of the disclosure, a kind of intelligent response device is provided, comprising:
Information receiving module for receiving the dialogue interactive information of the first user input, and obtains and first user
Corresponding historical session information;
Data obtaining module, for from preset time behavior table obtain with it is described dialogue interactive information generate it is current when
Between the corresponding behavioural information of information and attribute information corresponding with the behavioural information;
Information generating module is used for the dialogue interactive information, historical session information and the current time information pair
The behavioural information and attribute information answered are input in behavior perception dialog model, generate response message.
According to one aspect of the disclosure, a kind of computer readable storage medium is provided, computer program is stored thereon with,
The computer program realizes intelligent response method described in above-mentioned any one embodiment when being executed by processor.
According to one aspect of the disclosure, a kind of electronic equipment is provided, comprising:
Processor;And
Memory, for storing the executable instruction of the processor;
Wherein, the processor is configured to execute described in above-mentioned any one embodiment via the executable instruction is executed
Intelligent response method.
A kind of intelligent response method and device of the embodiment of the present invention, on the one hand, by the dialogue for receiving the first user input
Interactive information, and obtain historical session information corresponding with the first user;It obtains from preset time behavior table and is handed over dialogue again
The corresponding behavioural information of current time information and attribute information corresponding with behavioural information that mutual information generates;It finally will dialogue
Interactive information, historical session information, behavioural information corresponding with current time information and attribute information are input to behavior perception
In dialog model, response message is generated;It solves and is needed in the prior art by manually finishing writing question and answer template, and then result in the need for
Consume a large amount of human resources, and more time-consuming problem;On the other hand, by the way that interactive information, historical session letter will be talked with
Breath, behavioural information corresponding with current time information and attribute information are input in behavior perception dialog model, generate response
Information solves and carries out question and answer, therefore the negligible amounts for the answer that can be replied by then passing through template in the prior art, and
The problem that matched accuracy rate is lower and then causes the accuracy rate of session lower can be caused due to artificial, improve response
The generation quantity of information improves the accuracy rate of response message simultaneously;In another aspect, by the way that interactive information, historical session will be talked with
Information, behavioural information corresponding with current time information and attribute information are input in behavior perception dialog model, and generation is answered
Information is answered, improves the personalization of response message, while also the user experience is improved.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
It can the limitation present invention.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and be used to explain the principle of the present invention together with specification.It should be evident that the accompanying drawings in the following description is only the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 schematically shows a kind of flow chart of intelligent response method of example embodiment according to the present invention.
Fig. 2 schematically shows a kind of first according to belonging to the goal behavior data of example embodiment according to the present invention
Preset time period constructs the method flow diagram of the preset time behavior table.
Fig. 3 schematically shows the flow chart of another intelligent response method of example embodiment according to the present invention.
Fig. 4 schematically shows the flow chart of another intelligent response method of example embodiment according to the present invention.
Fig. 5 schematically shows a kind of block diagram of intelligent response device of example embodiment according to the present invention.
Fig. 6 schematically shows a kind of electronics for realizing above-mentioned intelligent response method of example embodiment according to the present invention
Equipment.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein;On the contrary, thesing embodiments are provided so that the present invention will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.Described feature, knot
Structure or characteristic can be incorporated in any suitable manner in one or more embodiments.In the following description, it provides perhaps
More details fully understand embodiments of the present invention to provide.It will be appreciated, however, by one skilled in the art that can
It is omitted with practicing technical solution of the present invention one or more in the specific detail, or others side can be used
Method, constituent element, device, step etc..In other cases, be not shown in detail or describe known solution to avoid a presumptuous guest usurps the role of the host and
So that each aspect of the present invention thickens.
In addition, attached drawing is only schematic illustrations of the invention, it is not necessarily drawn to scale.Identical attached drawing mark in figure
Note indicates same or similar part, thus will omit repetition thereof.Some block diagrams shown in the drawings are function
Energy entity, not necessarily must be corresponding with physically or logically independent entity.These function can be realized using software form
Energy entity, or these functional entitys are realized in one or more hardware modules or integrated circuit, or at heterogeneous networks and/or place
These functional entitys are realized in reason device device and/or microcontroller device.
In a kind of versatility reply scheme from seq2seq frame, the content for generally referring to reply does not have information content, example
Such as " ha ha ha ", " I is also " etc, inconsistency refers to that, for identical problem, robot may reply mutual lance twice
The answer of shield, such as " how old are you ", the reply of different time may export two kinds of " I 16 " and " I 18 ", so
More and more people, which begin one's study, promotes the diversity that reply generates, consistency.It is generated since seq2seq frame is easy general
, inconsistent reply, therefore how to generate the problem of having reverted back to urgent need to resolve of multifarious and self-consistent property.
During studying diversity and consistency, most users gradually wish that it is personalized for talking with robot.
That is, user can set the personality of robot oneself, age, hobby, occupation etc., and the reply generated is to meet machine
Device people's individual character.Personalized generation development experience from the beginning only to the embedded encoded of personage, meet knot to probing into
Structure people sets the reply in table (name, age, place etc.), then meets daily individual information to some (" liking having tea ", " rides
The setting of Che Qu school etc ") description reply.
But when the usually chat of people, more focused on the behavioural information at other side's current time, and everyone different time
Behavior be all various, so establishing one more meets that the movable chat robots of daily behavior are significant and have must
It wants.
And the modeling of customized information is generated, be mainly the following form: one is the modes based on template.
The individual character setting of general given robot, artificial finishes writing some question and answer templates, then carries out template every time to the input of user
Matching, reaches certain threshold value and is returned to the template answer finished writing.Another kind be different individuals is done it is embedded encoded, implicitly
Be added to dialogue generate in.Indicate to represent different users with different insertion vectors, then decoding end by this to
Amount and decoding terminal word vector splicing, are then sent through in decoder module, then train, and it is expected that the user is embedded in vector and can influence and change
Become the result that dialogue generates.Another is the individual character setting for giving robot, provides the reply that can meet setting.Such as it is given
The problem related to setting is replied by the name of robot, age, place etc., robot.Current technology has trial to generation
When judged, need when the individual information with robot then to go guidance to generate with individual information.Also, given machine
The description of the individual character of people needs its generation that can meet the answer of description setting.Such as the data of the match Facebook of ConvAI2
Collection, the customized information that two players have oneself long-term, determining, both sides chat according to personalized information.Pass through this
The data of sample allow model go study how generate when use customized information.Relatively good model respectively believes personalization
Breath and dialog history are encoded, and are then decoded in decoding end.
But on the one hand the above method has following defects that, the method based on template, it is necessary to have the language artificially finished writing
Material, and reply template and all finish writing, so replying space very little;In addition, once personalized information becomes complex,
Method based on template just consumes manpower very much, and matching error has certain rising.On the other hand, at present to personalized exploration
Some long-term customized informations are all built upon, but in actual chat, we often chat the behavior state of both sides, and
And different time behavior states is variation.And seldom go merely to set relevant content about identity, so by dynamically changeable
Behavior state be necessary as the research of the customized information of robot.In another aspect, current personalization machine people
Customized information can be structuring or non-structured information, and be directed at structuring and non-structural customized information
Reason is different, and how unified structure and unstructured customized information are also to need to study in a model.
A kind of intelligent response method is provided firstly in this example embodiment, this method can run on intelligent robot
Terminal;Certainly, those skilled in the art can also run method of the invention in other platforms according to demand, this exemplary implementation
Particular determination is not done to this in example.Refering to what is shown in Fig. 1, the intelligent response method may comprise steps of:
Step S110. receives the dialogue interactive information of the first user input, and obtains go through corresponding with first user
History session information.
Step S120. obtains the current time information pair generated with the dialogue interactive information from preset time behavior table
The behavioural information answered and attribute information corresponding with the behavioural information.
Step S130. is by the dialogue interactive information, historical session information, behavior corresponding with the current time information
Information and attribute information are input in behavior perception dialog model, generate response message.
In above-mentioned intelligent response method, on the one hand, by receive the first user input dialogue interactive information, and obtain with
The corresponding historical session information of first user;Again from preset time behavior table obtain with dialogue interactive information generate it is current when
Between the corresponding behavioural information of information and attribute information corresponding with behavioural information;Interactive information, historical session will finally be talked with
Information, behavioural information corresponding with current time information and attribute information are input in behavior perception dialog model, and generation is answered
Answer information;It solves and is needed in the prior art by manually finishing writing question and answer template, and then result in the need for consuming a large amount of manpower money
Source, and more time-consuming problem;On the other hand, by the way that interactive information, historical session information and current time information will be talked with
Corresponding behavioural information and attribute information are input in behavior perception dialog model, generate response message, solve existing skill
Question and answer, therefore the negligible amounts for the answer that can be replied are carried out by then passing through template in art, and can be due to artificial
The problem that matched accuracy rate is lower and then causes the accuracy rate of session lower is caused, the generation quantity for improving response message is same
When improve the accuracy rate of response message;In another aspect, by the way that interactive information, historical session information and current time will be talked with
The corresponding behavioural information of information and attribute information are input in behavior perception dialog model, are generated response message, are improved and answer
The personalization of information is answered, while also the user experience is improved.
In the following, by detailed solution is carried out to each step in intelligent response method above-mentioned in this example embodiment in conjunction with attached drawing
It releases and illustrates.
In step s 110, the dialogue interactive information of the first user input is received, and is obtained corresponding with first user
Historical session information.
In this exemplary embodiment, the first user (this sentences the first user and is illustrated for P1) can be with intelligent machine
Device people carries out interactive voice, can after intelligent robot receives dialogue interactive information (voice messaging) of the first user input
With start internal NLP (Natural Language Processing, natural language processing) module to talk with interactive information into
Row parsing, and historical session information corresponding with first user is obtained according to parsing result.Wherein, which can wrap
The session for including the first user is intended to (the problem of e.g. session robotic being needed to answer etc.), can also include first user
Acoustic information etc.;Specifically, can be matched by the acoustic information with the existing acoustic information of history, when matched
It, then can be using the corresponding session information of the existing acoustic information of the history as first user's after threshold value reaches preset threshold
Historical session information.
In the step s 120, the current time generated with the dialogue interactive information is obtained from preset time behavior table to believe
Cease corresponding behavioural information and attribute information corresponding with the behavioural information.
In this exemplary embodiment, in order to can in order to from preset time behavior table obtain behavioural information and attribute letter
Breath, it is necessary first to which preset time behavior table is constructed.Specifically, firstly, daily behavior data to first user
Classify, and includes the goal behavior data of status attribute for the addition of the daily behavior data of each classification;Secondly, according to institute
The first preset time period belonging to goal behavior data is stated, the preset time behavior table is constructed.Wherein, the status attribute packet
Include keyword and attribute value corresponding with the keyword.
For example, firstly, defining daily behavior first-level class.Specifically, daily behavior can be divided into following eight classes
Not, such as: cuisines go window-shopping, move, going on a tour, performing, amusement game, daily and study and work etc., then will be each specific
Behavior will all belong in one kind in these eighth types.Then, the behavior set in every class is designed.Specifically, for above-mentioned eight
Each of classification classification can enumerate common behavioral activity, for example have blocked shot below cuisines, eat sweets, is poly-
Meal, the following You Guang supermarket that goes window-shopping go shopping, stroll flower market etc..Further, it is also necessary to design the status attribute of each behavior.Example
Such as, for above-mentioned behavior, their applicable time (morning-A, noon-B, at night-C), the specific shape of each behavior can be designed
The attribute and value set of state information.It specifically can be as shown in table 1 below.
Table 1
Further, it after obtaining above-mentioned preset time behavior table, can be obtained from preset time behavior table and institute
State the corresponding behavioural information of current time information and attribute corresponding with behavioural information letter that dialogue interactive information generates
Breath.For example, when the current time information that dialog information generates is 14, available behavior letter corresponding with the temporal information
Breath: it has a meal;And attribute information corresponding with behavior information: source: dining room, food: home cooking, taste: general, companion:
Friend etc..
In step s 130, by the dialogue interactive information, historical session information, corresponding with the current time information
Behavioural information and attribute information are input in behavior perception dialog model, generate response message.
What in this exemplary embodiment, for example, being intended to according to the session that dialogue interactive information obtains are as follows: noon eats? it goes through
History session information are as follows: noon goes to dining room to eat rice;The response message then generated for example can be with are as follows: in go to dining room to eat home cooking,
Staple food eats rice;Or noon goes to dining room to have fast food;Or noon goes to the dining room XX to have a meal etc., it is special that this example does not do this
Limitation.Herein it should be noted that behavior perception dialog model can according to dialogue interactive information, historical session information, with work as
The corresponding behavioural information of preceding temporal information and attribute information generate one or more response message.If generating multiple response letters
Breath, then can assess according to user satisfaction of the historical session information to each response message, and according to the height of assessed value
It is ranked up, so that user selects.
Further, in this exemplary embodiment, which can also include: to perceive dialog model to behavior
It is trained, the input that the behavior perceives dialog model is the personalized behavioural information set and/or second user of the first user
Personalized behavioural information set;And first user and/or second user dialog history.Specifically, training process can be with
Including following two steps: it is possible, firstly, to carry out GPT (Generative Pre-trained based on large-scale Chinese corpus
Transformer) the pre-training of language model.Wherein, model has 12 layers, and the dimension for being embedded in vector is 768, dropout parameter
It is 0.1, using Chinese character as lexical unit when training, training obtains the model parameter of pre-training in 1-2 weeks on large-scale corpus.
Then, the dialog model of personalized behavior perception is finely adjusted.It, will be defeated in training corpus on parameter basis obtained above
If behavior state information, conversation history and the needs entered are replied, it is stitched together using separator complete as one
Sentence is sent in the good language model of pre-training and is finely adjusted training.Training is restrained until loss, then is used as final behavior sense
The dialog model known.
Fig. 2 schematically shows a kind of first according to belonging to the goal behavior data of example embodiment according to the present invention
Preset time period constructs the method flow diagram of the preset time behavior table.Refering to what is shown in Fig. 2, according to the goal behavior number
According to the first affiliated preset time period, constructing the preset time behavior table may include step S210 and step S220, with
Under be described in detail.
In step S210, the second preset time period is divided according to preset rules, it is pre- to obtain multiple described first
If the period.
In step S220, according to the first preset time period belonging to the goal behavior data, by the goal behavior
Data are filled into preset time period.
Hereinafter, will step S210 and step S220 be explained and be illustrated.Firstly, above-mentioned second preset time period
Such as can be one day, it is also possible to one week or 1 year, there is no special restriction on this for this example.Above-mentioned preset rules for example may be used
To be the working and resting arrangement of every day, it is also possible to job placement etc. in one week, there is no special restriction on this for this example.Specifically
, this example with the second preset time period be one day, preset rules be every day working and resting arrangement for be illustrated.
For example, firstly, carrying out Time segments division at the time of by one day, such as 0-7,7-12,12-15,15- can be divided into
18,18-24 obtain multiple first preset time periods;It then, is each first preset time period choosing period of time behavior.For example, may be used
To select suitable behavior to be filled in above-mentioned behavior set.Such as blocked shot is inserted in the 12-15 period, in 15-18
Period filling is gone shopping, and is each action selection attribute value, such as place is small fertile sheep in blocked shot behavior, and taste is
Somewhat peppery, companion is household, and service is general.It specifically can be as shown in table 2 below.
Table 2
Fig. 3 schematically shows the flow chart of another intelligent response method according to disclosure example embodiment.With reference to Fig. 3
Shown, which can also include step S310 and step S320, be described in detail below.
In step s310, the status attribute of the goal behavior data of first user is sampled to obtain the first pumping
Sample is as a result, and determine the keyword and category corresponding with the keyword of first user according to first sampling results
Property value.
In step s 320, according to the keyword of first user and attribute value corresponding with the keyword, really
The personalized behavioural information set of fixed first user.
Hereinafter, will step S310 and step S320 be explained and be illustrated.For example, to the target line of the first user
It is sampled to obtain the first sampling results for the status attribute of each behavior in the behavior set of data, then be taken out according to first
Sample result determines keyword and attribute value, forms the set of the personalized behavioural information of the first determining user;For example, crucial
Word is place, taste and companion;Attribute value are as follows: seabed fishing, very nice, household etc.;Therefore the first user is obtained eating
The set of personalized behavioural information under this classification of chafing dish are as follows: place: seabed fishing;Taste: very nice;Companion: household.
Further, after obtaining above-mentioned personalized behavioural information set, above-mentioned steps S130 can also include: will be right
Talk about the dialogue interactive information, the historical session information, behavioural information corresponding with the current time information, attribute information
And the personalized behavioural information set of first user is input in behavior perception dialog model, generates response message.It is logical
This method is crossed, can further improve the personalization level of response message, while can also be improved the accuracy rate of response message,
Promote the experience and satisfaction of user.
Fig. 4 schematically shows the flow chart of another intelligent response method according to disclosure example embodiment.With reference to Fig. 4
Shown, which can also include step S410- step S430, be described in detail below.
In step S410, carried out in the first preset time period belonging to the goal behavior data being sampled again and/
Or multiple sampling obtains one or more the second sampling results.
In the step s 420, the pass of second user corresponding with first user is determined according to second sampling results
Key word and attribute value corresponding with the keyword.
In step S430, according to the keyword of the second user and attribute value corresponding with the keyword, really
The personalized behavioural information set of the fixed second user.
It is progressive, after obtaining the personalized behavioural information set of second user, above-mentioned steps S130 can also include: by
Talk with the dialogue interactive information, the historical session information, behavioural information corresponding with the current time information, attribute letter
The personalized behavioural information of breath, the personalized behavioural information of first user and the second user is input to behavior perception
In dialog model, the response message between first user and the second user is generated.Specifically, the first user generated
Response message between second user for example can be as shown in table 3 below:
Table 3
Hereinafter, the intelligent response method that this example embodiment refers to is further explained and is illustrated.
Firstly, the daily behavior set of designing user.The common daily routines behavior set of the module design, and to each
The detailed state background of Activity design, these behaviors and their corresponding detailed status informations, when being exactly robot chat
Property information.
Secondly, time behavior table designs.Its personalized behavior schedule, schedule can be designed to each dialogue robot
The behavior of each period of table is chosen from daily behavior set.It is closer to real world by this design, each
Robot schedule behavior table all has differentiation.
Then, the dialogue corpus of Behavior-based control state background is collected.Two behaviors of sampling in daily behavior state set
Then state allows two mark persons to respectively represent a behavior state background as one group, and daily around behavior state progress
Change chat, obtains the more wheels being consistent with this group of behavior state dialogue corpus.
Further, the dialog model training of behavior perception.By more wheels of the behavior state to collection talk with corpus into
Row training makes it possible to reply the reply being consistent with the behavior state background currently set.
Finally, the chat robots application of time behavior perception.It, can be with according to the time behavior table of design at each moment
The personalized behavior setting state of current time robot is obtained, an input is often carried out, by input and personalized behavior state
It is sent to trained model, output one meets the reply of its behavior state.
In above-mentioned intelligent response method, detailed behavior state in different time periods can be designed for dialogue machine device people, and
The dialogue robot can generate corresponding reply according to the behavior state information of current time.It can bring following beneficial to effect
Fruit, first, dialogue robot can have the customized information of Time Perception, different from existing constant customized information, this
Variable, daily behavioural information can allow the reply mobilism of robot a bit.Second, the setting of customized information is according to the mankind
Live what daily behavioral activity was designed, can be closer to real-life, this can allow the reply of robot more to personalize,
Actualization.In short, the present invention can promote the anthropomorphic experience of chat robots, the reply generated is made to be closer to true feelings
Condition.
The disclosure additionally provides a kind of intelligent response device.Refering to what is shown in Fig. 5, the intelligent response device may include information
Receiving module 510, data obtaining module 520 and information generating module 530.Wherein:
Information receiving module 510 can be used for receiving the dialogue interactive information of the first user input, and obtain and described the
The corresponding historical session information of one user.
Data obtaining module 520 can be used for obtaining from preset time behavior table and the dialogue interactive information generation
The corresponding behavioural information of current time information and attribute information corresponding with the behavioural information.
Information generating module 530 can be used for the dialogue interactive information, historical session information and the current time
The corresponding behavioural information of information and attribute information are input in behavior perception dialog model, generate response message.
In a kind of example embodiment of the disclosure, obtained from preset time behavior table raw with the dialogue interactive information
At the corresponding behavioural information of current time information and attribute information corresponding with the behavioural information before, the intelligence is answered
Answer method further include:
Classify to the daily behavior data of first user, and adds packet for the daily behavior data of each classification
Include the goal behavior data of status attribute;
According to the first preset time period belonging to the goal behavior data, the preset time behavior table is constructed.
In a kind of example embodiment of the disclosure, the status attribute includes keyword and corresponding with the keyword
Attribute value.
In a kind of example embodiment of the disclosure, according to the first preset time period belonging to the goal behavior data,
Constructing the preset time behavior table includes:
The second preset time period is divided according to preset rules, obtains multiple first preset time periods;
According to the first preset time period belonging to the goal behavior data, the goal behavior data are filled to default
In period.
In a kind of example embodiment of the disclosure, the intelligent response device further include:
First decimation blocks can be used for being sampled the status attribute of the goal behavior data of first user
To the first sampling results, and according to first sampling results determine first user keyword and with the keyword
Corresponding attribute value;
First information determining module can be used for according to the keyword of first user and corresponding with the keyword
Attribute value, determine the personalized behavioural information set of first user.
In a kind of example embodiment of the disclosure, will talk with the dialogue interactive information, the historical session information, with
The corresponding behavioural information of the current time information and attribute information are input in behavior perception dialog model, generate response letter
Breath includes:
The dialogue interactive information, the historical session information, behavior corresponding with the current time information will be talked with
Information, attribute information and the personalized behavioural information set of first user are input in behavior perception dialog model, raw
At response message.
In a kind of example embodiment of the disclosure, the intelligent response device further include:
Second decimation blocks can be used in the first preset time period belonging to the goal behavior data being sampled carrying out
Again and/or multiple sampling obtains one or more the second sampling results;
Attribute value determining module can be used for determining corresponding with first user the according to second sampling results
The keyword of two users and attribute value corresponding with the keyword;
Second information determination module, according to the keyword of the second user and attribute corresponding with the keyword
Value, determines the personalized behavioural information set of the second user.
In a kind of example embodiment of the disclosure, will talk with the dialogue interactive information, the historical session information, with
The corresponding behavioural information of the current time information and attribute information are input in behavior perception dialog model, generate response letter
Breath includes:
The dialogue interactive information, the historical session information, behavior corresponding with the current time information will be talked with
Information, attribute information, the personalized behavioural information of the personalized behavioural information of first user and the second user are defeated
Enter into behavior perception dialog model, generates the response message between first user and the second user.
The detail of each module has carried out in corresponding intelligent response method in detail in above-mentioned intelligent response device
Thin description, therefore details are not described herein again.
It should be noted that although being referred to several modules or list for acting the equipment executed in the above detailed description
Member, but this division is not enforceable.In fact, embodiment according to the present invention, it is above-described two or more
Module or the feature and function of unit can embody in a module or unit.Conversely, an above-described mould
The feature and function of block or unit can be to be embodied by multiple modules or unit with further division.
In addition, although describing each step of method in the present invention in the accompanying drawings with particular order, this does not really want
These steps must be executed in this particular order by asking or implying, or having to carry out step shown in whole could realize
Desired result.Additional or alternative, it is convenient to omit multiple steps are merged into a step and executed by certain steps, and/
Or a step is decomposed into execution of multiple steps etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the present invention
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, mobile terminal or network equipment etc.) executes embodiment according to the present invention
Method.
In an exemplary embodiment of the present invention, a kind of electronic equipment that can be realized the above method is additionally provided.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, it may be assumed that complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as circuit, " module " or " system ".
The electronic equipment 600 of this embodiment according to the present invention is described referring to Fig. 6.The electronics that Fig. 6 is shown
Equipment 600 is only an example, should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in fig. 6, electronic equipment 600 is showed in the form of universal computing device.The component of electronic equipment 600 can wrap
It includes but is not limited to: at least one above-mentioned processing unit 610, at least one above-mentioned storage unit 620, the different system components of connection
The bus 630 of (including storage unit 620 and processing unit 610).
Wherein, the storage unit is stored with program code, and said program code can be held by the processing unit 610
Row, so that various according to the present invention described in the execution of the processing unit 610 above-mentioned " illustrative methods " part of this specification
The step of illustrative embodiments.For example, the processing unit 610 can execute step S110 as shown in fig. 1: receiving the
The dialogue interactive information of one user input, and obtain historical session information corresponding with first user;Step S120: from pre-
If obtained in time behavior table with the corresponding behavioural information of current time information of the dialogue interactive information generation and with institute
State the corresponding attribute information of behavioural information;Step S130: by the dialogue interactive information, historical session information, with it is described current
The corresponding behavioural information of temporal information and attribute information are input in behavior perception dialog model, generate response message.
Storage unit 620 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 6201 and/or cache memory unit 6202, it can further include read-only memory unit (ROM) 6203.
Storage unit 620 can also include program/utility with one group of (at least one) program module 6205
6204, such program module 6205 includes but is not limited to: operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.
Bus 630 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 600 can also be with one or more external equipments 700 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 600 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 600 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 650.Also, electronic equipment 600 can be with
By network adapter 660 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.As shown, network adapter 660 is communicated by bus 630 with other modules of electronic equipment 600.
It should be understood that although not shown in the drawings, other hardware and/or software module can not used in conjunction with electronic equipment 600, including but not
Be limited to: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and
Data backup storage system etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the present invention
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, terminal installation or network equipment etc.) executes embodiment according to the present invention
Method.
In an exemplary embodiment of the present invention, a kind of computer readable storage medium is additionally provided, energy is stored thereon with
Enough realize the program product of this specification above method.In some possible embodiments, various aspects of the invention may be used also
In the form of being embodied as a kind of program product comprising program code, when described program product is run on the terminal device, institute
Program code is stated for executing the terminal device described in above-mentioned " illustrative methods " part of this specification according to this hair
The step of bright various illustrative embodiments.
The program product for realizing the above method of embodiment according to the present invention can use Portable, compact
Disk read-only memory (CD-ROM) and including program code, and can be run on terminal device, such as PC.However,
Program product of the invention is without being limited thereto, and in this document, readable storage medium storing program for executing, which can be, any includes or storage program has
Shape medium, the program can be commanded execution system, device or device use or in connection.
Described program product can be using any combination of one or more readable mediums.Readable medium can be readable letter
Number medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or
System, device or the device of semiconductor, or any above combination.The more specific example of readable storage medium storing program for executing is (non exhaustive
List) include: electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory
(CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable signal medium can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.
The program code for including on readable medium can transmit with any suitable medium, including but not limited to wirelessly, have
Line, optical cable, RF etc. or above-mentioned any appropriate combination.
The program for executing operation of the present invention can be write with any combination of one or more programming languages
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It calculates and executes in equipment, partly executes on a user device, being executed as an independent software package, partially in user's calculating
Upper side point is executed on a remote computing or is executed in remote computing device or server completely.It is being related to far
Journey calculates in the situation of equipment, and remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
In addition, above-mentioned attached drawing is only the schematic theory of processing included by method according to an exemplary embodiment of the present invention
It is bright, rather than limit purpose.It can be readily appreciated that the time that above-mentioned processing shown in the drawings did not indicated or limited these processing is suitable
Sequence.In addition, be also easy to understand, these processing, which can be, for example either synchronously or asynchronously to be executed in multiple modules.
Those skilled in the art will readily occur to of the invention its after considering specification and the invention invented here of practice
His embodiment.This application is intended to cover any variations, uses, or adaptations of the invention, these modifications, purposes or
Adaptive change follow general principle of the invention and the common knowledge in the art do not invented including the present invention or
Conventional techniques.The description and examples are only to be considered as illustrative, and true scope and spirit of the invention are by claim
It points out.
Claims (11)
1. a kind of intelligent response method characterized by comprising
The dialogue interactive information of the first user input is received, and obtains historical session information corresponding with first user;
Behavioural information corresponding with the current time information that the dialogue interactive information generates is obtained from preset time behavior table
And attribute information corresponding with the behavioural information;
By the dialogue interactive information, historical session information, behavioural information corresponding with the current time information and attribute
Information input generates response message into behavior perception dialog model.
2. intelligent response method according to claim 1, which is characterized in that from preset time behavior table obtain with it is described
Talk with the corresponding behavioural information of current time information and attribute information corresponding with the behavioural information that interactive information generates
Before, the intelligent response method further include:
Classify to the daily behavior data of first user, and includes shape for the addition of the daily behavior data of each classification
The goal behavior data of state attribute;
According to the first preset time period belonging to the goal behavior data, the preset time behavior table is constructed.
3. intelligent response method according to claim 2, which is characterized in that the status attribute include keyword and with
The corresponding attribute value of the keyword.
4. intelligent response method according to claim 2, which is characterized in that according to belonging to the goal behavior data
One preset time period, constructing the preset time behavior table includes:
The second preset time period is divided according to preset rules, obtains multiple first preset time periods;
According to the first preset time period belonging to the goal behavior data, the goal behavior data are filled to preset time
Duan Zhong.
5. intelligent response method according to claim 3, which is characterized in that in acquisition and institute from preset time behavior table
State the corresponding behavioural information of current time information and attribute corresponding with behavioural information letter that dialogue interactive information generates
After breath, the intelligent response method further include:
The status attribute of the goal behavior data of first user is sampled to obtain the first sampling results, and according to described
First sampling results determine the keyword and attribute value corresponding with the keyword of first user;
According to the keyword of first user and attribute value corresponding with the keyword, of first user is determined
Property behavioural information set.
6. intelligent response method according to claim 5, which is characterized in that the dialogue interactive information, described will be talked with
Historical session information, behavioural information corresponding with the current time information and attribute information are input to behavior perception dialogue mould
In type, generating response message includes:
To talk with the dialogue interactive information, the historical session information, behavioural information corresponding with the current time information,
The personalized behavioural information set of attribute information and first user are input in behavior perception dialog model, generate response
Information.
7. intelligent response method according to claim 5, which is characterized in that the intelligent response method further include:
It is carried out in the first preset time period belonging to the goal behavior data being sampled again and/or multiple sampling obtains one
Or multiple second sampling results;
According to second sampling results determine corresponding with first user second user keyword and with the pass
The corresponding attribute value of key word;
According to the keyword of the second user and attribute value corresponding with the keyword, of the second user is determined
Property behavioural information set.
8. intelligent response method according to claim 7, which is characterized in that the dialogue interactive information, described will be talked with
Historical session information, behavioural information corresponding with the current time information and attribute information are input to behavior perception dialogue mould
In type, generating response message includes:
To talk with the dialogue interactive information, the historical session information, behavioural information corresponding with the current time information,
The personalized behavioural information of attribute information, the personalized behavioural information of first user and the second user is input to row
To generate the response message between first user and the second user in perception dialog model.
9. a kind of intelligent response device characterized by comprising
Information receiving module for receiving the dialogue interactive information of the first user input, and obtains corresponding with first user
Historical session information;
Data obtaining module is believed for obtaining the current time generated with the dialogue interactive information from preset time behavior table
Cease corresponding behavioural information and attribute information corresponding with the behavioural information;
Information generating module, for by the dialogue interactive information, historical session information, corresponding with the current time information
Behavioural information and attribute information are input in behavior perception dialog model, generate response message.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
Claim 1-8 described in any item intelligent response methods are realized when being executed by processor.
11. a kind of electronic equipment characterized by comprising
Processor;And
Memory, for storing the executable instruction of the processor;
Wherein, the processor is configured to require 1-8 described in any item via executing the executable instruction and carry out perform claim
Intelligent response method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910637337.6A CN110347817B (en) | 2019-07-15 | 2019-07-15 | Intelligent response method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910637337.6A CN110347817B (en) | 2019-07-15 | 2019-07-15 | Intelligent response method and device, storage medium and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110347817A true CN110347817A (en) | 2019-10-18 |
CN110347817B CN110347817B (en) | 2022-03-18 |
Family
ID=68175358
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910637337.6A Active CN110347817B (en) | 2019-07-15 | 2019-07-15 | Intelligent response method and device, storage medium and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110347817B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111552779A (en) * | 2020-04-28 | 2020-08-18 | 深圳壹账通智能科技有限公司 | Man-machine conversation method, device, medium and electronic equipment |
CN112131367A (en) * | 2020-09-24 | 2020-12-25 | 民生科技有限责任公司 | Self-auditing man-machine conversation method, system and readable storage medium |
CN112989013A (en) * | 2021-04-30 | 2021-06-18 | 武汉龙津科技有限公司 | Conversation processing method and device, electronic equipment and storage medium |
CN116627261A (en) * | 2023-07-25 | 2023-08-22 | 安徽淘云科技股份有限公司 | Interaction method, device, storage medium and electronic equipment |
CN116881429A (en) * | 2023-09-07 | 2023-10-13 | 四川蜀天信息技术有限公司 | Multi-tenant-based dialogue model interaction method, device and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105094315A (en) * | 2015-06-25 | 2015-11-25 | 百度在线网络技术(北京)有限公司 | Method and apparatus for smart man-machine chat based on artificial intelligence |
US20160070695A1 (en) * | 2013-08-06 | 2016-03-10 | Nuance Communications, Inc. | Method And Apparatus For A Multi I/O Modality Language Independent User-Interaction Platform |
CN105654950A (en) * | 2016-01-28 | 2016-06-08 | 百度在线网络技术(北京)有限公司 | Self-adaptive voice feedback method and device |
CN106297789A (en) * | 2016-08-19 | 2017-01-04 | 北京光年无限科技有限公司 | The personalized interaction method of intelligent robot and interactive system |
CN107153685A (en) * | 2017-04-25 | 2017-09-12 | 竹间智能科技(上海)有限公司 | The cognitive approach and device remembered in interactive system based on timeline |
CN108124008A (en) * | 2017-12-20 | 2018-06-05 | 山东大学 | A kind of old man under intelligent space environment accompanies and attends to system and method |
WO2018175291A1 (en) * | 2017-03-20 | 2018-09-27 | Ebay Inc. | Detection of mission change in conversation |
CN109176535A (en) * | 2018-07-16 | 2019-01-11 | 北京光年无限科技有限公司 | Exchange method and system based on intelligent robot |
-
2019
- 2019-07-15 CN CN201910637337.6A patent/CN110347817B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160070695A1 (en) * | 2013-08-06 | 2016-03-10 | Nuance Communications, Inc. | Method And Apparatus For A Multi I/O Modality Language Independent User-Interaction Platform |
CN105094315A (en) * | 2015-06-25 | 2015-11-25 | 百度在线网络技术(北京)有限公司 | Method and apparatus for smart man-machine chat based on artificial intelligence |
CN105654950A (en) * | 2016-01-28 | 2016-06-08 | 百度在线网络技术(北京)有限公司 | Self-adaptive voice feedback method and device |
CN106297789A (en) * | 2016-08-19 | 2017-01-04 | 北京光年无限科技有限公司 | The personalized interaction method of intelligent robot and interactive system |
WO2018175291A1 (en) * | 2017-03-20 | 2018-09-27 | Ebay Inc. | Detection of mission change in conversation |
CN107153685A (en) * | 2017-04-25 | 2017-09-12 | 竹间智能科技(上海)有限公司 | The cognitive approach and device remembered in interactive system based on timeline |
CN108124008A (en) * | 2017-12-20 | 2018-06-05 | 山东大学 | A kind of old man under intelligent space environment accompanies and attends to system and method |
CN109176535A (en) * | 2018-07-16 | 2019-01-11 | 北京光年无限科技有限公司 | Exchange method and system based on intelligent robot |
Non-Patent Citations (4)
Title |
---|
DYLAN F. GLAS ETC.: "Personal service: A robot that greets people individually based on observed behavior patterns", 《 2013 8TH ACM/IEEE INTERNATIONAL CONFERENCE ON HUMAN-ROBOT INTERACTION (HRI)》 * |
LI XIAO ETC.: "The tuneup and integration of resources in Web-based learning", 《 PROCEEDINGS IEEE INTERNATIONAL CONFERENCE ON ADVANCED LEARNING TECHNOLOGIES》 * |
曾丽霞: "基于情境感知的智能穿戴设备交互设计研究——以IT白领人群的智能手环设计为例", 《中国优秀硕士学位论文全文数据库(信息科技辑)》 * |
高廷丽: "面向自然交互的多通道人机对话***中答句自动生成方法研究", 《第十届和谐人机环境联合学术会议论文集》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111552779A (en) * | 2020-04-28 | 2020-08-18 | 深圳壹账通智能科技有限公司 | Man-machine conversation method, device, medium and electronic equipment |
CN112131367A (en) * | 2020-09-24 | 2020-12-25 | 民生科技有限责任公司 | Self-auditing man-machine conversation method, system and readable storage medium |
CN112989013A (en) * | 2021-04-30 | 2021-06-18 | 武汉龙津科技有限公司 | Conversation processing method and device, electronic equipment and storage medium |
CN116627261A (en) * | 2023-07-25 | 2023-08-22 | 安徽淘云科技股份有限公司 | Interaction method, device, storage medium and electronic equipment |
CN116881429A (en) * | 2023-09-07 | 2023-10-13 | 四川蜀天信息技术有限公司 | Multi-tenant-based dialogue model interaction method, device and storage medium |
CN116881429B (en) * | 2023-09-07 | 2023-12-01 | 四川蜀天信息技术有限公司 | Multi-tenant-based dialogue model interaction method, device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110347817B (en) | 2022-03-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110347817A (en) | Intelligent response method and device, storage medium, electronic equipment | |
US10922866B2 (en) | Multi-dimensional puppet with photorealistic movement | |
CN106448670B (en) | Conversational system is automatically replied based on deep learning and intensified learning | |
CN108491514B (en) | Method and device for questioning in dialog system, electronic equipment and computer readable medium | |
CN110347792A (en) | Talk with generation method and device, storage medium, electronic equipment | |
CN108345692A (en) | A kind of automatic question-answering method and system | |
CN105975622A (en) | Multi-role intelligent chatting method and system | |
CN106020488A (en) | Man-machine interaction method and device for conversation system | |
CN109242164A (en) | Optimize method and device, the computer storage medium, electronic equipment in product path | |
US11308312B2 (en) | System and method for reconstructing unoccupied 3D space | |
CN111831798A (en) | Information processing method, information processing device, electronic equipment and computer readable storage medium | |
US10785489B2 (en) | System and method for visual rendering based on sparse samples with predicted motion | |
CN111538456A (en) | Human-computer interaction method, device, terminal and storage medium based on virtual image | |
CN108280218A (en) | A kind of flow system based on retrieval and production mixing question and answer | |
CN110147936A (en) | Service evaluation method, apparatus based on Emotion identification, storage medium | |
CN112204565A (en) | System and method for inferring scenes based on visual context-free grammar model | |
Kao et al. | Model of multi-turn dialogue in emotional chatbot | |
EP4181011A1 (en) | Method of generating response using utterance and apparatus therefor | |
CN113761156A (en) | Data processing method, device and medium for man-machine interaction conversation and electronic equipment | |
CN113342948A (en) | Intelligent question and answer method and device | |
CN116543082A (en) | Digital person generation method and device and digital person generation system | |
CN108053826A (en) | For the method, apparatus of human-computer interaction, electronic equipment and storage medium | |
CN107967304A (en) | Session interaction processing method, device and electronic equipment | |
Huang et al. | Applied to mobile multimedia intelligent speech system interactive topic guiding model | |
CN116010899A (en) | Multi-mode data processing and pre-training method of pre-training model and electronic equipment |
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 |