CN110019824A - Man-machine interaction method, the apparatus and system of knowledge based map - Google Patents
Man-machine interaction method, the apparatus and system of knowledge based map Download PDFInfo
- Publication number
- CN110019824A CN110019824A CN201710595078.6A CN201710595078A CN110019824A CN 110019824 A CN110019824 A CN 110019824A CN 201710595078 A CN201710595078 A CN 201710595078A CN 110019824 A CN110019824 A CN 110019824A
- Authority
- CN
- China
- Prior art keywords
- user
- knowledge
- map
- robot
- human
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003993 interaction Effects 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000013507 mapping Methods 0.000 claims abstract description 105
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 9
- 230000002452 interceptive effect Effects 0.000 claims description 34
- 230000000977 initiatory effect Effects 0.000 claims description 10
- 230000006399 behavior Effects 0.000 claims description 8
- 230000008451 emotion Effects 0.000 abstract description 17
- 238000001228 spectrum Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000234435 Lilium Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000036578 sleeping time Effects 0.000 description 1
- 235000011888 snacks Nutrition 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Manipulator (AREA)
Abstract
Man-machine interaction method, the apparatus and system of knowledge based map provided by the invention, method are as follows: whether detection user comes into the identification range of robot, if recognizing user in the identification range of robot, analyzes the identity of user；According to the identity of user, the personal knowledge map of user is found, the personal knowledge map of multiple users is stored in robot；According to the personal knowledge map of user, robot actively initiates to talk with to user, carries out human-computer interaction；In human-computer interaction process, the intention of user is analyzed, and a plurality of types of knowledge mappings pre-established is combined to interact with user, a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.The present invention is by designing a plurality of types of knowledge mappings, and the experience of user, makes robot understand the true intention of user during human-computer interaction when improving human-computer interaction, accomplishes to accompany the emotion of user, improves user experience.
Description
Technical field
The present invention relates to the man-machine interaction methods of artificial intelligence field more particularly to knowledge based map, apparatus and system
Field.
Background technique
Since the relevant technology of Google publication knowledge mapping in 2012 and research, many knowledge mappings are relevant to be ground
Study carefully, books, library etc. are published and are paid attention to.But current knowledge mapping is studied and is absorbed in world knowledge map, or profession
Domain knowledge map, and above-mentioned knowledge mapping is as an aid in the fields such as search, aid decision, auxiliary question and answer interaction more.Instead
Yan Zhi, for a certain concrete application of completion, specially designed knowledge mapping is fewer and fewer, and therefore knowing based on human-computer interaction
Know map and still lacks progress at present.The basis that human-computer interaction is gone on smoothly is that machine has knowledge base powerful enough, is used for machine
Itself understand the intention of user and organizes corresponding answer.
The existing technical solution for establishing knowledge mapping is mostly by search engine companies (google baidu sougou) or to grind
Study carefully tissue to take the lead exploitation, the knowledge mapping developed has the characteristic sparse towards full field, relationship, while these knowledge mappings
Also has the characteristics that relative quiescent, entity therein, entity relationship, entity attribute may keep steady in longer period of time
It is fixed, and involved knowledge is all objectively, based on a large amount of isomorphism information forms two meta-knoeledge.
User-friendly human-computer interaction needs machine to have subjective perception to the language of user, movement, expression, traditional
In man-machine interactive system, machine can only do feedback once, such as people's key-machine movement to the input of people, due to inabundant
Understand the intention and subjective emotion of user, therefore cannot accomplish customer-centric well, the life to subscriber identity information
There is also defects for object identification, exist simultaneously the limitation of use scope and interactive mode, are not completely free of interactive interface also to easy-to-use
The influence of property.
During human-computer interaction, robot is needed to understand that user is intended to, organizes the answer for meeting user's demand.Mesh
The poor reason of preceding man-machine interaction experience:
1, current robot cannot effectively identify the subjective intention of user, only mechanically identification user input
Content, then mechanically retrieved, the answer organized is also excessively objective.
2, the mostly of machine (information) retrieval are objective contents, be not associated with current subjective emotion, personality, social relationships of user etc. because
Element, that is, do not accomplish centered on user during retrieving answer.
3, for machine when retrieving answer, the content of retrieval is unordered, lower with the correlation degree of user's real demand, has done very
Mostly invalid retrieval, the accuracy that not only inefficiency but also influence are answered.
4, many intention assessment schemes are all based on the Technology designs such as markov random file or bayesian algorithm, identification
Accuracy rate is lower, and priori knowledge is needed to carry out model training, require manual intervention, and efficiency is lower.
5, current techniques are based on, machine can not actively initiate Effective Dialogue, face the test of dialogue " cold start-up ".
Based on the reason that above-mentioned man-machine interaction experience is poor, defect in the prior art is: existing human-computer interaction process
In, knowledge mapping it is single, lead to the true intention that cannot identify user, intelligence can not be provided based on different knowledge mappings
The answer of energy, and robot can not carry out active interlocution, keep interactive experience poor.
Summary of the invention
In view of the above technical problems, the present invention provides man-machine interaction method, the apparatus and system of a kind of knowledge based map,
By designing a plurality of types of knowledge mappings, the experience of user, makes robot in the mistake of human-computer interaction when improving human-computer interaction
The true intention that user is understood in journey accomplishes to accompany the emotion of user, improves user experience.
In order to solve the above technical problems, present invention provide the technical scheme that
In a first aspect, the present invention provides a kind of man-machine interaction method of knowledge based map, comprising:
Step S1, whether detection user comes into the identification range of robot, if known in the identification range of robot
It is clipped to the user, actively wake up the robot and analyzes the identity of the user, obtains the identity of the user；
Step S2 finds the personal knowledge map of the user according to the identity of the user, in the robot
Store the personal knowledge map of multiple users；
Step S3, according to the personal knowledge map of the user, the robot is actively to user initiation pair
Words carry out human-computer interaction；
Step S4 analyzes the intention of the user, and combine the multiple types pre-established in human-computer interaction process
Knowledge mapping interacted with the user, a plurality of types of knowledge mappings be according to the knowledge of different field establish
Knowledge mapping.
A kind of man-machine interaction method of knowledge based map provided by the invention, technical solution are as follows: detecting user is
The no identification range for coming into robot, if recognizing the user in the identification range of robot, actively described in wake-up
Robot and the identity for analyzing the user, obtain the identity of the user；According to the identity of the user, institute is found
The personal knowledge map of user is stated, the personal knowledge map of multiple users is stored in the robot；According to the use
The personal knowledge map of person, the robot actively initiate to talk with to the user, carry out human-computer interaction；In human-computer interaction
Cheng Zhong analyzes the intention of the user, and carries out in conjunction with a plurality of types of knowledge mappings and the user pre-established
Interaction, a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.
The man-machine interaction method of a kind of knowledge based map provided by the invention, by designing a plurality of types of knowledge graphs
Spectrum, the experience of user, makes robot understand the true intention of user during human-computer interaction when improving human-computer interaction,
Accomplish to accompany the emotion of user, improves user experience.
Further, a plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, individual
Knowledge mapping and service and task scene map.
Further, the step S3, specifically:
According to the personal knowledge map of the user, obtained there and then by the service and task scene map
Basic information；
According to the basic information, the robot actively initiates to talk with to the user, carries out human-computer interaction.
Further, in the step S4, the intention of the user is analyzed, specifically:
Word and word analysis are carried out to the interactive information of the user, and the movement and behavior of the user are divided
Analysis, obtains the intention of the user.
Further, the step S4, specifically:
According to the intention of the user, the knowledge mapping of the interactive information between the robot and the user is judged
Classification；
According to the knowledge mapping classification, answer is combined to generate template in a plurality of types of knowledge mappings pre-established
It generates and replys；
According to the reply, interacted with the user.
Further, in the step S2, the individual of the user is found by obtained use relationship registered in advance
Knowledge mapping.
Second aspect, the human-computer interaction device of knowledge based map, comprising:
Whether user's identification module comes into the identification range of robot for detecting user, if in robot
The user is recognized in identification range, actively wake up the robot and analyzes the identity of the user, is obtained described
The identity of user；
Personal knowledge map obtains module, and for the identity according to the user, the individual for finding the user knows
Know map, the personal knowledge map of multiple users is stored in the robot；
Talk with initiation module, for the personal knowledge map according to the user, the robot actively makes to described
User initiates dialogue, carries out human-computer interaction；
Interactive module, in human-computer interaction process, analyzing the intention of the user, and it is more in conjunction with pre-establishing
The knowledge mapping of seed type is interacted with the user, and a plurality of types of knowledge mappings are knowing according to different field
Know the knowledge mapping established.
The human-computer interaction device of knowledge based map provided by the invention, technical solution are as follows: mould is identified by user
Block, whether detection user comes into the identification range of robot, if recognizing the use in the identification range of robot
Person actively wakes up the robot and analyzes the identity of the user, obtains the identity of the user；
Module is obtained by personal knowledge map, according to the identity of the user, the individual for finding the user knows
Know map, the personal knowledge map of multiple users is stored in the robot；By talking with initiation module, according to the use
The personal knowledge map of person, the robot actively initiate to talk with to the user, carry out human-computer interaction；Pass through interaction mould
Block analyzes the intention of the user in human-computer interaction process, and combine a plurality of types of knowledge mappings for pre-establishing with
The user interacts, and a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.
The human-computer interaction device of knowledge based map provided by the invention is changed by designing a plurality of types of knowledge mappings
Into the experience of user when human-computer interaction, so that robot is understood the true intention of user during human-computer interaction, accomplish
The emotion of user is accompanied, user experience is improved.
Further, a plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, individual
Knowledge mapping and service and task scene map.
Further, the dialogue initiation module, is specifically used for:
According to the personal knowledge map of the user, obtained there and then by the service and task scene map
Basic information；
According to the basic information, the robot actively initiates to talk with to the user, carries out human-computer interaction.
The third aspect, the present invention provide a kind of man-machine interactive system of knowledge based map, including robot, setting in institute
State camera, processor and the memory in robot；
The processor is connect with the camera and the memory respectively；
The image of camera user for identification carries out the intention assessment of identification and the user；
The memory is for storing program；
The processor executes following steps for running program described in the memory: whether detection user walks
Into the identification range of robot, if recognizing the user in the identification range of robot, analyze the user's
Identity obtains the identity of the user；
According to the identity of the user, the personal knowledge map of the user is found, storage is more in the robot
The personal knowledge map of a user；According to the personal knowledge map of the user, the robot is actively to the use
Person initiates dialogue, carries out human-computer interaction；In human-computer interaction process, the intention of the user is analyzed, and combine and pre-establish
A plurality of types of knowledge mappings interacted with the user, a plurality of types of knowledge mappings be according to different field
Knowledge establish knowledge mapping.
The man-machine interactive system of knowledge based map provided by the invention is changed by designing a plurality of types of knowledge mappings
Into the experience of user when human-computer interaction, so that robot is understood the true intention of user during human-computer interaction, accomplish
The emotion of user is accompanied, user experience is improved.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described.
Fig. 1 shows a kind of flow chart of the man-machine interaction method of knowledge based map provided by the embodiment of the present invention；
Fig. 2 shows a kind of schematic diagrames of the human-computer interaction device of knowledge based map provided by the embodiment of the present invention.
Specific embodiment
It is described in detail below in conjunction with embodiment of the attached drawing to technical solution of the present invention.Following embodiment is only used for
Clearly illustrate technical solution of the present invention, therefore be intended only as example, and cannot be used as a limitation and limit protection of the invention
Range.
Embodiment one
In a first aspect, Fig. 1 shows a kind of man-machine interaction method of knowledge based map provided by the embodiment of the present invention
Flow chart；As shown in Figure 1, a kind of man-machine interaction method for knowledge based map that embodiment one provides, comprising:
Step S1, whether detection user comes into the identification range of robot, if known in the identification range of robot
It is clipped to user, actively wake up robot and analyzes the identity of user, obtains the identity of user；
Step S2 finds the personal knowledge map of user, stores multiple uses in robot according to the identity of user
The personal knowledge map of person；
Step S3, according to the personal knowledge map of user, robot actively initiates to talk with to user, carries out man-machine friendship
Mutually；
Step S4 analyzes the intention of user in human-computer interaction process, and pre-establish a plurality of types of is combined to know
Know map to interact with user, a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.
A kind of man-machine interaction method of knowledge based map provided by the invention, technical solution are as follows: detecting user is
The no identification range for coming into robot actively wakes up robot simultaneously if recognizing user in the identification range of robot
Analyze the identity of user；According to the identity of user, the personal knowledge map of user is found, stores multiple make in robot
The personal knowledge map of user；According to the personal knowledge map of user, robot actively initiates to talk with to user, carries out people
Machine interaction；In human-computer interaction process, analyze the intention of user, and combine a plurality of types of knowledge mappings for pre-establishing with
User interacts, and a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.
The man-machine interaction method of a kind of knowledge based map provided by the invention, by designing a plurality of types of knowledge graphs
Spectrum, the experience of user, makes robot understand the true intention of user during human-computer interaction when improving human-computer interaction,
Accomplish to accompany the emotion of user, improves user experience.The not only content that mechanically identification user inputs, provides reply,
But the factors such as current subjective emotion, personality, social relationships of combined use person, it is targetedly replied, is improved back
Multiple efficiency and accuracy.
Particularly, the content that human-computer interaction is related to is many and diverse, in addition to needing to consider the current subjective desire of user and life
It arranges, it is also necessary in view of human-computer interaction may relate to the content of the different aspects such as common sense, scientific knowledge, amusement, social activity.It is single
Knowledge mapping it is too huge sparse, be unable to ensure machine in human-computer interaction to the abundant of user's subjective emotion and intention
Identification is also unable to ensure the complete answer for being effectively drawn into specific topics, therefore the present invention is by establishing various types
Knowledge mapping, the complete stock of knowledge library of human-computer interaction framework is carried out for machine, so that machine is in the mistake for carrying out human-computer interaction
Cheng Zhongke understands comprehensive user's intention by these knowledge mappings and extracts correct answer from knowledge mapping.
Preferably, a plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, personal knowledge figure
Spectrum and service and task scene map.
Wherein, general fact knowledge map contains the various rules of nature and the fact has occurred；General knowledge map
Social common sense comprising various aspects；Service and task scene map link the third party service provider and provide such as traffic for user
The third party's services such as situation, weather condition；Personal knowledge map is divided into two kinds, one is setting comprising producers such as machine IDs for machine
Machine itself knowledge mapping of fixed other attributes, the second is the personal knowledge map centered on user.
In use, only personal knowledge map is unified initialization, for the machine/robot just started,
Its personal knowledge map be it is identical, during by use, personal knowledge map just gradually appears the differentiation of content, than
After the identity for such as identifying user, corresponding personal knowledge figure can be found out in personal knowledge map according to the identity of user
Spectrum, the personal knowledge map based on the user interacts later, with the interaction habits of personalized adaptation different users.
Particularly, present invention provide that each machine has and only one machine knowledge mapping itself, and centered on user
Every machine of knowledge mapping can have it is multiple.User's picture can be effectively performed in knowledge mapping centered on user
Picture, for machine, the information such as combining environmental, the tone, context understand user's subjectivity emotion in interactive process and intention has established base
Plinth.
It should be noted that the method for establishing knowledge mapping in the present invention is the prior art.
Preferably, in step S2, the personal knowledge map of user is found by obtained use relationship registered in advance.
The present invention is by obtained use relationship registered in advance, by the knowledge mapping of machine itself and user's knowledge mapping
It is chained up, after machine identifies user in human-computer interaction process, can be closed by obtained use registered in advance
System finds itself portrait to the user, the identification being intended to this guidance machine to user.Specifically, the individual of user knows
Knowing map can be established when registered in advance, and the knowledge mapping of user and robot is connected.
Wherein, user's personal knowledge map includes personal essential information, the letter such as age, height, residence, hobby
Breath.The reply for providing robot is that the factors such as personality and social relationships based on user provide, and it is quasi- to reply more intelligence
Really.
It wherein, include the self information of robot, including achievable function in the personal knowledge profile information of robot
Information, status information etc..Such as robot is in sleep state at this time, when user at this time with robot interactive when, possible machine
The reply of people especially actively, or can directly will not inform user, at this time its sleeping time, it is undesirable to be disturbed, increase machine
The emotion communication of device people and user, the personification of hoisting machine people.
Preferably, by registered in advance, the personal knowledge map based on user carries out the differentiation of permission, and each user establishes
Personal knowledge map, the tender subject in each user knowledge map, can only be accessed in the interactive process with specific user,
Without being known by other users, guarantee that each user has the right for protecting oneself privacy, such as when user A and robot are carried out
When interaction, the privacy concern (such as marital status, property status etc.) about user B is mentioned, robot will recognise that user A not
It is user B, therefore for the privacy concern about user B that user A is proposed, will not gives and answer.Or work as user A and machine
In people's interactive process, when referring to related privacy concern, it is on the scene to recognize other people, robot can remind user A, avoid privacy
It leaks, has further ensured the personal secrets of user, while improving user experience.
It is highly preferred that step S3, specifically:
According to the personal knowledge map of user, believed by the basis serviced and task scene map obtains there and then
Breath；
According to basic information, robot actively initiates to talk with to user, carries out human-computer interaction.
According to service and task scene map, robot can be interacted actively with user, according in basic information
The information such as weather, red-letter day, traffic conditions actively greet user, rather than user is passively waited to propose that topic just carries out
Interaction avoids the phenomenon that being cold-started, and improves user experience.
In addition, in the case where there is awkward silence at a meeting, may also be combined with specific interaction scenarios in human-computer interaction process, in conjunction with
At least one of general fact knowledge map, general knowledge map, personal knowledge map and service and task scene map figure
Spectrum, actively proposes topic, interacts with user.The unasked topic of robot includes if user is interested recently
Topic, can extract from dialog history information and obtain, and the topic of proposition may also comprise the topic of nearest hot topic, can talk about from network hot topic
The personal preference of combined use person obtains in topic.By actively proposing topic in human-computer interaction process, awkward silence at a meeting situation is reduced
Occur, improves interactive experience.
Preferably, in step S4, the intention of user is analyzed, specifically:
Word and word analysis are carried out to the interactive information of user, and the movement and behavior of user are analyzed, is obtained
The intention of user.
Specifically, in human-computer interaction process, monitoring user that can be real-time continuous by the camera of robot itself
Behavior and movement, then analyzed, combine the information such as specific environment, the tone, context to understand in interactive process and use
Person's subjectivity emotion and intention only comprehensively understand the intention of user, can just provide more intelligent reply.
Wherein, word is carried out to the interactive information of user and word is analyzed, it can be according to word of the words in interactive information sentence
Property, syntax rule, the tone, intonation etc. carry out the identification of sentence intention, carry out user from the interactive information that user inputs
The analysis of intention.
Wherein, the image of the user acquired by camera carries out image recognition, distinguishes the expression of user, acts
Deng, by analyze user behavior and movement, further appreciate that the true intention of user.
It is highly preferred that step S4, specifically:
According to the intention of user, the knowledge mapping classification of the interactive information between robot and user is judged；
According to knowledge mapping classification, answer is combined to generate template generation in a plurality of types of knowledge mappings pre-established
It replys；
According to reply, interacted with user.
Wherein, the generating mode of reply is not limited in the present invention, the generating mode of reply can be realized using the prior art,
Such as the keyword inputted according to user, entity and entity relationship are extracted, correspondence finds knowledge mapping, generates mould according to answer
Plate generates candidate answers in conjunction with knowledge mapping.Or the information inputted according to user, the letter of user input is judged according to intonation
Breath is instruction or dialogue, and if it is dialogue, the retrieval just replied according to existing question and answer mode provides reply.
Specifically, it after the true intention for identifying user, targetedly to be returned according to different knowledge mappings
It is multiple, improve the accuracy of reply.
Specifically, the image that user can be acquired by camera, identifies the identity of user, or pass through voice collecting
Equipment acquires the identity of the voice recognition user of user.
Be exemplified below in the present embodiment knowledge mapping be human-computer interaction process bring the utility model has the advantages that
1, under machine standby mode, when user comes into the sensing range of machine, the infrared module of machine is perceived
User walks close to, and obtains the information such as weather, red-letter day, traffic at that time by service map and task scene map, machine is actively
It initiates to talk with to user.
2, when user comes into the identification range of machine, the camera of machine acquires user's image information, and passes through
The identity of the biological informations such as image, voice identification user.
3, after determining user's identity, machine based on personal knowledge map determine user social relationships and machine with make
The relationship of user.
4, machine continues movement, the behavior of tracking observation user, and movement, behavior based on user automatically initiate pair
Words, and question and answer are carried out according to the four kinds of knowledge mappings and user that pre-establish.
Referring to fig. 2, second aspect, the human-computer interaction device 10 of knowledge based map, comprising:
Whether user's identification module 101 comes into the identification range of robot for detecting user, if in robot
Identification range in recognize user, actively wake up robot and analyze the identity of user, obtain the identity of user；
Personal knowledge map obtains module 102 and finds the personal knowledge figure of user for the identity according to user
It composes, the personal knowledge map of multiple users is stored in robot；
Talk with initiation module 103, for the personal knowledge map according to user, robot is actively to user's initiation pair
Words carry out human-computer interaction；
Interactive module 104, in human-computer interaction process, analyzing the intention of user, and it is more in conjunction with pre-establishing
The knowledge mapping of seed type is interacted with user, and a plurality of types of knowledge mappings are to be established according to the knowledge of different field
Knowledge mapping.
The human-computer interaction device 10 of knowledge based map provided by the invention, technical solution are as follows: identified by user
Module 101, whether detection user comes into the identification range of robot, if recognizing use in the identification range of robot
Person actively wakes up robot and analyzes the identity of user, obtains the identity of user；
Module 102 is obtained by personal knowledge map, and the personal knowledge figure of user is found according to the identity of user
It composes, the personal knowledge map of multiple users is stored in robot；By talking with initiation module 103, according to the individual of user
Knowledge mapping, robot actively initiate to talk with to user, carry out human-computer interaction；By interactive module 104, in human-computer interaction
Cheng Zhong analyzes the intention of user, and a plurality of types of knowledge mappings pre-established is combined to interact with user, a variety of
The knowledge mapping of type is the knowledge mapping established according to the knowledge of different field.
The human-computer interaction device 10 of knowledge based map provided by the invention, by designing a plurality of types of knowledge mappings,
The experience of user, makes robot understand the true intention of user during human-computer interaction, does when improving human-computer interaction
It is accompanied to the emotion to user, improves user experience.
Preferably, a plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, personal knowledge figure
Spectrum and service and task scene map.
Preferably, talk with initiation module, be specifically used for:
According to the personal knowledge map of user, believed by the basis serviced and task scene map obtains there and then
Breath；
According to basic information, robot actively initiates to talk with to user, carries out human-computer interaction.
The third aspect, the present invention provide a kind of man-machine interactive system of knowledge based map, including robot, setting in machine
Camera, processor and memory on device people；
Processor is connect with camera and memory respectively；
The image of camera user for identification carries out the intention assessment of identification and user；
Memory is for storing program；
Processor executes following steps for program in run memory: whether detection user comes into the knowledge of robot
Other range analyzes the identity of user if recognizing user in the identification range of robot；
According to the identity of user, the personal knowledge map of user is found, of multiple users is stored in robot
People's knowledge mapping；According to the personal knowledge map of user, robot actively initiates to talk with to user, carries out human-computer interaction；
In human-computer interaction process, the intention of user is analyzed, and combine a plurality of types of knowledge mappings pre-established and user
It interacts, a plurality of types of knowledge mappings are the knowledge mapping established according to the knowledge of different field.
The man-machine interactive system of knowledge based map provided by the invention is changed by designing a plurality of types of knowledge mappings
Into the experience of user when human-computer interaction, so that robot is understood the true intention of user during human-computer interaction, accomplish
The emotion of user is accompanied, user experience is improved.
Embodiment two
Based on general fact knowledge map, general knowledge map, personal knowledge map and service and task scene map, knot
Concrete scene is closed, is further detailed:
Under machine standby mode, when user comes into the sensing range of machine, by the infrared module of machine, such as
Infrared sensor perceives user and walks close to, and is believed by weather, red-letter day, the traffic etc. that service and task scene map obtains at that time
Breath, machine actively initiate to talk with to user.Such as: " you are good, for a moment can be rainy, you, which go out, gets out umbrella? " pass through master
It is dynamic to initiate dialogue, some daily weather of user can be not only reminded, traffic information etc. can also further draw in user
At a distance from robot.
If user is interested in the topic of robot or has other to need to seek advice from robot, further with machine
Device people engages in the dialogue, and user will come into robot, and when user comes into the identification range of machine, the camera of machine is adopted
Collect user's image information, and identifies the identity of user by biological informations such as image, voices.After determining user's identity,
Machine determines the social relationships of user and the relationship of robot and user based on personal knowledge map.In user and use
During person engages in the dialogue, the intention of user, the personal knowledge map of combined use person and specifically used person couple can be analyzed
The field scope that words content is related to, provides answer.
For example, user can then ask that " road conditions near XX are present according to the unasked topic of a upper robot
How? " robot can transfer rapidly current traffic information, and combined use person based on service and task scene map
The route usually walked in people's knowledge mapping, is answered: " road conditions are good at present, you can trust trip."
If user wants to go to a unfamiliar place self-driving travel, this place is not quite understood, machine can also be seeked advice from
People, such as geographical common sense, " which Daocheng save at? " robot can provide answer " in Sichuan Province " according to general knowledge map；It connects
Get off, user may specifically seek advice from local conditions and customs local, and the individual of clothing, food, lodging and transportion -- basic necessities of life, robot combined use person knows
Know map and fact knowledge map, general knowledge map；For example, showing to like quietly to occupy in the personal knowledge map of user
Firmly environment is liked eating distinctive snack, and robot will recommend some peaces of user in conjunction with local political geography characteristic
Quiet people place, not only beautiful environment, and there are also local characteristics, if by chance catching up with local red-letter day, can also be practised according to local custom
It is used to give user some good suggestions.For example based on the fact that knowledge mapping introduces the history of some Daochengs；Based on general knowledge figure
Compose some points for attention of illustratively self-driving travel.
If at this time after some introductions and recommendation of user Ting Guo robot, hence it is evident that show very interested and very high
Emerging state, robot can also further do interaction by analyzing the dialogue intention of user with user, then ask user
Some interested topics, such as: " do you like encamping? in XX, this season has many people in that camping, the starry sky in evening
It is as fair as a lily ", of this sort topic can establish good interactive environment with user, according to the specifically intended of user, in conjunction with
Four kinds of knowledge mappings provide accurate quickly reply abundant, promote the interactive experience of user, allow user experience seem
As good friend is talked.
Based on the interactive mode of four kinds of knowledge mappings, compare the interactive mode based on a knowledge mapping, knows in conjunction with four kinds
The interactive mode for knowing map can fully and effectively identify the subjective intention of user, rather than mechanically identification user input
Content provide mechanical answer in conjunction with fixed dialog template, which is not only not bound in a knowledge mapping
The actual conditions of user are not associated with user current subjective emotion, personality, social relationships etc. without passionate color
Factor, that is, do not accomplish centered on user during answer retrieving usually because of the answering model of dialog template
Limitation, bad experience is provided to user.
It should be noted that not limiting in the present invention is four kinds of knowledge mappings, the interactive mode based on a variety of knowledge mappings
It is within the scope of the invention, such as weather map and music map, is substantially third party's service and task field
Scape map.
In addition, machine, when retrieving answer, the content of retrieval is unordered, lower with the correlation degree of user's real demand, do
Many invalid retrievals not only inefficiency but also influence the accuracy answered.Based on current techniques, machine can not actively be sent out
Effective Dialogue is played, the test of dialogue " cold start-up " is faced, that is usually only user carries out actively starting dialogue mode, machine
Device talent conference provides reply based on the conversation content of user, and the robot in the present invention can actively wake up, and then leads
It is dynamic to engage in the dialogue with user, further improve the user experience of user.
Particularly, the content that human-computer interaction is related to is many and diverse, in addition to needing to consider the current subjective desire of user and life
It arranges, it is also necessary in view of human-computer interaction may relate to the content of the different aspects such as common sense, scientific knowledge, amusement, social activity.It is single
Knowledge mapping it is too huge sparse, be unable to ensure machine in human-computer interaction to the abundant of user's subjective emotion and intention
Identification is also unable to ensure the complete answer for being effectively drawn into specific topics, therefore the present invention is by establishing various types
Knowledge mapping, the complete stock of knowledge library of human-computer interaction framework is carried out for machine, so that machine is in the mistake for carrying out human-computer interaction
Cheng Zhongke understands comprehensive user's intention by these knowledge mappings and extracts correct answer from knowledge mapping.
In human-computer interaction process, robot can continue the movement of tracking observation user, behavior, analyze the meaning of user
Figure, provides accurate answer in conjunction with four kinds of knowledge mappings in the present invention.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover within the scope of the claims and the description of the invention.
Claims (10)
1. the man-machine interaction method of knowledge based map characterized by comprising
Step S1, whether detection user comes into the identification range of robot, if recognized in the identification range of robot
The user actively wakes up the robot and analyzes the identity of the user, obtains the identity of the user；
Step S2 finds the personal knowledge map of the user, stores in the robot according to the identity of the user
The personal knowledge map of multiple users；
Step S3, according to the personal knowledge map of the user, the robot actively initiates to talk with to the user, into
Row human-computer interaction；
Step S4 analyzes the intention of the user in human-computer interaction process, and pre-establish a plurality of types of is combined to know
Know map to interact with the user, a plurality of types of knowledge mappings are to be known according to what the knowledge of different field was established
Know map.
2. the man-machine interaction method of knowledge based map according to claim 1, which is characterized in that
A plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, personal knowledge map kimonos
Business and task scene map.
3. the man-machine interaction method of knowledge based map according to claim 2, which is characterized in that
The step S3, specifically:
According to the personal knowledge map of the user, basis there and then is obtained by the service and task scene map
Information；
According to the basic information, the robot actively initiates to talk with to the user, carries out human-computer interaction.
4. the man-machine interaction method of knowledge based map according to claim 1, which is characterized in that
In the step S4, the intention of the user is analyzed, specifically:
Word and word analysis are carried out to the interactive information of the user, and the movement and behavior of the user are analyzed,
Obtain the intention of the user.
5. the man-machine interaction method of knowledge based map according to claim 4, which is characterized in that
The step S4, specifically:
According to the intention of the user, the knowledge mapping class of the interactive information between the robot and the user is judged
Not；
According to the knowledge mapping classification, answer is combined to generate template generation in a plurality of types of knowledge mappings pre-established
It replys；
According to the reply, interacted with the user.
6. the man-machine interaction method of knowledge based map according to claim 1, which is characterized in that
In the step S2, the personal knowledge map of the user is found by obtained use relationship registered in advance.
7. the human-computer interaction device of knowledge based map characterized by comprising
Whether user's identification module comes into the identification range of robot for detecting user, if in the identification of robot
The user is recognized in range, actively wake up the robot and analyzes the identity of the user, obtains the use
The identity of person；
Personal knowledge map obtains module and finds the personal knowledge figure of the user for the identity according to the user
It composes, the personal knowledge map of multiple users is stored in the robot；
Talk with initiation module, for the personal knowledge map according to the user, the robot is actively to the user
Dialogue is initiated, human-computer interaction is carried out；
Interactive module in human-computer interaction process, analyzing the intention of the user, and combines the multiple types pre-established
The knowledge mapping of type is interacted with the user, and a plurality of types of knowledge mappings are to be built according to the knowledge of different field
Vertical knowledge mapping.
8. the human-computer interaction device of knowledge based map according to claim 7, which is characterized in that
A plurality of types of knowledge mappings include general fact knowledge map, general knowledge map, personal knowledge map kimonos
Business and task scene map.
9. the human-computer interaction device of knowledge based map according to claim 8, which is characterized in that
The dialogue initiation module, is specifically used for:
According to the personal knowledge map of the user, basis there and then is obtained by the service and task scene map
Information；
According to the basic information, the robot actively initiates to talk with to the user, carries out human-computer interaction.
10. the man-machine interactive system of knowledge based map, which is characterized in that including robot, be arranged in the robot
Camera, processor and memory；
The processor is connect with the camera and the memory respectively；
The image of camera user for identification carries out the intention assessment of identification and the user；
The memory is for storing program；
The processor executes following steps for running program described in the memory: whether detection user comes into machine
The identification range of device people analyzes the identity of the user if recognizing the user in the identification range of robot,
Obtain the identity of the user；
According to the identity of the user, the personal knowledge map of the user is found, stores multiple make in the robot
The personal knowledge map of user；According to the personal knowledge map of the user, the robot is actively sent out to the user
Dialogue is played, human-computer interaction is carried out；In human-computer interaction process, the intention of the user is analyzed, and more in conjunction with pre-establishing
The knowledge mapping of seed type is interacted with the user, and a plurality of types of knowledge mappings are knowing according to different field
Know the knowledge mapping established.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710595078.6A CN110019824A (en) | 2017-07-20 | 2017-07-20 | Man-machine interaction method, the apparatus and system of knowledge based map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710595078.6A CN110019824A (en) | 2017-07-20 | 2017-07-20 | Man-machine interaction method, the apparatus and system of knowledge based map |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110019824A true CN110019824A (en) | 2019-07-16 |
Family
ID=67185862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710595078.6A Pending CN110019824A (en) | 2017-07-20 | 2017-07-20 | Man-machine interaction method, the apparatus and system of knowledge based map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110019824A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909159A (en) * | 2019-09-29 | 2020-03-24 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN111858236A (en) * | 2020-06-23 | 2020-10-30 | 深圳精匠云创科技有限公司 | Knowledge graph monitoring method and device, computer equipment and storage medium |
| CN112015919A (en) * | 2020-09-15 | 2020-12-01 | 重庆广播电视大学重庆工商职业学院 | Dialogue management method based on learning auxiliary knowledge graph |
| CN112331201A (en) * | 2020-11-03 | 2021-02-05 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and electronic device |
| CN113377947A (en) * | 2021-07-06 | 2021-09-10 | 北京明略昭辉科技有限公司 | Theme recommendation method and device, computer equipment and storage medium |
| CN113408690A (en) * | 2021-07-01 | 2021-09-17 | 之江实验室 | Robot personalized emotion interaction device and method based on multi-mode knowledge graph |
| CN113486233A (en) * | 2020-08-21 | 2021-10-08 | 青岛海信电子产业控股股份有限公司 | Content recommendation method, device and medium |
| CN114363277A (en) * | 2020-12-31 | 2022-04-15 | 万翼科技有限公司 | Intelligent chatting method and device based on social relationship and related products |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710543A (en) * | 2008-12-31 | 2012-10-03 | 北京华旗资讯数码科技有限公司 | Interaction instant communication device including three-dimensional device and system thereof |
| CN203399142U (en) * | 2013-02-27 | 2014-01-15 | 深圳好未来智能科技有限公司 | Working and learning application robot |
| CN104010014A (en) * | 2013-02-27 | 2014-08-27 | 深圳好未来智能科技有限公司 | Work learning application robot |
| CN204463028U (en) * | 2015-01-16 | 2015-07-08 | 深圳市中控生物识别技术有限公司 | A kind of identification apparatus with infrared arousal function |
| US20150370787A1 (en) * | 2014-06-18 | 2015-12-24 | Microsoft Corporation | Session Context Modeling For Conversational Understanding Systems |
| CN105856257A (en) * | 2016-06-08 | 2016-08-17 | 以恒激光科技（北京）有限公司 | Intelligent robot suitable for reception |
| CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
| CN106126503A (en) * | 2016-07-12 | 2016-11-16 | 海信集团有限公司 | Business scope localization method and terminal |
| CN106326440A (en) * | 2016-08-26 | 2017-01-11 | 北京光年无限科技有限公司 | Human-computer interaction method and device facing intelligent robot |
| CN106462384A (en) * | 2016-06-29 | 2017-02-22 | 深圳狗尾草智能科技有限公司 | Multi-modal based intelligent robot interaction method and intelligent robot |
| CN106874261A (en) * | 2017-03-17 | 2017-06-20 | 中国科学院软件研究所 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
-
2017
- 2017-07-20 CN CN201710595078.6A patent/CN110019824A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710543A (en) * | 2008-12-31 | 2012-10-03 | 北京华旗资讯数码科技有限公司 | Interaction instant communication device including three-dimensional device and system thereof |
| CN203399142U (en) * | 2013-02-27 | 2014-01-15 | 深圳好未来智能科技有限公司 | Working and learning application robot |
| CN104010014A (en) * | 2013-02-27 | 2014-08-27 | 深圳好未来智能科技有限公司 | Work learning application robot |
| US20150370787A1 (en) * | 2014-06-18 | 2015-12-24 | Microsoft Corporation | Session Context Modeling For Conversational Understanding Systems |
| CN204463028U (en) * | 2015-01-16 | 2015-07-08 | 深圳市中控生物识别技术有限公司 | A kind of identification apparatus with infrared arousal function |
| CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
| CN105856257A (en) * | 2016-06-08 | 2016-08-17 | 以恒激光科技（北京）有限公司 | Intelligent robot suitable for reception |
| CN106462384A (en) * | 2016-06-29 | 2017-02-22 | 深圳狗尾草智能科技有限公司 | Multi-modal based intelligent robot interaction method and intelligent robot |
| CN106126503A (en) * | 2016-07-12 | 2016-11-16 | 海信集团有限公司 | Business scope localization method and terminal |
| CN106326440A (en) * | 2016-08-26 | 2017-01-11 | 北京光年无限科技有限公司 | Human-computer interaction method and device facing intelligent robot |
| CN106874261A (en) * | 2017-03-17 | 2017-06-20 | 中国科学院软件研究所 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909159A (en) * | 2019-09-29 | 2020-03-24 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN110909159B (en) * | 2019-09-29 | 2022-09-16 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN111858236A (en) * | 2020-06-23 | 2020-10-30 | 深圳精匠云创科技有限公司 | Knowledge graph monitoring method and device, computer equipment and storage medium |
| CN111858236B (en) * | 2020-06-23 | 2022-12-16 | 深圳精匠云创科技有限公司 | Knowledge graph monitoring method and device, computer equipment and storage medium |
| CN113486233A (en) * | 2020-08-21 | 2021-10-08 | 青岛海信电子产业控股股份有限公司 | Content recommendation method, device and medium |
| CN112015919A (en) * | 2020-09-15 | 2020-12-01 | 重庆广播电视大学重庆工商职业学院 | Dialogue management method based on learning auxiliary knowledge graph |
| CN112331201A (en) * | 2020-11-03 | 2021-02-05 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and electronic device |
| CN114363277A (en) * | 2020-12-31 | 2022-04-15 | 万翼科技有限公司 | Intelligent chatting method and device based on social relationship and related products |
| CN113408690A (en) * | 2021-07-01 | 2021-09-17 | 之江实验室 | Robot personalized emotion interaction device and method based on multi-mode knowledge graph |
| CN113377947A (en) * | 2021-07-06 | 2021-09-10 | 北京明略昭辉科技有限公司 | Theme recommendation method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110019824A (en) | Man-machine interaction method, the apparatus and system of knowledge based map | |
| US11526720B2 (en) | Artificial intelligence system for supporting communication | |
| CN106095833B (en) | Human-computer dialogue content processing method | |
| KR100688087B1 (en) | method for modeling context information using rules and ontology and context aware system | |
| US20170186429A1 (en) | Better resolution when referencing to concepts | |
| US8838505B2 (en) | Schedule management system using interactive robot and method and computer-readable medium thereof | |
| CN110288985A (en) | Voice data processing method, device, electronic equipment and storage medium | |
| CN108804698A (en) | Man-machine interaction method, system, medium based on personage IP and equipment | |
| CN107493353A (en) | A kind of intelligent robot cloud computing method based on contextual information | |
| CN105912692B (en) | A kind of method and apparatus of Intelligent voice dialog | |
| WO2018098932A1 (en) | Proactive searching method and device based on speech recognition | |
| CN106294774A (en) | User individual data processing method based on dialogue service and device | |
| WO2016104736A1 (en) | Communication provision system and communication provision method | |
| US20230014775A1 (en) | Intelligent task completion detection at a computing device | |
| US11521111B2 (en) | Device and method for recommending contact information | |
| US10770072B2 (en) | Cognitive triggering of human interaction strategies to facilitate collaboration, productivity, and learning | |
| US11694690B2 (en) | Method and system providing service based on user voice | |
| CN108595609A (en) | Generation method, system, medium and equipment are replied by robot based on personage IP | |
| CN109119095A (en) | Level of fatigue recognition methods, device, computer equipment and storage medium | |
| CN106815321A (en) | Chat method and device based on intelligent chat robots | |
| CN113746978A (en) | Device and method for recommending contact information | |
| KR20200092166A (en) | Server, method and computer program for recognizing emotion | |
| KR20180116104A (en) | Method and apparatus of dialog scenario database constructing for dialog system | |
| KR20190046062A (en) | Method and apparatus of dialog scenario database constructing for dialog system | |
| KR20180009408A (en) | Placeness Extraction and Inference System based on Social Network Analysis and Ontology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190716 |