CN109086391A - A kind of method and system constructing knowledge mapping - Google Patents
A kind of method and system constructing knowledge mapping Download PDFInfo
- Publication number
- CN109086391A CN109086391A CN201810842317.8A CN201810842317A CN109086391A CN 109086391 A CN109086391 A CN 109086391A CN 201810842317 A CN201810842317 A CN 201810842317A CN 109086391 A CN109086391 A CN 109086391A
- Authority
- CN
- China
- Prior art keywords
- data
- knowledge mapping
- knowledge
- user
- vertical field
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of methods and system for constructing knowledge mapping.The described method includes: obtaining semi-structured data；For target domain, information extraction is carried out from webpage, obtains the vertical FIELD Data comprising entity node and/or node relationships；The vertical FIELD Data is added in the semi-structured data, the knowledge mapping comprising vertical FIELD Data is generated.Compared to the prior art, according to the method for the present invention and system, generation that can be simple and quick is directed to the knowledge mapping in specific vertical field, so that the human-computer interaction for intelligent robot is provided with force data support.
Description
Technical field
The present invention relates to computer fields, and in particular to a kind of method and system for constructing knowledge mapping.
Background technique
With the continuous development of artificial intelligence technology, intelligent robot is increasingly used in the production and living of the mankind
In the middle.
In the interactive process of intelligent robot and the mankind, it is accurate to respond to improve intelligent robot to user demand
Degree, improves the user experience of intelligent robot, and during generating interaction response, intelligent robot would generally call relevant
Knowledge mapping generates interactive response according to the knowledge information data in knowledge mapping.For example, puing question in user, " potato is cooked
When method ", intelligent robot can call relevant knowledge map, therefrom obtain the menu in relation to " potato ".In above process, knowledge
The correctness of knowledge information data and integrality will have a direct impact on the correctness of intelligent robot response in map.
Currently, the building of knowledge mapping is all based on the good structural data of manual compiling.However, manual compiling is not only too
The time-consuming and laborious and easy knowledge point omitted compared with side door.This results in the knowledge mapping that can not only come into operation at this stage suitable
Scarcity, even and the knowledge mapping that comes into operation spent comprehensively there is also correctness and knowledge and be unable to asking for meet demand
Topic.
Summary of the invention
The present invention provides a kind of methods for constructing knowledge mapping, which comprises
Obtain semi-structured data；
For target domain, information extraction is carried out from webpage, obtains the vertical neck comprising entity node and/or node relationships
Numeric field data；
The vertical FIELD Data is added in the semi-structured data, the knowledge comprising vertical FIELD Data is generated
Map.
In one embodiment, information extraction is carried out from webpage, comprising:
Text identification identifies web page text；
Data are crawled, data relevant to the target domain are obtained based on text identification result；
Entity extraction and relationship are extracted, and the entity node of the semi-structured data missing is extracted from the data of acquisition
And/or node relationships.
It is in one embodiment, described to crawl data step further include:
Data cleansing, the data cleansing are non-physical cleaning, wherein are filtered for specific vocabulary.
In one embodiment, the entity extraction includes:
Name Entity recognition, wherein be named Entity recognition using the method for deep learning.
In one embodiment, node relationships extraction is carried out based on neural network.
The invention also provides a kind of exchange methods, which comprises
It obtains user's multi-modal data and parses, determine that user's interaction is intended to；
Determine that user's interaction is intended to corresponding vertical field；
Knowledge mapping is called, the knowledge mapping includes the vertical FIELD Data in the corresponding vertical field；
The knowledge information data responded needed for user's interaction is intended to are extracted from the knowledge mapping；
According to the knowledge information data, it is intended to generate multi-modal interactive response data for user interaction.
In one embodiment, the method also includes:
Obtain the feedback data that user is directed to the multi-modal interactive response data；
Determine that the user is directed to the satisfaction of the multi-modal interactive response data according to the feedback data；
According to the satisfaction determine the data correctness of knowledge information data extracted from the knowledge mapping and/or
Data integrity degree；
According to the data correctness and/or the data integrity degree to the knowledge information extracted from the knowledge mapping
Data carry out data update and/or carry out data supplement to the knowledge mapping.
The invention also provides a kind of storage medium, it is stored on the storage medium and method as described herein can be achieved
Program code.
The invention also provides a kind of building system for knowledge mapping, the system comprises:
Foundation structure constructing module is configured to obtain semi-structured data；
Data extraction module is configured to determine target domain, for the target domain, carries out information from webpage and mentions
It takes, obtains the vertical FIELD Data comprising entity node and/or node relationships；
Knowledge mapping generation module is configured to add to the vertical FIELD Data in the semi-structured data,
Generate the knowledge mapping comprising vertical FIELD Data.
The invention also provides a kind of interactive system, the system comprises:
Input obtains module, is configured to acquisition user's multi-modal data；
Interaction parsing module is configured to parse user's multi-modal data, obtains user's interaction and is intended to and determines institute
It states user's interaction and is intended to corresponding vertical field；
Knowledge mapping calling module is configured to the knowledge mapping for calling building system as described herein to generate, described
Knowledge mapping includes the vertical FIELD Data in the corresponding vertical field；
Generation module is responded in interaction, is configured to needed for being intended to from the user's interaction of extraction response in the knowledge mapping
Knowledge information data be intended to generate multi-modal interaction for user interaction and respond number according to the knowledge information data
According to.
Compared to the prior art, according to the method for the present invention and system, generation that can be simple and quick is for specific vertical
The knowledge mapping in field, so that the human-computer interaction for intelligent robot is provided with force data support.
Other feature or advantage of the invention will illustrate in the following description.Also, Partial Feature of the invention or
Advantage will be become apparent by specification, or be appreciated that by implementing the present invention.The purpose of the present invention and part
Advantage can be realized or be obtained by step specifically noted in the specification, claims and drawings.
Detailed description of the invention
Attached drawing is used to provide further understanding of the present invention, and constitutes part of specification, with reality of the invention
It applies example and is used together to explain the present invention, be not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the method flow diagram of building knowledge mapping according to an embodiment of the invention；
Fig. 2 is the method partial process view of building knowledge mapping according to an embodiment of the present invention；
Fig. 3 is exchange method flow chart according to an embodiment of the invention；
Fig. 4 is exchange method partial process view according to an embodiment of the invention；
Fig. 5 is the system structure schematic diagram of building knowledge mapping according to an embodiment of the invention；
Fig. 6 is the components of system as directed structure diagram of building knowledge mapping according to an embodiment of the invention；
Fig. 7 is interactive system structure diagram according to an embodiment of the invention.
Specific embodiment
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and examples, implementation personnel of the invention whereby
Can fully understand that how the invention applies technical means to solve technical problems, and reach technical effect realization process and according to
The present invention is embodied according to above-mentioned realization process.As long as each embodiment it should be noted that do not constitute conflict, in the present invention
And each feature in each embodiment can be combined with each other, be formed by technical solution protection scope of the present invention it
It is interior.
With the continuous development of artificial intelligence technology, intelligent robot is increasingly used in the production and living of the mankind
In the middle.
In the interactive process of intelligent robot and the mankind, it is accurate to respond to improve intelligent robot to user demand
Degree, improves the user experience of intelligent robot, and during generating interaction response, intelligent robot would generally call relevant
Knowledge mapping generates interactive response according to the knowledge information data in knowledge mapping.For example, puing question in user, " potato is cooked
When method ", intelligent robot can call relevant knowledge map, therefrom obtain the menu in relation to " potato ".In above process, knowledge
The correctness of knowledge information data and integrality will have a direct impact on the correctness of intelligent robot response in map.
Currently, the building of knowledge mapping is all based on the good structural data of manual compiling.However, manual compiling is not only too
The time-consuming and laborious and easy knowledge point omitted compared with side door.This results in the knowledge mapping that can not only come into operation at this stage suitable
Scarcity, even and the knowledge mapping that comes into operation spent comprehensively there is also correctness and knowledge and be unable to asking for meet demand
Topic.
In view of the above-mentioned problems, the invention proposes a kind of methods for constructing knowledge mapping.
In order to guarantee that correctness and knowledge of the knowledge mapping on specific vertical field are spent comprehensively, it is desirable to knowledge mapping
In the knowledge data as much as possible comprising corresponding vertical field.Under current techniques environment, have opposite full knowledge data
One important information source of amount is exactly webpage.Therefore, in the method for the invention, knowledge data is carried out using webpage as data source
Collection, thus obtain be directed to object knowledge field comprehensive knowledge data.
Further, simple vertical domain knowledge data cannot constitute a complete knowledge mapping.In order to the greatest extent
The building process of possible simplified knowledge mapping, in the method for the invention, the data structure conduct that use is existing, more typical
Trunk depends on vertical domain knowledge data on the trunk to simple and fast composition knowledge mapping.
Under current techniques environment, semi-structured data (semi-structured data) is one in Database Systems
Kind data type, has following characteristics:
(1) it is a kind of data model suitable for geo-database integration, that is to say, that is suitable for description included in two or more
Data in database (set of metadata of similar data that these databases contain different mode).
(2) it is a kind of basic model of label service, for shared information on Web.
It is compared with common plain text, semi-structured data has centainly structural, but and with stringent theoretical model
The data of relational database are compared, it has shortage of data again.
Based on the specificity analysis to semi-structured data, in the method for the invention, existing structural data is used
As trunk, knowledge data collection is carried out for object knowledge field, the knowledge data collected is filled into semi-structured number
In, to obtain the knowledge mapping comprising complete vertical FIELD Data.Compared to the prior art, according to the method for the present invention
And system, generation that can be simple and quick are directed to the knowledge mapping in specific vertical field, to be the man-machine friendship of intelligent robot
Mutually it is provided with force data support.
Next based on attached drawing detailed description detailed process according to the method for the embodiment of the present invention, in the flow chart of attached drawing
The step of showing can execute in the computer system comprising such as a group of computer-executable instructions.Although in flow charts
The logical order of each step is shown, but in some cases, it can be to be different from shown by sequence execution herein or retouch
The step of stating.
As shown in Figure 1, in one embodiment, method includes:
It obtains semi-structured data (S110)；
For target domain, information extraction is carried out from webpage, obtains the vertical neck comprising entity node and/or node relationships
Numeric field data (S120)；
The vertical FIELD Data obtained in step S120 is added in the semi-structured data of step S110 acquisition, is generated
Knowledge mapping (S130) comprising vertical FIELD Data.
In method flow of the invention, one of key point is to obtain correct, comprehensive vertical field number from webpage
According to.In one embodiment, it is primarily based on web page text identification, the knowledge information that webpage is included is determined, is then wrapped from webpage
The vertical domain knowledge information in object knowledge field is filtered out in the knowledge information contained.As the webpage as data source is constantly long-pending
Tired, the vertical domain knowledge information got also constantly accumulates.Since the webpage as data source has very big data volume, this
It is ensured that finally getting comprehensive vertical domain knowledge information.
Further, for semi-structured data, the data lacked compared with complete knowledge mapping include entity
Node and/or node relationships.
Therefore, in one embodiment, carrying out the data filling of vertical FIELD Data to semi-structured data is to half hitch
Entity node and/or node relationships are filled in structure data.Specifically, that is, the vertical domain knowledge obtained from webpage is believed
Breath is converted into specific entity node and/or node relationships.
Specifically, as shown in Fig. 2, in one embodiment, carrying out information extraction from webpage, comprising:
Text identification (S210) identifies web page text；
Data (S220) is crawled, the text identification result based on step S210 obtains data relevant to target domain；
Entity extraction and relationship extract (S230), and semi-structured data missing is extracted from the data that step S220 is obtained
Entity node and/or node relationships.
Specifically, the result of extraction can be simple entity node, be also possible to simple node in step S230
Relationship, the set that can also be entity node and node relationships.
Specifically, in one embodiment, the node relationships of semi-structured data missing can be semi-structured data and wrap
Node relationships between the entity node contained, the node relationships being also possible between the entity node of semi-structured data missing, may be used also
To be the node relationships between the entity node of semi-structured data missing and the entity node for having included.
Further, it is contemplated that the information on network is many and diverse, may include redundancy and/or error message.Therefore,
In one embodiment, crawling data step further includes data cleansing step.
Specifically, in one embodiment, the process that data cleansing step is examined and verified again to data, purpose exists
The mistake existing for deletion duplicate message, correction, and data consistency is provided.
Further, it is contemplated that the network information may include sensitive information, in one embodiment, in data cleansing step
Include non-physical cleaning step, wherein be filtered for specific vocabulary.
Further, in one embodiment, during entity extraction, entity extraction includes name Entity recognition step.
Further, in order to guarantee to name the accuracy of Entity recognition, entity node be avoided to be missed, in an embodiment
In, Entity recognition is named using the method for deep learning.Specifically, in one embodiment, using the method for deep learning,
Entity recognition is named using BiLSTM+CRF model.
Further, in order to guarantee to obtain accurate node relationships, node relationships be avoided to be missed, in one embodiment,
Node relationships extraction is carried out based on neural network.
Further, in order to avoid data redundancy, in one embodiment, when carrying out entity extraction and relationship is extracted,
Existing entity node and node relationships in synchronous contrast semi-structured data, so that repeated data be avoided to be filled into half hitch
In structure data.
Further, the method based on building knowledge mapping proposed by the present invention, the invention also provides a kind of interaction sides
Method.Specifically, as shown in figure 3, in one embodiment, method includes:
It obtains user's multi-modal data (S310)；
User's multi-modal data is parsed, determines that user's interaction is intended to (S320)；
Determine that the interaction of user determined by step S320 is intended to corresponding vertical field (S330)；
Knowledge mapping is called, which includes the vertical FIELD Data in vertical field determined by corresponding step S330
(S340)；
It is responded needed for the interaction of user determined by step S320 is intended to from being extracted in the knowledge mapping that step S340 is called
Knowledge information data (S350)；
According to the knowledge information data that step S350 is extracted, it is intended to generate for the interaction of user determined by step S320 more
Mode interaction response data (S360).
Further, in practical application scene, in theory, can not absolute guarantee got from webpage certain spy
All knowledge informations in fixed vertical field；Meanwhile also can not all knowledge informations for being got from webpage of absolute guarantee all
It is correct.Therefore, it is impossible to include all vertical FIELD Datas needed for user in absolute guarantee's knowledge mapping；Meanwhile also without
It is right-on that method absolute guarantee responds according to all multi-modal interactions that the vertical FIELD Data in knowledge mapping generates.
Based on the above issues, in one embodiment, by the way of dynamic corrections, constantly knowledge mapping is improved in amendment.
Even if that there are shortage of data and/or error in data is undesirable so as to cause user experience for current knowledge mapping, by continuous
Data correction is perfect, and shortage of data present in knowledge mapping and/or error in data are necessarily fewer and fewer, this is just finally realized
The continuous promotion of user experience.
Specifically, in one embodiment, web page monitored information (periodically or non-periodically) is compared new vertical in webpage information
Data with existing in domain knowledge information and knowledge mapping, when occurring the vertical neck for not including in knowledge mapping in webpage information
When domain knowledge information, new vertical domain knowledge information supplement realizes the self-perfection of knowledge mapping into knowledge mapping.
Further, in one embodiment, when the vertical FIELD Data that appearance is included with knowledge mapping in webpage information
When inconsistent vertical domain knowledge information, correct data are stored in knowledge mapping by the correctness both confirmed, thus
Realize the self-recision of knowledge mapping.
Further, in one embodiment, it is responded according to the interaction of user and determines whether current knowledge mapping meets use
Family demand, to according to user demand active correction, improve knowledge mapping.
Specifically, as shown in figure 4, in one embodiment, method further include:
Obtain the feedback data (S410) that user is directed to multi-modal interactive response data；
Determine that user is directed to the satisfaction of multi-modal interactive response data according to the feedback data that step S410 is obtained
(S420)；
Determine that the data correctness for the knowledge information data extracted from knowledge mapping and/or data are complete according to satisfaction
It spends (S430)；
The knowledge extracted from knowledge mapping is believed according to the step S430 data correctness determined and/or data integrity degree
Data are ceased to carry out data update and/or carry out data supplement (S440) to knowledge mapping.
Further, it based on method of the invention, the invention also provides a kind of storage medium, is stored on the storage medium
There is the program code that method as described herein can be achieved.
Further, based on method of the invention, the invention also provides a kind of building systems for knowledge mapping.Tool
Body, as shown in figure 5, in one embodiment, system includes:
Foundation structure constructing module 510 is configured to obtain semi-structured data；
Data extraction module 520 is configured to determine target domain, for target domain, carries out information extraction from webpage,
Obtain the vertical FIELD Data comprising entity node and/or node relationships；
Knowledge mapping generation module 530 is configured to add to vertical FIELD Data in semi-structured data, generates packet
Knowledge mapping containing vertical FIELD Data.
Specifically, in one embodiment, as shown in fig. 6, data extraction module includes:
Text recognizer 610 is configured to identification web page text；
Data cell 620 is crawled, the text identification result for being configured to text recognizer 610 obtains and target domain
Relevant data；
Data extracting unit 630, from the reality for crawling extraction semi-structured data missing in the data that data cell 620 obtains
Body node and/or node relationships.
Further, based on method of the invention, the invention also provides a kind of interactive systems.Specifically, implementing one
In example, as shown in fig. 7, system includes:
Input obtains module 710, is configured to acquisition user's multi-modal data；
Interaction parsing module 720 is configured to parsing user's multi-modal data, obtains user's interaction and is intended to and determines user
Interaction is intended to corresponding vertical field；
Knowledge mapping calling module 730 is configured to the knowledge mapping for calling building system proposed by the invention to generate,
The knowledge mapping includes the vertical FIELD Data in vertical field determined by corresponding interaction parsing module 720；
Generation module 740 is responded in interaction, is configured to extract from the knowledge mapping that knowledge mapping calling module 730 calls
The knowledge information data needed for user's interaction is intended to are responded, according to knowledge information data, are intended to generate multimode for user's interaction
State interacts response data.
Further, based on exchange method and interactive system of the invention, the invention also provides one kind based on virtual
The interactive system of people.Specifically, in one embodiment, system includes smart machine and cloud server, in which:
Cloud server includes interactive system as described in the present invention and multiple ability interfaces.Interactive system is configured to adjust
User's multi-modal data is obtained with the ability interface of cloud server and is parsed, generate and is exported multi-modal interactive response data.
Specifically, each ability interface calls corresponding logical process respectively in multi-modal data resolving.
Specifically, the ability interface of cloud server includes that semantic understanding interface, visual identity interface, affection computation connect
Mouth, cognition calculate interface.
Specifically, in one embodiment, the following are the explanations of each interface:
Semantic understanding interface receives the special sound instruction forwarded from the communication module of smart machine, carries out language to it
Sound identification and the natural language processing based on a large amount of corpus.
Visual identity interface, can be for human body, face, scene according to computer vision algorithms make, deep learning algorithm etc.
Carry out video content detection, identification, tracking etc..Image is identified according to scheduled algorithm, the detection knot of quantitative
Fruit.Have image preprocessing function, feature extraction functions and decision making function.Wherein:
Image preprocessing function, which can be, carries out basic handling to the vision collecting data of acquisition, including color space turns
It changes, edge extracting, image convert and image threshold；
Feature extraction functions can extract the features such as the colour of skin of target, color, texture, movement and coordinate in image and believe
Breath；
Decision making function can be to characteristic information, is distributed to according to certain decision strategy and needs the specific of this feature information
Multi-modal output equipment or multi-modal output application, such as realize Face datection, human limbs identification, motion detection function.
Affection computation interface is received the multi-modal data forwarded from communication module, (can be using affection computation logic
Emotion identification technology) calculate the current emotional state of user.Emotion identification technology is an important composition portion of affection computation
Point, the content of Emotion identification research includes facial expression, voice, behavior, text and physiological signal identification etc., by above
Content may determine that the emotional state of user.Emotion identification technology only can monitor user's by vision Emotion identification technology
Emotional state can also monitor the feelings of user using vision Emotion identification technology and sound Emotion identification technology in conjunction with by the way of
Not-ready status, and be not limited thereto.In the present embodiment, it is preferred to use the two in conjunction with mode monitor mood.
Affection computation interface is to collect human face's table by using image capture device when carrying out vision Emotion identification
Feelings image is then converted into that data can be analyzed, the technologies such as image procossing is recycled to carry out the analysis of expression mood.Understand facial table
Feelings, it usually needs the delicate variation of expression is detected, such as cheek muscle, mouth variation and choose eyebrow etc..
Cognition calculates interface, receives the multi-modal data forwarded from communication module, it is more to handle that cognition calculates interface
Modal data carries out data acquisition, identification and study, to obtain user's portrait, knowledge mapping etc., to multi-modal output data
Carry out Rational Decision.
It should be understood that disclosed embodiment of this invention is not limited to specific structure disclosed herein, processing step
Or material, and the equivalent substitute for these features that those of ordinary skill in the related art are understood should be extended to.It should also manage
Solution, term as used herein is used only for the purpose of describing specific embodiments, and is not intended to limit.
" embodiment " mentioned in specification means that a particular feature, structure, or characteristic described in conjunction with the embodiments is included in
In at least one embodiment of the present invention.Therefore, the phrase " embodiment " that specification various places throughout occurs might not
Refer both to the same embodiment.
While it is disclosed that embodiment content as above but described only to facilitate understanding the present invention and adopting
Embodiment is not intended to limit the invention.Method of the present invention can also have other various embodiments.Without departing substantially from
In the case where essence of the present invention, those skilled in the art make various corresponding changes or change in accordance with the present invention
Shape, but these corresponding changes or deformation all should belong to scope of protection of the claims of the invention.
Claims (10)
1. a kind of method for constructing knowledge mapping, which is characterized in that the described method includes:
Obtain semi-structured data；
For target domain, information extraction is carried out from webpage, obtains the vertical field number comprising entity node and/or node relationships
According to；
The vertical FIELD Data is added in the semi-structured data, the knowledge graph comprising vertical FIELD Data is generated
Spectrum.
2. the method according to claim 1, wherein carrying out information extraction from webpage, comprising:
Text identification identifies web page text；
Data are crawled, data relevant to the target domain are obtained based on text identification result；
Entity extraction and relationship are extracted, extracted from the data of acquisition the semi-structured data missing entity node and/
Or node relationships.
3. according to the method described in claim 2, it is characterized in that, described crawl data step further include:
Data cleansing, the data cleansing are non-physical cleaning, wherein are filtered for specific vocabulary.
4. according to the method in claim 2 or 3, which is characterized in that the entity extraction includes:
Name Entity recognition, wherein be named Entity recognition using the method for deep learning.
5. the method according to any one of claim 2~4, which is characterized in that carry out node relationships based on neural network
It extracts.
6. a kind of exchange method based on any one of Claims 1 to 5, which comprises
It obtains user's multi-modal data and parses, determine that user's interaction is intended to；
Determine that user's interaction is intended to corresponding vertical field；
Knowledge mapping is called, the knowledge mapping includes the vertical FIELD Data in the corresponding vertical field；
The knowledge information data responded needed for user's interaction is intended to are extracted from the knowledge mapping；
According to the knowledge information data, it is intended to generate multi-modal interactive response data for user interaction.
7. according to the method described in claim 6, it is characterized in that, the method also includes:
Obtain the feedback data that user is directed to the multi-modal interactive response data；
Determine that the user is directed to the satisfaction of the multi-modal interactive response data according to the feedback data；
The data correctness and/or data of the knowledge information data extracted from the knowledge mapping are determined according to the satisfaction
Integrity degree；
According to the data correctness and/or the data integrity degree to the knowledge information data extracted from the knowledge mapping
It carries out data update and/or data supplement is carried out to the knowledge mapping.
8. a kind of storage medium, which is characterized in that being stored on the storage medium can be achieved such as any one of claim 1-7
The program code of the method.
9. a kind of building system for knowledge mapping, which is characterized in that the system comprises:
Foundation structure constructing module is configured to obtain semi-structured data；
Data extraction module is configured to determine target domain, for the target domain, carries out information extraction from webpage, obtains
Take the vertical FIELD Data comprising entity node and/or node relationships；
Knowledge mapping generation module is configured to add to the vertical FIELD Data in the semi-structured data, generates
Knowledge mapping comprising vertical FIELD Data.
10. a kind of interactive system, which is characterized in that the system comprises:
Input obtains module, is configured to acquisition user's multi-modal data；
Interaction parsing module is configured to parse user's multi-modal data, obtains user's interaction and is intended to and determines the use
Family interaction is intended to corresponding vertical field；
Knowledge mapping calling module is configured to the knowledge mapping for calling system as claimed in claim 9 to generate, the knowledge graph
Vertical FIELD Data of the spectrum comprising the corresponding vertical field；
Generation module is responded in interaction, is configured to extract from the knowledge mapping and is responded knowing needed for user's interaction is intended to
Know information data, according to the knowledge information data, is intended to generate multi-modal interactive response data for user interaction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810842317.8A CN109086391B (en) | 2018-07-27 | 2018-07-27 | Method and system for constructing knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810842317.8A CN109086391B (en) | 2018-07-27 | 2018-07-27 | Method and system for constructing knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109086391A true CN109086391A (en) | 2018-12-25 |
| CN109086391B CN109086391B (en) | 2022-07-01 |
Family
ID=64831028
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810842317.8A Active CN109086391B (en) | 2018-07-27 | 2018-07-27 | Method and system for constructing knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109086391B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851612A (en) * | 2019-08-29 | 2020-02-28 | 国家计算机网络与信息安全管理中心 | Encyclopedic knowledge-based mobile application knowledge graph composite completion method and device |
| CN114491085A (en) * | 2022-04-15 | 2022-05-13 | 支付宝(杭州)信息技术有限公司 | Graph data storage method and distributed graph data calculation method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090187444A1 (en) * | 2007-05-11 | 2009-07-23 | Yefim Zhuk | Service knowledge map |
| CN105956052A (en) * | 2016-04-27 | 2016-09-21 | 青岛海尔软件有限公司 | Building method of knowledge map based on vertical field |
| CN106815293A (en) * | 2016-12-08 | 2017-06-09 | 中国电子科技集团公司第三十二研究所 | System and method for constructing knowledge graph for information analysis |
| CN106960005A (en) * | 2017-02-20 | 2017-07-18 | 北京光年无限科技有限公司 | A kind of output intent and robot for artificial intelligence robot |
| CN107491555A (en) * | 2017-09-01 | 2017-12-19 | 北京纽伦智能科技有限公司 | Knowledge mapping construction method and system |
| CN107665252A (en) * | 2017-09-27 | 2018-02-06 | 深圳证券信息有限公司 | A kind of method and device of creation of knowledge collection of illustrative plates |
-
2018
- 2018-07-27 CN CN201810842317.8A patent/CN109086391B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090187444A1 (en) * | 2007-05-11 | 2009-07-23 | Yefim Zhuk | Service knowledge map |
| CN105956052A (en) * | 2016-04-27 | 2016-09-21 | 青岛海尔软件有限公司 | Building method of knowledge map based on vertical field |
| CN106815293A (en) * | 2016-12-08 | 2017-06-09 | 中国电子科技集团公司第三十二研究所 | System and method for constructing knowledge graph for information analysis |
| CN106960005A (en) * | 2017-02-20 | 2017-07-18 | 北京光年无限科技有限公司 | A kind of output intent and robot for artificial intelligence robot |
| CN107491555A (en) * | 2017-09-01 | 2017-12-19 | 北京纽伦智能科技有限公司 | Knowledge mapping construction method and system |
| CN107665252A (en) * | 2017-09-27 | 2018-02-06 | 深圳证券信息有限公司 | A kind of method and device of creation of knowledge collection of illustrative plates |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851612A (en) * | 2019-08-29 | 2020-02-28 | 国家计算机网络与信息安全管理中心 | Encyclopedic knowledge-based mobile application knowledge graph composite completion method and device |
| CN110851612B (en) * | 2019-08-29 | 2023-08-18 | 国家计算机网络与信息安全管理中心 | Mobile application knowledge graph composite completion method and device based on encyclopedia knowledge |
| CN114491085A (en) * | 2022-04-15 | 2022-05-13 | 支付宝(杭州)信息技术有限公司 | Graph data storage method and distributed graph data calculation method |
| CN114491085B (en) * | 2022-04-15 | 2022-08-09 | 支付宝(杭州)信息技术有限公司 | Graph data storage method and distributed graph data calculation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109086391B (en) | 2022-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Multimodal large language models: A survey | |
| CN106997243B (en) | Speech scene monitoring method and device based on intelligent robot | |
| CN109522835A (en) | Children's book based on intelligent robot is read and exchange method and system | |
| CN109271018A (en) | Exchange method and system based on visual human's behavioral standard | |
| CN107301168A (en) | Intelligent robot and its mood exchange method, system | |
| CN109871450A (en) | Based on the multi-modal exchange method and system for drawing this reading | |
| CN107797663A (en) | Multi-modal interaction processing method and system based on visual human | |
| CN108942919A (en) | A kind of exchange method and system based on visual human | |
| CN109176535A (en) | Exchange method and system based on intelligent robot | |
| CN109948447A (en) | The discovery of personage's cyberrelationship and evolution rendering method based on video image identification | |
| Lin et al. | Deep structured scene parsing by learning with image descriptions | |
| CN109145168A (en) | A kind of expert service robot cloud platform | |
| CN107016046A (en) | The intelligent robot dialogue method and system of view-based access control model displaying | |
| CN113971208B (en) | Video object positioning method and system based on mixed attention mechanism | |
| CN105355200A (en) | System and method for training and modifying interactive content of robot directly | |
| CN109343695A (en) | Exchange method and system based on visual human's behavioral standard | |
| CN109086351A (en) | A kind of method and user tag system obtaining user tag | |
| CN107784355A (en) | The multi-modal interaction data processing method of visual human and system | |
| CN109278051A (en) | Exchange method and system based on intelligent robot | |
| CN110472532A (en) | A kind of the video object Activity recognition method and apparatus | |
| CN108595012A (en) | Visual interactive method and system based on visual human | |
| CN106126636B (en) | A kind of man-machine interaction method and device towards intelligent robot | |
| CN109086391A (en) | A kind of method and system constructing knowledge mapping | |
| CN109977238A (en) | Generate the system for drawing this, method and apparatus | |
| CN108415561A (en) | Gesture interaction method based on visual human and system |
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 |