CN106874261A - A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle - Google Patents
A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle Download PDFInfo
- Publication number
- CN106874261A CN106874261A CN201710160270.2A CN201710160270A CN106874261A CN 106874261 A CN106874261 A CN 106874261A CN 201710160270 A CN201710160270 A CN 201710160270A CN 106874261 A CN106874261 A CN 106874261A
- Authority
- CN
- China
- Prior art keywords
- concept
- entity
- illustrative plates
- domain knowledge
- physical layer
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
-
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle.Collection of illustrative plates of the invention includes conceptual level, physical layer；Conceptual level is combined into by the collection of concept, and each concept is provided with unique mark, and represents lexical item by one and candidate's lexical item set is represented；According to the concept set of the related knowledge in field and conceptual level, the corresponding multiple entities of each concept are instantiated, constitute physical layer, entity is the extension of concept, and each entity is provided with unique mark, and represent lexical item by one and candidate's lexical item set is represented；The incidence relation set up according to field relevant knowledge is provided between related notion；The incidence relation set up according to field relevant knowledge is provided between conceptual level and physical layer；The incidence relation built according to field relevant knowledge is provided between entity.This method realizes concept and is separated with entity, facilitates the combing of knowledge, and has distinguished the not same-action of concept and entity in knowledge understanding and utilization, improves search efficiency.
Description
Technical field
The invention belongs to artificial intelligence field, and in particular to a kind of domain knowledge collection of illustrative plates and issuer based on semantic triangle
Method.
Background technology
With developing rapidly for Internet technology, when the mankind successively experienced " Web 1.0 " with document as principal character
Generation and " Web 2.0 " epoch being characterized with data interconnection data sharing, and moving towards knowledge based interconnection
" Web3.0 " epoch.An intelligible knowledge network, a major challenge as the modern the Internet epoch are represented in big data.
Equally, the basis of man-machine interaction is to understand the verbal information of user, infer the intention of user and then return to user by machine
As a result.Knowledge mapping is arisen at the historic moment with its powerful semantic processing ability and the ability of opening and interconnecting, can be by internet
Magnanimity, the data of isomery are collected as intelligible knowledge network, readily appreciate, apply.Therefore, certain specific area how is represented
Knowledge mapping and be applied to specific business scenario to solve corresponding problem, tool is of great significance.
The semantic network that the expression of existing knowledge mapping describe between entity mostly, each entity (concept outward
Prolong) identified with an ID for globally unique determination, represent knowledge network in the form of master-meaning-guest's triple.But it is this
Method for expressing can bring obscuring for concept and entity, so as to cause the disturbance of understanding of user, and be unfavorable for the understanding of entity.Pin
To specific problem, it is necessary to be inquired about in it there is the knowledge mapping of mass data and relation, cost is of a relatively high.Traditional knowledge
Also there are polysemy and adopted many word problems in collection of illustrative plates so that semantic network is chaotic, be unfavorable for the word enriched from internet
In sort out knowledge and carry out quick search.
Semantic triangle is a kind of theory on meaning, by British scholar Ao Gedeng (Ogden) and Richards
(Richards) the semantics seminal book published in nineteen twenty-three《The meaning of meaning》In (The Meaning of Meaning)
Propose, represent the exemplary aspect of traditional semantics.A kind of phase of symbol, concept (meaning) and objective things between is described
Mutually restriction, the relation for interacting, emphasize that linguistic notation is the reference to things, and reference process is exactly symbol, concept (meaning)
The process being related with things.Symbol is used for representing concept that concept is the think of for reflecting objective things or object essential attribute
Dimension form, is the thought of word；Concept reflects reality the i.e. signified thing of objective things in the world.Semantic triangle includes tripartite
Relation of plane, is respectively the direct relation of concept/between meaning and objective things, the direct relation between concept and symbol/word
And the indirect relationship between symbol or word and denotion thing/things.
How using semantic triangle thought, specific area be magnanimity, the big data of isomery expression, tissue, management and
Using providing a kind of more efficiently mode so that semantic network can realize specific area concept and entity organization and administration and
The function of quick search, is one of current problem demanding prompt solution of knowledge mapping.
The content of the invention
The present invention is regarding to the issue above, it is proposed that a kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle.Should
Concept and entity separation are represented word and multiple candidate words representing as concept or entity by collection of illustrative plates with one, entity it
Between relation set up by super side based on concept node, it is achieved thereby that effective management of knowledge network and quick search.
The technical solution adopted by the present invention is as follows：
Domain knowledge collection of illustrative plates based on semantic triangle is made up of conceptual level, physical layer, is embodied as following five part：
1., according to the knowledge architecture conceptual level that field is related, conceptual level is combined into by the collection of concept, and each concept is provided with only
One mark, and represent lexical item by one and candidate's lexical item set is represented.Concept correspondence semantic triangle in concept/meaning, be
Summarize on the basis of the objective things in the field.Each candidate word in candidate's lexical item set can be understood as representing word
The synonym or near synonym of item, and the concept is represented with certain probability.In addition, conceptual level can be more than one layer, can be according to difference
Business scenario refinement concepts layer, it is more accurately portrayed domain knowledge, meet specific task.
2., according to the concept set of the related knowledge in field and conceptual level, the corresponding multiple entities of each concept are instantiated,
Constitute physical layer.Entity is the extension of concept, and each entity is provided with unique mark, and represents lexical item and candidate word item collection by one
Close and represent.Entity is obtained by the conceptual example (instanceOf) in conceptual level, i.e., according to domain knowledge that entity is corresponding
Onto related notion.Each candidate word in candidate's lexical item set can be understood as representing the synonym of lexical item, with certain probability
Represent the entity.Here probability can be set by experience, and the method that may be based on statistical analysis learns out.
3., according to field relevant knowledge, by methods such as empirical rule reasoning or statistical learnings, represent that conceptual level is related
Incidence relation between concept.There is following several relations between concept：
(1) inclusion relation (has) between concept and its attribute.If the attribute of concept is also as one in conceptual level
Concept occurs, then the relation between both is inclusion relation (has).
(2) membership (isA) between concept.If a concept is the subset of another concept connotation, two
Relation between person is membership (isA).
(3) incidence relation between concept.If there are other incidence relations between concept, can be defined according to specific field
Specific incidence relation, such as expansion relation etc..
4., according to field relevant knowledge, the incidence relation between conceptual level and physical layer is represented.Concept is most direct with entity
Relation be conceptual example (instanceOf), it is recorded in physical layer generating process.In addition, also exist and use super
While the concept for representing and entity based on relation (basedOn), the relation between entity that can determine whether is based on certain category of conceptual level
When property concept, entity relationship is just connected with attributive concept by a super side, for representing based on relation (basedOn).
5., according to field relevant knowledge, build independently of the incidence relation between the entity of conceptual level.
The method for expressing of above knowledge mapping is suitable for each specific area, with universality and versatility.But for not
Same domain, its specific construction method and technology are then different, and being potentially based on domain knowledge carries out artificial judgment, it is also possible to base
Be automatically performed in the method for statistical learning, this is accomplished by the accumulation degree according to domain knowledge, and field related data rule
Mould considers and design with that can obtain the combined factors such as degree.The present invention is directed to propose a kind of general domain knowledge collection of illustrative plates side of expression
Method, and specific building process will for different field the characteristics of, studied respectively and designed, also will in other patents body
It is existing.
Based on above-mentioned knowledge mapping, efficiently the concept and entity in field can be inquired about, and feed back and inquire about right
As the Query Result being associated.For a request to be checked, query process is divided into following steps：
1. the query statement participle for user being given, and remove participle storage after stop words therein in set S.
2. the entity in each lexical item and the concept and physical layer of knowledge mapping conceptual level in S is matched：
(1) it is general by the attribute concept related with this is associated with by knowledge mapping when certain concept of conceptual level is matched
Read, and the entity sets obtained after the conceptual example, and returned as Query Result；
(2) when certain entity of physical layer is matched, the attribute with the entity associated will be associated with by knowledge mapping,
And the corresponding concept of the entity, and returned as Query Result；
(3) when multiple entities of physical layer are matched, check whether they have side phase in the physical layer of knowledge mapping
Connection, if while be connected and this while constitute a super side of BasedOn, then the concept that this super side connects also is included into inquiry and tied
Really, so that relationship type that may be between default entity in understanding query statement exactly；
(4) when the concept for matching conceptual level, and the concept and match physical layer simultaneously an entity between exist
During the super sides of BasedOn, another entity that this super side connects also is included into Query Result, so as to be fully understood by concept and reality
Incidence relation between body.
Compared with prior art, beneficial effects of the present invention are as follows：
1) this method realizes concept and is separated with entity, and user is when specific area faces particular task, it is easy to
Start with searching solution from abstract concept, it is convenient compared to the method that entitative concept in traditional knowledge mapping is defined together
The combing of knowledge, and distinguished the not same-action of concept and entity in knowledge understanding and utilization.
2) this method realizes the quick search of knowledge in knowledge mapping.The method for expressing of concept and entity separation is by knowledge
The arrangement of stratification is carried out, when user carries out certain specific knowledge to be inquired about, the step from concept to entity can have been followed
Suddenly carry out, reduce inquiry cost, improve search efficiency.
3) present method solves adopted many word problems.Each concept and entity represent lexical item and a candidate by one
Lexical item set represents, can soon understand the intention of user, finds the synonymy between word.
4) present method solves the problem of polysemy.The either "Yes" in Chinese and " having ", or in English
" IsA " and " Has " is all applied too heavily the various relations between presentation-entity or between concept, and this method is several important
Relation explicitly indicate that out, eliminate some Ambiguities in traditional knowledge mapping.
5) this method contributes to the intention of computer understanding user, improves the experience of user, helps user more efficient
Ground completes the particular task of specific area, and then lifts the quality of nature man-machine interaction.
Brief description of the drawings
Fig. 1 is the semantic triangle schematic diagram of this method；
Fig. 2 is the knowledge mapping schematic diagram in Finance Audit field；
Fig. 3 is the knowledge mapping schematic diagram in Finance Audit field；
Fig. 4 is the knowledge mapping schematic diagram in Finance Audit field；
Fig. 5 is the knowledge mapping schematic diagram of ecological system regions.
Specific embodiment
In order that those skilled in the art more fully understand the present invention, further retouched in detail below in conjunction with example and accompanying drawing
The present invention is stated, but is not construed as limiting the invention.
Fig. 1 is the semantic triangle of this method, is close-coupled relation between concept and entity, concept and expression, entity and
It is loose couplings relation between expression.
Fig. 2 is the example of the knowledge mapping in Finance Audit field, and its structure can be divided into conceptual network and physical network two
Layer：
1. conceptual network includes five concepts, is respectively loan, contract, non-performing loan, Contract NO and the amount of the loan.
(1) each concept is identified with unique CID, and represents lexical item by one and candidate's lexical item set is represented, contract
Number, Contract Document number there is identical implication as candidate's lexical item set and Contract NO.Similarly, in audit field, loan gold
Volume, loan settlement remaining sum can also be expressed and loan balance identical implication.
(2) attribute of contract is Contract NO, and Contract NO is used for identifying a contract, therefore contract is compiled with contract
It is inclusion relation (has) between number.Similarly, between non-performing loan and the amount of the loan it is also inclusion relation (has).
(3) in audit field, the process of loan is often related to the signing of contract, and auditor checks related loan
When information, the treaty content that the loan is related to can be generally also checked, therefore have extension between contract and loan
(extend) relation.The foundation of this relation simplifies the operation of auditor.
(4) in audit field, loan wherein some be non-performing loan, therefore loan and non-performing loan between be to include
Relation (include).
2. physical network includes 4 entities, is respectively contract entity 1, XXXX100,1,1000000 yuan of entity of providing a loan.This
A little entities can be obtained by the conceptual example (InstanceOf) in conceptual level.Relation between entity is inherited in conceptual level
Relation between concept.Such as, there is expansion relation, therefore contract entity 1 and loan between conceptual level contract and non-performing loan
Also there is corresponding expansion relation between money entity.
3. conceptual level be most directly instantiation with associating for physical layer.In addition, one of attribute of contract entity 1 is
Contract NO XXXX100, and this relation is based on this concept of Contract NO, therefore the relation on attributes between this example
There is BasedOn relations between the Contract NO as concept.
Auditor utilizes this knowledge mapping based on semantic triangle, can more quickly solve the problems, such as correlation.
Traditional knowledge mapping obscures concept and entity, cause user to occur that unclear classification is clear in use, returning result not
The problems such as specifying.For example, in Finance Audit domain knowledge collection of illustrative plates (as shown in Figure 2), when user's inquiry, " which non-performing loan has
When a bit ", the present invention is stored in set S={ " non-performing loan ", " which " }, then by S first by query statement participle
Lexical item is matched with the entity in concept and physical layer in conceptual level, and " non-performing loan " is found in conceptual level, then can be with
It is instantaneously mapped to belong to the entity sets of non-performing loan in physical layer, and is returned as Query Result.However, in traditional knowledge
Then need to travel through the non-performing loan entity that all entity nodes find correlation in collection of illustrative plates, search efficiency is relatively low.
For another example, when user's inquiry, " amount of the loan of the non-performing loan of loan entity 1 is how many" when, first will inquiry
Sentence participle, stores at set S={ " loan entity 1 ", " non-performing loan ", " amount of the loan ", " how many " }, then in conceptual level
In find " non-performing loan " and " amount of the loan ", entity " loan entity 1 " is found in physical layer, and then by " loan entity
1 " with the super side where " amount of the loan ", the entity " 1000000 yuan " of physical layer is expanded to, is returned as Query Result.
Further, collection of illustrative plates of the invention summarises the knowledge in field from abstract concept, is a kind of for domain knowledge
Taxonomic revision, is conducive to knowledge system construction of the user from abstract angle understanding field.For example, auditor can design not
Same audit analysis model carries out the task of correlation, and the knowledge mapping of this method design can help auditor to arrange thinking simultaneously
Design correlation model.Knowledge mapping as shown in Figure 3, when the risk of identification is " financing deposition ", output model " sink by loan fund
Form sediment "；During identification relevant risk " break the whole up into parts, to borrow repayings ... ", output recommended models " payment funding breaks the whole up into parts ", " with loan
Repay " etc..
In terms of instance layer, the query function except that can carry out entity can also be by knowledge mapping reality of the invention
The task such as existing deceptive information identification and abnormality detection.As shown in figure 4, consistency check can be used to monitor deceptive information, such as
When there is identical legal person, mailing address, telephone number, business license in different enterprises, auditor just should to this two
The information of enterprise of family is examined；Especially, when telephone number and business licence number repeat, if not data record
Enter mistake, then just there is an enterprise to provide deceptive information certainly.
Fig. 5 is the ecosphere knowledge mapping built for Xinjiang big data ecological center, by ecological environment
Fundamental and the fact are modeled, all kinds of knowledge of combing, form complete system.Knowledge mapping is divided into three layers, wherein two
Layer is conceptual level, and one layer is physical layer.It is specific as follows：
1) ecological conceptual level.The main key concept for describing ecological environment, such as water, soil and air etc..Wherein
In the presence of corresponding membership (isA), such as underground water, precipitation and urban river water belong to water this concept.
2) conceptual level is measured.The various Measure Indexes of ecological environment are essentially described, is such as wrapped in this concept of underground water
Containing another measurement concept, underground water content.
3) physical layer.Concept in conceptual level is instantiated, for example this concept of Tianshan Area laterite succession laterite,
20ml is inherited from soil moisture content this concept.Wherein, 20ml is the water content of Tianshan Area laterite, there is attribute between the two
Relation (attr), this relation has basedOn based on this concept of soil moisture content between soil moisture content
Super frontier juncture system.
Above-mentioned knowledge mapping can be applied to the monitoring of Xinjiang region ecological environment.By the measurement number to various kinds of sensors
According to tentatively being sorted out, and according to incidence relation, real time data is monitored and reasoning.Sensing data (structuring) and
Text data (destructuring) combines, and the monitored results of each given area are dynamically built into physical layer.Due to general
Read layer reflection be more normal ecological recycle system, when find entity layer building the circulatory system and conceptual level difference compared with
When big, i.e., when figure is changed greatly, it is believed that generate abnormal ecological phenomenon, related personnel needs the finger according to change
The reason for mark investigation problem occurs, and carry out respective handling.Ecological environment knowledge mapping can also be provided and operating personnel's nature
Interactive function, is sorted out by the inquiry and action type that represent sentence, for the key being related in enquirement sentence
Attribute, the part to lacking is filled or is interacted with user automatically, progressively understands the intention of operating personnel, gives phase
The feedback information answered, improves the operating efficiency of operating personnel.
The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, the ordinary skill of this area
Personnel can modify or equivalent to technical scheme, without departing from the spirit and scope of the present invention, this
The protection domain of invention should be to be defined described in claim.
Claims (10)
1. a kind of domain knowledge collection of illustrative plates based on semantic triangle, it is characterised in that including conceptual level, physical layer；Wherein,
The conceptual level is a concept set related to target domain knowledge, and each concept is provided with unique mark, and uses one
Represent lexical item and candidate's lexical item set is represented；Concept or meaning in the concept correspondence semantic triangle, the time of concept i
Select the synonym or near synonym of the representative lexical item that each candidate word in lexical item set is concept i, the corresponding each times of concept i
Select word that one probable value for representing concept i is set；
The physical layer is an entity sets, and each concept in the concept set is carried out according to the knowledge that target domain is related
Instantiation obtains the corresponding multiple entities of each concept, constitutes the entity sets；Each entity is provided with unique mark, is used in combination
One represents lexical item and candidate's lexical item set；Each candidate word in candidate's lexical item set of entity i is the generation of entity i
The synonym of table lexical item, the corresponding each candidate words of entity i set a probable value for representing entity i；
The incidence relation set up according to field relevant knowledge is provided between the related notion of the conceptual level；Conceptual level and entity
The incidence relation set up according to field relevant knowledge is provided between layer；It is provided between the entity according to field relevant knowledge
The incidence relation of structure.
2. domain knowledge collection of illustrative plates as claimed in claim 1, it is characterised in that the incidence relation bag between the related notion
Include：
A) inclusion relation, is inclusion relation if concept and its attribute are in the concept set, between concept and its attribute；
B) membership, if concept is the subset of another concept connotation, relation between the two is and is subordinate to pass
System.
3. domain knowledge collection of illustrative plates as claimed in claim 1, it is characterised in that between the concept and the entity, described general
It is close-coupled relation to read and the expression of the concept between, is that loose couplings are closed between the expression of the entity and the entity
System.
4. the domain knowledge collection of illustrative plates as described in claim 1 or 2 or 3, it is characterised in that the conceptual level is according to different business
Scene is divided into some concept sublayers, one concept subset of each concept sublayer correspondence.
5. the domain knowledge collection of illustrative plates as described in claim 1 or 2 or 3, it is characterised in that between the conceptual level and physical layer
Incidence relation includes：Corresponding relation during conceptual example between concept and entity, the concept represented with super side and entity
BasedOn relations.
6. the domain knowledge collection of illustrative plates as described in claim 1 or 2 or 3, it is characterised in that the probable value is set by experience.
7. the domain knowledge collection of illustrative plates as described in claim 1 or 2 or 3, it is characterised in that the method setting institute based on statistical analysis
State probable value.
8. the domain knowledge collection of illustrative plates as described in claim 1 or 2 or 3, it is characterised in that by empirical rule reasoning or statistics
Learning method, sets the incidence relation between the related notion.
9. a kind of querying method based on domain knowledge collection of illustrative plates described in claim 1, its step is：
1) by query statement participle, and by participle storage in a set S；
2) entity in the concept and physical layer in the conceptual level by each participle in set S with the domain knowledge collection of illustrative plates
Matched：
A) when a concept of conceptual level is matched, the attributive concept concept related with this is associated with by domain knowledge collection of illustrative plates,
And the entity sets obtained after the conceptual example, and returned as Query Result；
B) when an entity of physical layer is matched, the attribute with the entity associated will be associated with by domain knowledge collection of illustrative plates,
And the corresponding concept of the entity, and returned as Query Result；
C) when multiple entities of physical layer are matched, check whether they have side in the physical layer of the domain knowledge collection of illustrative plates
Be connected, if while be connected and this while constitute a super side of BasedOn, then the concept that the super sides of the BaseOn connect also is returned
Enter Query Result；
D) when the concept for matching conceptual level, and the concept and match physical layer simultaneously an entity between exist
During the super sides of BasedOn, another entity that this super side connects also is included into Query Result.
10. method as claimed in claim 9, it is characterised in that step 1) in, by query statement participle, and remove therein
After stop words by participle storage in a set S.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710160270.2A CN106874261A (en) | 2017-03-17 | 2017-03-17 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710160270.2A CN106874261A (en) | 2017-03-17 | 2017-03-17 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106874261A true CN106874261A (en) | 2017-06-20 |
Family
ID=59171337
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710160270.2A Pending CN106874261A (en) | 2017-03-17 | 2017-03-17 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106874261A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609052A (en) * | 2017-08-23 | 2018-01-19 | 中国科学院软件研究所 | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle |
| CN108446769A (en) * | 2018-01-23 | 2018-08-24 | 深圳市阿西莫夫科技有限公司 | Knowledge mapping relation inference method, apparatus, computer equipment and storage medium |
| CN108563749A (en) * | 2018-04-16 | 2018-09-21 | 中山大学 | On-line education system resource recommendation method based on various dimensions information and knowledge network |
| CN108959433A (en) * | 2018-06-11 | 2018-12-07 | 北京大学 | A kind of method and system extracting knowledge mapping and question and answer from software project data |
| CN109472033A (en) * | 2018-11-19 | 2019-03-15 | 华南师范大学 | Entity relation extraction method and system in text, storage medium, electronic equipment |
| CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
| CN110008354A (en) * | 2019-04-10 | 2019-07-12 | 华侨大学 | A kind of construction method of the external Chinese studying content of knowledge based map |
| CN110019560A (en) * | 2017-12-28 | 2019-07-16 | 中国移动通信集团上海有限公司 | A kind of querying method and device of knowledge based map |
| CN110019824A (en) * | 2017-07-20 | 2019-07-16 | 苏州狗尾草智能科技有限公司 | Man-machine interaction method, the apparatus and system of knowledge based map |
| CN110740196A (en) * | 2018-07-19 | 2020-01-31 | 中国移动通信有限公司研究院 | Processing method and device for Internet of things identifiers and terminal equipment |
| CN111091006A (en) * | 2019-12-20 | 2020-05-01 | 北京百度网讯科技有限公司 | Entity intention system establishing method, device, equipment and medium |
| CN111259166A (en) * | 2020-01-22 | 2020-06-09 | 清华大学 | Scientific research entity linking method and device based on knowledge graph |
| CN111611343A (en) * | 2020-04-28 | 2020-09-01 | 北京智通云联科技有限公司 | Knowledge graph shortest path query-based search system, method and equipment |
| CN111949828A (en) * | 2019-05-17 | 2020-11-17 | 南京农业大学 | Automatic construction method of rice cultivation mode knowledge graph facing semi-structured form |
| CN112528291A (en) * | 2020-12-15 | 2021-03-19 | 青岛海信智慧生活科技股份有限公司 | Code auditing method and device based on knowledge graph |
| CN112784064A (en) * | 2021-01-28 | 2021-05-11 | 电子科技大学 | Hierarchical super-dimensional knowledge graph construction method facing social network |
| CN113553444A (en) * | 2021-07-27 | 2021-10-26 | 之江实验室 | Audit knowledge graph representation model based on excess edges and associated reasoning method |
| CN113988724A (en) * | 2021-12-28 | 2022-01-28 | 深圳市迪博企业风险管理技术有限公司 | Risk analysis method for financial activity knowledge graph of listed company |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663122A (en) * | 2012-04-20 | 2012-09-12 | 北京邮电大学 | Semantic query expansion algorithm based on emergency ontology |
-
2017
- 2017-03-17 CN CN201710160270.2A patent/CN106874261A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663122A (en) * | 2012-04-20 | 2012-09-12 | 北京邮电大学 | Semantic query expansion algorithm based on emergency ontology |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110019824A (en) * | 2017-07-20 | 2019-07-16 | 苏州狗尾草智能科技有限公司 | Man-machine interaction method, the apparatus and system of knowledge based map |
| CN107609052A (en) * | 2017-08-23 | 2018-01-19 | 中国科学院软件研究所 | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle |
| CN107609052B (en) * | 2017-08-23 | 2019-09-24 | 中国科学院软件研究所 | A kind of generation method and device of the domain knowledge map based on semantic triangle |
| CN110019560A (en) * | 2017-12-28 | 2019-07-16 | 中国移动通信集团上海有限公司 | A kind of querying method and device of knowledge based map |
| CN110019560B (en) * | 2017-12-28 | 2021-09-21 | 中国移动通信集团上海有限公司 | Knowledge graph-based query method and device |
| CN108446769A (en) * | 2018-01-23 | 2018-08-24 | 深圳市阿西莫夫科技有限公司 | Knowledge mapping relation inference method, apparatus, computer equipment and storage medium |
| CN108446769B (en) * | 2018-01-23 | 2020-12-08 | 深圳市阿西莫夫科技有限公司 | Knowledge graph relation inference method, knowledge graph relation inference device, computer equipment and storage medium |
| CN108563749B (en) * | 2018-04-16 | 2020-11-10 | 中山大学 | Online education system resource recommendation method based on multi-dimensional information and knowledge network |
| CN108563749A (en) * | 2018-04-16 | 2018-09-21 | 中山大学 | On-line education system resource recommendation method based on various dimensions information and knowledge network |
| CN108959433A (en) * | 2018-06-11 | 2018-12-07 | 北京大学 | A kind of method and system extracting knowledge mapping and question and answer from software project data |
| CN110740196A (en) * | 2018-07-19 | 2020-01-31 | 中国移动通信有限公司研究院 | Processing method and device for Internet of things identifiers and terminal equipment |
| CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
| CN109472033B (en) * | 2018-11-19 | 2022-12-06 | 华南师范大学 | Method and system for extracting entity relationship in text, storage medium and electronic equipment |
| CN109472033A (en) * | 2018-11-19 | 2019-03-15 | 华南师范大学 | Entity relation extraction method and system in text, storage medium, electronic equipment |
| CN110008354B (en) * | 2019-04-10 | 2022-06-07 | 华侨大学 | Method for constructing foreign Chinese learning content based on knowledge graph |
| CN110008354A (en) * | 2019-04-10 | 2019-07-12 | 华侨大学 | A kind of construction method of the external Chinese studying content of knowledge based map |
| CN111949828A (en) * | 2019-05-17 | 2020-11-17 | 南京农业大学 | Automatic construction method of rice cultivation mode knowledge graph facing semi-structured form |
| CN111091006A (en) * | 2019-12-20 | 2020-05-01 | 北京百度网讯科技有限公司 | Entity intention system establishing method, device, equipment and medium |
| CN111091006B (en) * | 2019-12-20 | 2023-08-29 | 北京百度网讯科技有限公司 | Method, device, equipment and medium for establishing entity intention system |
| CN111259166B (en) * | 2020-01-22 | 2023-02-03 | 清华大学 | Scientific research entity linking method and device based on knowledge graph |
| CN111259166A (en) * | 2020-01-22 | 2020-06-09 | 清华大学 | Scientific research entity linking method and device based on knowledge graph |
| CN111611343A (en) * | 2020-04-28 | 2020-09-01 | 北京智通云联科技有限公司 | Knowledge graph shortest path query-based search system, method and equipment |
| CN112528291B (en) * | 2020-12-15 | 2022-09-16 | 青岛海信智慧生活科技股份有限公司 | Code auditing method and device based on knowledge graph |
| CN112528291A (en) * | 2020-12-15 | 2021-03-19 | 青岛海信智慧生活科技股份有限公司 | Code auditing method and device based on knowledge graph |
| CN112784064A (en) * | 2021-01-28 | 2021-05-11 | 电子科技大学 | Hierarchical super-dimensional knowledge graph construction method facing social network |
| CN112784064B (en) * | 2021-01-28 | 2023-05-12 | 电子科技大学 | Hierarchical super-dimensional knowledge graph construction method for social network |
| CN113553444A (en) * | 2021-07-27 | 2021-10-26 | 之江实验室 | Audit knowledge graph representation model based on excess edges and associated reasoning method |
| CN113988724B (en) * | 2021-12-28 | 2022-03-22 | 深圳市迪博企业风险管理技术有限公司 | Risk analysis method for financial activity knowledge graph of listed company |
| CN113988724A (en) * | 2021-12-28 | 2022-01-28 | 深圳市迪博企业风险管理技术有限公司 | Risk analysis method for financial activity knowledge graph of listed company |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106874261A (en) | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle | |
| Loukachevitch et al. | RuThes linguistic ontology vs. Russian wordnets | |
| Cimiano et al. | Automatic acquisition of taxonomies from text: FCA meets NLP | |
| Hofferer | Achieving business process model interoperability using metamodels and ontologies | |
| CN107609052A (en) | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle | |
| Janowicz et al. | Sim-dla: A novel semantic similarity measure for description logics reducing inter-concept to inter-instance similarity | |
| CN104050224A (en) | Combining different type coercion components for deferred type evaluation | |
| Abels et al. | Identification of common methods used for ontology integration tasks | |
| CN102982095A (en) | Noumenon automatic generating system and method thereof based on thesaurus | |
| Bhatia et al. | Ontology based framework for detecting ambiguities in software requirements specification | |
| Solskinnsbakk et al. | A hybrid approach to constructing tag hierarchies | |
| Mellace et al. | Temporal embeddings and transformer models for narrative text understanding | |
| Lyu et al. | CRUD-RAG: A comprehensive chinese benchmark for retrieval-augmented generation of large language models | |
| Rao et al. | Enhancing multi-document summarization using concepts | |
| Hakimpour et al. | Resolution of semantic heterogeneity in database schema integration using formal ontologies | |
| Weida et al. | Subsumption and recognition of heterogeneous constraint networks | |
| Kwon et al. | Deriving similarity for Semantic Web using similarity graph | |
| Punitha et al. | Partition document clustering using ontology approach | |
| Singh et al. | Neural network guided fast and efficient query-based stemming by predicting term co-occurrence statistics | |
| Narendra et al. | Designing a transactional smart assistant in Indonesian using Rasa framework | |
| Wang | Answering contextual questions based on ontologies and question templates | |
| Manjula Shenoy et al. | NN based ontology mapping | |
| Kiyoki | Managing co-reference knowledge for data integration | |
| Wang et al. | A graph-based approach for semantic similar word retrieval | |
| Wan et al. | A new semantic model for domain-ontology learning |
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 |
Application publication date: 20170620 |
|
| RJ01 | Rejection of invention patent application after publication |