CN106570566A - Camellia sinensis insect pest knowledge expression and sharing method based on ontology - Google Patents
Camellia sinensis insect pest knowledge expression and sharing method based on ontology Download PDFInfo
- Publication number
- CN106570566A CN106570566A CN201610972344.8A CN201610972344A CN106570566A CN 106570566 A CN106570566 A CN 106570566A CN 201610972344 A CN201610972344 A CN 201610972344A CN 106570566 A CN106570566 A CN 106570566A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- tea
- domain
- concept
- plant pests
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
Abstract
The invention introduces ontology into a Camellia sinensis insect pest knowledge expression and sharing model, and provides the Camellia sinensis insect pest knowledge expression and sharing method based on ontology; the method comprises the following steps: firstly obtaining domain knowledge through demand analysis; then using ontology technology and OWL language to solve the domain knowledge effective expressing problem; finally realizing domain knowledge reuse and sharing through a ProtegeAPI interface. The method can realize standard formal expression of the camellia sinensis insect pest domain knowledge, thus improving camellia sinensis insect pest domain knowledge shearing reuse efficiency, providing a more effectively knowledge service platform for camellia sinensis insect pest control and tea grower production decision, and promoting tea production quality and tea industry informatization development for certain level.
Description
Technical field
The present invention is application of the computer information technology in agriculture field, mainly proposes a kind of virtual body in Tea Science field
Modeling method.
Background technology
Tea is beverage popular in the world, has the effect of life lengthening to human body.In recent years, with tea-drinking market
The raising of share, tea industry are developed rapidly, and to tea industry knowledge services requirements at the higher level are proposed.Wherein, tea-plant pests preventing and treating,
The study hotspot that even more market is paid close attention to.Tea-plant pests preventing and treating is related to multiple subject crossing necks such as Tea Science, biology, Plant Protection
Domain, includes a large amount of, complicated concept and relation in knowledge system.And effectively expressing tea-plant pests domain knowledge be diagnosis,
Preventing and treating tea-plant pests, realize basis and the premise of tea industry knowledge services.
Domain knowledge expression mainly have logical representation, production KR, frame representation, semantic notation,
XML representation, ontology representation method etc..Logical representation represents action Subjective and Objective in predicate form, is described using logical formula
Practical judgment, property, relation etc., but be difficult to realize uncertain reasoning;Production KR, i.e. IF-THEN representations, behind IF
The prerequisite of description rule, the conclusion of description rule behind THEN, be used for state various processes knowledge between control and
Its interaction mechanism, but structured knowledge beyond expression of words.
Body is the clear and definite Formal Specification explanation of shared ideas model, by concept in the field for determining common accreditation
And the relationship of the concepts, the model of Description of Knowledge on semantic and knowledge level.Body generally by concept, relation, example, axiom,
The first language composition of five modelings of function, i.e. O=(C, R, I, A, F).Wherein, C represents concept set, and R is set of relationship, and I represents real
Example, A represents axiomatic set theory, F then representative function set.
The content of the invention
It is of the invention on the basis of having studied for tea-plant pests field knowledge hierarchy in large scale, by body skill
Art introduces tea industry knowledge services, a kind of tea-plant pests representation of knowledge and sharing method based on body is proposed, from bulky complex
Tea-plant pests knowledge system in take out the domain knowledge for being easy to reuse, determine in field between the concept and concept of common accreditation
Relation, expresses tea-plant pests domain knowledge, builds domain knowledge ontology library, generates domain knowledge shared platform, realizes tea insect
The effective expression of evil domain knowledge, shared reuse, organization and administration etc., are that the such as tea-plant pests diagnosis of development tea industry knowledge services is anti-
Control, the element task of tea Culture problem decision-making.
1st, a kind of tea-plant pests representation of knowledge and sharing method based on body, is characterized in that step includes:
1) knowledge acquisition:
First, all relevant knowledges that knowledge representation is tea-plant pests field are determined by demand analyses;
Then, by multiple Knowledge Sources in tea-plant pests field, finishing field knowledge is collected, fully obtains field related
Concept and the relationship of the concepts;
Finally, body logical framework i.e. O=(C, R, I, A, F) is built based on the domain knowledge for arranging, defines knowledge table
Show class C, attribute R, example I, axiom A and the rule F of body;Representation of knowledge body with biological classification method as standard, to Camellia sinensis
Insect concept is classified;Tea-plant pests field concept and the relationship of the concepts include father's subclass relation, equivalence class relation, object category
Property, data attribute;
2) knowledge representation:
2.1) structure of class
In tea-plant pests field, by Pests of Tea-Plants, Camellia sinensis position, Cha Qu, hazard approach, the extent of injury, morphological characteristic, life
State is prevented and treated and is defined as the top layer class of body with pest natural enemy;Subclass or example are respectively provided with again;
2.2) attribute builds
Noumenon property includes object properties and data attribute;
Object properties are used to associate two examples, express the non-categorical relation between concept, with anti-reciprocity, symmetry and
Transitivity is constrained;The domain of definition of object properties determines respectively two classes for associating with codomain;Camellia sinensis defined in tea-plant pests field
Insect and Cha Qu, Camellia sinensis position, hazard approach, the extent of injury, pest natural enemy, insect and insect morphological characteristic are the class for associating,
I.e. object properties are set to liveIn (tea area), harm (Camellia sinensis position), harmMode (hazard approach), harmDegree (danger
Evil degree), eated (pest natural enemy), eat (Pests of Tea-Plants), pestMorphology (insect morphological characteristic), wherein
PestMorphology (insect morphological characteristic) attribute is provided with sub- attribute;
Data attribute describes the characteristic of example itself using basic data type;Tea-plant pests FIELD Data attribute includes
Tea area feature, generation life cycle, life habit, hazard symptoms, prevention and controls, insect picture and natural enemy morphological characteristic, it is various anti-
Control the sub- attribute that method is then set to data attribute;
2.3) example is created
Instances of ontology is the member of class, with atomicity;Create example need fully to include each subclass included into
Member, and its data attribute and object properties are respectively provided with, improve between the concrete property and example and example of each example
Objective connection;Tea-plant pests EXAMPLE OF FIELD concentrates on two concepts of Pests of Tea-Plants species and morphological characteristic;
2.4) axiom builds
Axiom rule further illustrates concept and the relationship of the concepts;Body axiom includes axiom, attribute axiom, the reality of class
Example axiom, axiom of constraint and custom rule;
The axiom of class has DisjiontClasses, SubClassOf, EquivalentClasses, and mutual exclusion is expressed respectively
Class, father and son's class, the definition of equivalence class;Three axioms of class are defined while class is built;
Attribute axiom includes that attribute definition domain, number field restriction and symmetry and transitivity are constrained;
Example axiom then refers to that instances of ontology is stated;
Axiom of constraint is divided into value constraint and constraint base, for defining the necessary and necessary and sufficient condition of class;
Value constraint includes All valuesSome valuesFor the codomain of limitation attribute;
Constraint base includes Max cardinality (≤), Min cardinality (>=) and Exact
Cardinality (=), for the number of limitation attribute value;
2.5) knowledge encoding
Body formally expresses concept and relation by modeling language, the tea-plant pests neck determined using OWL language pair
Domain knowledge is encoded, and realizes the semantic formal expression of domain knowledge, generates user-friendly tea-plant pests ontology library;
Cataloged procedure is to carry out manual coding with reference to the OWL grammatical ruless issued by World Wide Web Consortium W3C, or is utilized
The prot é g é softwares developed by Stanford University are encoded by OO form.
2.6) ontology inference
FaCT++ inference machines in by being integrated in prot é g é softwares make inferences detection to tea-plant pests domain body;
There is logic or example conflict if detecting, corrected according to prompting, and again reasoning inspection until errorless;
2.7) ontology evaluation
Ontology evaluation includes concept evaluation and the relationship of the concepts evaluation:
2. concept evaluation, including:
Conceptual integrity is evaluated, i.e., concept should as far as possible comprising the most of basic and important concepts in the field;
Concept evaluation of the accuracy, the concept defined in body is not only needed with integrity, and needs correctness;
Integrity detection and standard are carried out respectively to tea-plant pests domain body using the concept evaluation methodology based on corpus
Really property detection, measures accuracy rate and recall rate;
2. the relationship of the concepts evaluation, including:
Relation evaluation of the accuracy, that is, judge whether the classification of the relationship of the concepts is correct, meets objective fact;
Relationship consistency detect, refer to judge concept in body, assert and with the relation between other each conceptions of species, Qian Houding
Whether justice has semantic conflict;
Relation terseness detects, refers to and judge whether the relationship of the concepts duplicates in body, redundancy error;
Using based on Concept Semantic Similarity algorithm to the concordance of tea-plant pests domain body the relationship of the concepts, accurately
Property, terseness etc. are evaluated respectively;
Concept is associated by the relation between it, therefore, whether the relation between concept is evaluated correctly it is necessary to first evaluate
Whether the concept associated by the relation is correct, and this programme evaluates first the language of the concept and concept in corpus being associated with relation
Adopted similarity, if the semantic similarity exceedes threshold value, the concept associated by explanation relation is correct, so as to derive association
The relation of these concepts is also correct.
It is judged as whether tea-plant pests Domain Ontology Modeling is good according to Ontological concept evaluation and assessed in relation acquired results,
Domain knowledge indicates whether to reach original target, if realizes the shared of knowledge and reuses;
3) knowledge sharing
Domain knowledge effectively expressing is realized by tea-plant pests body, and is known domain knowledge as tea-plant pests field
Know storehouse, add Prot é g é api interfaces, realize that platform is shared and reused to the tea-plant pests domain knowledge based on ontology library;
Prot é g é API are the java class storehouses of increasing income realized for Web Ontology Languages OWL and RDF (S), and the API is provided
Loading and preservation OWL files, inquiry and operation OWL data models.Tea-plant pests body is by adding Prot é g é api interfaces, energy
It is enough preferably to provide support for tea-plant pests knowledge services, so as to realize the tea-plant pests domain knowledge for being based on ontology library share with
Reuse.
During ontology library input knowledge services, new conceptual knowledge is modified and increased to old concept,
Body logical framework is formally defined and improves, finally the knowledge to changing is encoded, and realizes tea-plant pests field sheet
Adjustment, maintenance and the upgrading of the corresponding persistence in body storehouse.Furtherly, also including tea-plant pests Acquirement of field knowledge module, tea
Tree insect pest domain knowledge expression module and tea-plant pests domain knowledge sharing module;
The step of knowledge acquisition module, comprises the steps one~tri-:
Step one:Demand analyses are carried out to market and user, the territory of knowledge representation is determined;
Step 2:By the cooperation with domain expert, obtain from multiple sources such as expert's monograph, document website, knowhow
Domain knowledge (term) is taken, the concept and the relationship of the concepts in combing domain knowledge;
Step 3:According to the domain knowledge through arranging, the logical framework that the domain knowledge represents body, i.e. O=are built
(C,R,I,A,F);Define class, example, object properties, data attribute and the axiom rule of domain knowledge;
The step of knowledge representation module, comprises the steps four~seven:
Step 4:The establishment such as domain body class, attribute, example, axiom is completed, relation, each reality of each class is built
Axiom rule in object properties, data attribute and the body of example;
Step 5:The domain knowledge determined using OWL language pair is encoded, and realizes the formal of domain knowledge semanteme
Expression;
Step 6:Body is carried out to be represented to the domain knowledge for having built based on description logic and Tableau (x) algorithms consistent
Property reasoning, check its knowledge representation with the presence or absence of logic conflict and example conflict;If there is conflict after reasoning, need according to mistake
Prompting return to step four, completes body modification and reasoning inspection again, until body is errorless;
Step 7:Evaluated based on the body expression of corpus and Arithmetic of Semantic Similarity to domain knowledge;If evaluating
As a result it is bad, return to step two, step 3 is needed, redefine domain knowledge, the logical framework of amendment body;If result is good
It is good, that is, generate domain knowledge ontology library;
The step of knowledge sharing module, comprises the steps eight~nine:
Step 8:Domain knowledge ontology library is connected by Prot é g é api interfaces with WWW, you can produce for tea grower,
The knowledge services such as Knowledge Decision-making, Semantic Web field provides the representation of knowledge based on body and shared platform;
Step 9:During the representation of knowledge and shared platform carry out knowledge services, need to collect user side in time
Feedback information, the new domain knowledge for updating and demand for services, and send to knowledge acquisition module, i.e., into step 3, update
Rare book body logical framework, realizes that domain knowledge ontology library is routinely safeguarded and upgraded.
Body is introduced the tea-plant pests representation of knowledge and Share Model by the present invention, is proposed based on the tea-plant pests knowledge of body
Represent and sharing method, obtain domain knowledge by demand analyses first, then solve field using ontology and OWL language
Knowledge effective expression ground problem, is finally based on Prot é g é api interfaces and provides the multiplexing of domain knowledge and share.The method is realized
The formal expression of tea-plant pests domain knowledge specifications, improves the shared reuse efficiency of tea-plant pests domain knowledge, is tea
Tree pest control, tea grower's production decision etc. provide more fruitful Knowledge Service Platform, the raising, tea for tea production quality
The development of production informatization has certain facilitation.
Description of the drawings
Fig. 1 Pests of Tea-Plants classification charts;
Fig. 2 Pests of Tea-Plants concept relation graphs;
The tea-plant pests representations of knowledge and shared system framework schematic diagram of the Fig. 3 based on body.
Specific embodiment
Below in conjunction with the accompanying drawings the present invention is further described with specific embodiment.
1 knowledge acquisition
First, the territory of knowledge representation is determined by demand analyses such as market surveys.The present invention is tea-plant pests
All relevant knowledges in field.
Secondly, by multiple sources such as the expert in the field, monograph, document, website, knowhow, finishing field is collected
Knowledge, fully obtains the related concept in field and the relationship of the concepts.The present invention by with Folium Camelliae sinensis biochemistry and biotechnology state
The cooperation of key lab of family cultivation base, to vocabulary, term in tea-plant pests field etc. combing is carried out, and is determined common in field
The concept of accreditation, and classification and non-categorical relation between concept is arranged into clear, complete the arrangement of tea-plant pests domain knowledge.Such as tea
Tree insect pest field includes the concepts such as Homoptera, green plant bug, larva, wing expanse, cultural control, south China Cha Qu, bud-leaf and Miridae-green
The relations such as fleahopper, false eye leafhopper-suction juice.
Finally, its body logical framework is built based on the domain knowledge for arranging, i.e. O=(C, R, I, A, F) defines knowledge
Represent class, attribute, example, axiom, rule of body etc..
Representation of knowledge body is classified, as shown in Figure 1 with biological classification method as standard to Pests of Tea-Plants concept.Tea
Tree insect pest field concept and the relationship of the concepts, including is-a, instance-of, attribute-of etc., as shown in Figure 2.With green
As a example by fleahopper, father's subclass relation:Arthropoda Insecta Semiptera Miridae;Equivalence class relation:Scientific name, another name;Object belongs to
Property:Life area, morphological characteristic, hazard approach, the extent of injury, hazard symptoms;Data attribute:Generation life cycle, life habit,
Environmental factorss, prevention and controls, natural enemy etc..
2 knowledge representations
(1) structure of class
Body represents that some have common denominator group of individuals using class, and by father and son's class, fraternal class, equivalence
The definition of the class relations such as class, mutual exclusion class, improves concept classification relation, it is ensured that the integrity and accuracy of field concept.
Tea-plant pests field is by Pests of Tea-Plants, Camellia sinensis position, Cha Qu, hazard approach, the extent of injury, morphological characteristic, ecology
Preventing and treating, pest natural enemy etc. are defined as top layer class, then are respectively provided with subclass or example, and such as Miridae and green plant bug are father and son's class, blind
Pentatomiddae is fraternal class with Scutelleridae, and green plant bug is equivalence class with floral leaf worm, and Miridae is mutual exclusion class etc. with Scutelleridae.
(2) attribute builds
Noumenon property includes object properties and data attribute.Object properties are used to associate two examples, between expression concept
Non-categorical relation, with the constraint such as anti-reciprocity, symmetry, transitivity, its domain of definition determines respectively two for associating with codomain
Class.Pests of Tea-Plants and Cha Qu, Camellia sinensis position, hazard approach, the extent of injury, pest natural enemy, form defined in tea-plant pests field
Feature etc. for association class, i.e., object properties be set to liveIn, harm, harmMode, harmDegree, eated, eat,
PestMorphology etc., wherein pestMorphology attributes are provided with sub- attribute.
Data attribute describes the characteristic of example itself using basic data type.Tea-plant pests FIELD Data attribute includes
Tea area feature, generation life cycle, life habit, hazard symptoms, prevention and controls, insect picture, natural enemy morphological characteristic etc., it is various anti-
Control the sub- attribute that method is then set to data attribute.
(3) example is created
Instances of ontology is the member of class, with atomicity.Create example need fully to include each subclass included into
Member, and its data attribute and object properties are respectively provided with, improve between the concrete property and example and example of each example
Objective connection.Tea-plant pests EXAMPLE OF FIELD is had focused largely among two concepts of Pests of Tea-Plants and morphological characteristic, is such as defined green blind
Stinkbug, tea geometrid etc. for Pests of Tea-Plants class example, yellow, ellipse etc. for morphological characteristic class example.
(4) axiom builds
Axiom rule further illustrates concept and the relationship of the concepts, makes body logic tighter, contributes to inconsistent
The detection of property and knowledge reasoning.Body axiom includes axiom, attribute axiom, example axiom, axiom of constraint, the custom rule of class
Etc. form.
Three big axioms DisjiontClasses, SubClassOf of class, EquivalentClasses, express respectively mutual exclusion
Class, father and son's class, the definition of equivalence class, the classification relation between perfect concept.Three grand dukes of class are defined while class is built
Reason, such as the mutual exclusion class of green plant bug are tea angle fleahopper, and equivalence class is floral leaf worm, and parent is Miridae etc..
Attribute axiom includes the constraint such as attribute definition domain, number field restriction and symmetry, transitivity, such as tea-plant pests body
Attribute eat domain of definition is pest natural enemy, codomain is Pests of Tea-Plants, and attribute eated domain of definition is Pests of Tea-Plants, codomain is insect day
Enemy, and with eat each other against attribute.
Example axiom then refers to that instances of ontology is stated, such as same instance, different instances constrained.
Axiom of constraint is divided into value constraint and constraint base, is used mainly to define the necessary and necessary and sufficient condition of class.Value constraint bag
Include All valuesSome valuesFor the codomain of limitation attribute.Such as harm of Miridae insect
Mode only inhales juice, therefore it is only constraints to add All values from, if the adjective restriction of attribute is some or deposits
Then adding some constraints.Constraint base include Max cardinality (≤), Min cardinality (>=) and
Exact cardinality (=), for the number of limitation attribute value.
Body supplements the deficiency of axiomatic specification by custom rule, expands the ability and scope of body expression knowledge, such as
SWRL rules etc..SWRL (Semantic Web Rule Language) rule is presented by systematic fashion, with OWL sublanguages
Based on OWL DL and OWL Lite, with reference to the regular describing mode of Unary/Binay Datalog Rule ML, OWL is supplemented
Deficiency of the language in terms of rule description and reasoning, there is provided more powerful logical expression ability.
(5) knowledge encoding
Body formally expresses concept and relation, such as SHOE, CycL, RDF, RDF-S, OWL by modeling language.
SHOE is an extension based on HTML.CycL is the Knowledge Description Language of Cyc systems.RDF is a kind of mark of description web resource
Note language, for processing metadata.OWL be based on RDF and RDF-S, using based on XML RDF syntax gauges, definition and
Write a kind of labelling language of Semantic Web body.
Body OWL language has clearly grammer system, abundant semantic meaning representation ability, effective computability, can
Explicit, formal description is carried out to the concept and the relationship of the concepts in field, and carries out rationally consistent reasoning.Therefore,
The present invention selects the tea-plant pests domain knowledge that OWL language pair determines to encode, and realizes the formal of domain knowledge semanteme
Expression, generates user-friendly tea-plant pests ontology library.
(6) ontology inference
Based on manual modeling, workload is big, process is complicated for domain body, it is therefore desirable to which ontology inference is patrolled checking knowledge
The conflict such as discordance of framework and Formal Representation is collected, and the information that wherein contain is inferred based on the domain knowledge for providing.
OWL bodies are based on description logic theory form, and employing (optimization) Tableaux algorithms complete logic detection and sheet
Body reasoning.The inference machine for being applied to OWL language ontologies is more, and the present invention selects FaCT++ inference machines to tea-plant pests domain body
Make inferences detection.There is logic or example conflict if detecting, need to be corrected according to prompting, and reasoning inspection again
Until errorless.The reasoning results show that tea-plant pests domain body structure is consistent, without logic conflict and example conflict.
(7) ontology evaluation
Ontology evaluation is concentrated mainly on concept evaluation and the relationship of the concepts evaluates two aspects.
1. concept evaluation:Should be as far as possible basic comprising the great majority in the field including conceptual integrity evaluation, i.e. concept
With important concept;Concept evaluation of the accuracy, the concept defined in body is not only needed with integrity, and is needed correct
Property.Integrity detection and accuracy inspection are carried out respectively to tea-plant pests domain body using based on the concept evaluation methodology of corpus
Survey, measure accuracy rate and recall rate as shown in table 1, table 2:
The accuracy rate that the tea-plant pests Ontological concept of table 1 is evaluated
The recall rate that the tea-plant pests Ontological concept of table 2 is evaluated
2. the relationship of the concepts evaluation:Whether the classification of inclusion relation evaluation of the accuracy, i.e. the relationship of the concepts is correct, meets visitor
See true;Relationship consistency detect, refer to concept in body, assert and with the relation between other each conceptions of species, in front and back definition be
It is no to have semantic conflict;The detection of relation terseness, refers to whether the relationship of the concepts duplicates in body, redundancy error.Using being based on
Concept Semantic Similarity algorithm enters respectively to concordance, accuracy, terseness of tea-plant pests domain body the relationship of the concepts etc.
Row is evaluated, and its result is as shown in table 3:
The tea-plant pests ontological relationship of table 3 evaluates catalog
Wherein, relationship consistency testing result is shown as 0 mistake, and relation accuracy testing result is shown as 7 mistakes,
Relation terseness testing result be 4 mistakes, and to point out mistake be corrected.According to Ontological concept evaluation and relation
Evaluating acquired results carries out comprehensive analysis, can determine whether that tea-plant pests Domain Ontology Modeling is good, domain knowledge represent reach it is original
Target, it is possible to achieve the shared and reuse of knowledge.
3 knowledge sharings
The tea-plant pests field ontology library for building at present amounts to 560 examples of attribute 960 of class 56, including domestic normal
Relevant knowledge in the Pests of Tea-Plants seen, and tea-plant pests field.Can be with graphic software platform, intuitively by OntoGraf plug-in units
Ground expression tea-plant pests field concept and the relationship of the concepts., respectively show tea-plant pests body general frame, tea-plant pests sheet
Body Lepidoptera framework, tea-plant pests body tea geometrid framework.
Domain knowledge effectively expressing is realized by tea-plant pests body, and as tea-plant pests domain knowledge base,
Addition Prot é g é api interfaces, realize that the tea-plant pests domain knowledge based on ontology library is shared and reuses platform, be tea grower's production,
The numerous areas such as scientific research, Semantic Web, knowledge engineering provide extensive knowledge services.
During ontology library input knowledge services, the feedback information for constantly noting each user side, Yi Jigeng are needed
New conceptual knowledge and demand for services, constantly collect to knowledge acquisition module, by analyze existing knowledge be need modification or
It is to delete, then according to the knowledge and demand for updating, new conceptual knowledge, formalization is modified and increased to old concept
Ground is defined and improves body logical framework, and finally the knowledge to changing is encoded, and realizes tea-plant pests field ontology library phase
Adjustment, maintenance and the upgrading of the persistence answered, is that inquiry diagnosis, decision support, the shared multiplexing in follow-up knowledge based storehouse etc. are carried
For more preferable tea industry knowledge services.
It is that checking is based on the tea-plant pests representation of knowledge of body and the feasibility of sharing method and effectiveness, the present invention is adopted
Protege platforms, Mysql data bases and Jsp/Servlet technologies, develop the tea-plant pests representation of knowledge based on body with
Shared system, the development environment for being used is Windows 7.The system both can carry out the formal of semanteme to domain knowledge
Expression, it is also possible to the shared of knowledge is provided the user on client service platform and service is reused.
The tea-plant pests representation of knowledge based on body is with shared system mainly by tea-plant pests Acquirement of field knowledge module, tea
The tree insect pest domain knowledge expression composition such as module and tea-plant pests domain knowledge sharing module, its basic frame structure such as Figure 83
It is shown:
The system concrete steps flow process is as follows:
Step one:Demand analyses are carried out to market and user, the territory of knowledge representation is determined.
Step 2:By the cooperation with domain expert, obtain from multiple sources such as expert's monograph, document website, knowhow
Domain knowledge (term) is taken, the concept and the relationship of the concepts in combing domain knowledge.
Step 3:According to the domain knowledge through arranging, the logical framework that the domain knowledge represents body, i.e. O=are built
(C,R,I,A,F).Define class, example, object properties, data attribute and the axiom rule of domain knowledge.
Step 4:The establishment such as domain body class, attribute, example, axiom is completed, relation, each reality of each class is built
Axiom rule in object properties, data attribute and the body of example etc..
Step 5:The domain knowledge determined using OWL language pair is encoded, and realizes the formal of domain knowledge semanteme
Expression.
Step 6:Body is carried out to be represented to the domain knowledge for having built based on description logic and Tableau (x) algorithms consistent
Property reasoning, check its knowledge representation with the presence or absence of logic conflict and example conflict.If there is conflict after reasoning, need according to mistake
Prompting return to step four, completes body modification and reasoning inspection again, until body is errorless.
Step 7:Evaluated based on the body expression of corpus and Arithmetic of Semantic Similarity to domain knowledge.If evaluating
As a result it is bad, return to step two, step 3 is needed, redefine domain knowledge, the logical framework of amendment body;If result is good
It is good, that is, generate domain knowledge ontology library.
Step 8:Domain knowledge ontology library is connected by Prot é g é api interfaces with WWW, you can produce for tea grower,
The knowledge services such as Knowledge Decision-making, Semantic Web field provides the representation of knowledge based on body and shared platform.
Step 9:During the representation of knowledge and shared platform carry out knowledge services, need to collect user side in time
Feedback information, the new domain knowledge for updating and demand for services, and send to knowledge acquisition module, i.e., into step 3, update
Rare book body logical framework, realizes that domain knowledge ontology library is routinely safeguarded and upgraded.
By the test analysis to system, good effect is achieved, demonstrate the feasibility and effectiveness of the method.
Body is introduced the tea-plant pests representation of knowledge and Share Model by the present invention, is proposed based on the tea-plant pests knowledge of body
Represent and sharing method, obtain domain knowledge by demand analyses first, then solve field using ontology and OWL language
Knowledge effective expression ground problem, is finally based on Prot é g é api interfaces and provides the multiplexing of domain knowledge and share.The method is realized
The formal expression of tea-plant pests domain knowledge specifications, improves the shared reuse efficiency of tea-plant pests domain knowledge, is tea
Tree pest control, tea grower's production decision etc. provide more fruitful Knowledge Service Platform, the raising, tea for tea production quality
The development of production informatization has certain facilitation.
Claims (2)
1. a kind of tea-plant pests representation of knowledge and sharing method based on body, is characterized in that step includes:
1) knowledge acquisition:
First, all relevant knowledges that knowledge representation is tea-plant pests field are determined by demand analyses;
Then, by multiple Knowledge Sources in tea-plant pests field, finishing field knowledge is collected, fully obtains the general of field correlation
Read and the relationship of the concepts;
Finally, body logical framework i.e. O=(C, R, I, A, F) is built based on the domain knowledge for arranging, defines representation of knowledge sheet
Class C of body, attribute R, example I, axiom A and rule F;Representation of knowledge body with biological classification method as standard, to Pests of Tea-Plants
Concept is classified;Tea-plant pests field concept and the relationship of the concepts include father's subclass relation, equivalence class relation, object properties,
Data attribute;
2) knowledge representation:
2.1) structure of class
It is in tea-plant pests field, Pests of Tea-Plants, Camellia sinensis position, Cha Qu, hazard approach, the extent of injury, morphological characteristic, ecology is anti-
Control the top layer class that body is defined as with pest natural enemy;Subclass or example are respectively provided with again;
2.2) attribute builds
Noumenon property includes object properties and data attribute;
Object properties are used to associate two examples, the non-categorical relation between concept are expressed, with anti-reciprocity, symmetry and transmission
Property constraint;The domain of definition of object properties determines respectively two classes for associating with codomain;Pests of Tea-Plants defined in tea-plant pests field
Be the class that associates with tea area, Camellia sinensis position, hazard approach, the extent of injury, pest natural enemy, insect and insect morphological characteristic, i.e., it is right
As attribute is set to liveIn (tea area), harm (Camellia sinensis position), harmMode (hazard approach), harmDegree (harm journeys
Degree), eated (pest natural enemy), eat (Pests of Tea-Plants), pestMorphology (insect morphological characteristic), wherein
PestMorphology (insect morphological characteristic) attribute is provided with sub- attribute;
Data attribute describes the characteristic of example itself using basic data type;Tea-plant pests FIELD Data attribute includes tea area
Feature, generation life cycle, life habit, hazard symptoms, prevention and controls, insect picture and natural enemy morphological characteristic, various preventing and treating sides
Rule is set to the sub- attribute of data attribute;
2.3) example is created
Instances of ontology is the member of class, with atomicity;Creating example needs fully to include the member that each subclass is included, and
Its data attribute and object properties are respectively provided with, improve objective between the concrete property and example and example of each example
System;Tea-plant pests EXAMPLE OF FIELD concentrates on two concepts of Pests of Tea-Plants species and morphological characteristic;
2.4) axiom builds
Axiom rule further illustrates concept and the relationship of the concepts;Body axiom includes axiom, attribute axiom, the example public affairs of class
Reason, axiom of constraint and custom rule;
The axiom of class has DisjiontClasses, SubClassOf, EquivalentClasses, and mutual exclusion class, father are expressed respectively
The definition of subclass, equivalence class;Three axioms of class are defined while class is built;
Attribute axiom includes that attribute definition domain, number field restriction and symmetry and transitivity are constrained;
Example axiom then refers to that instances of ontology is stated;
Axiom of constraint is divided into value constraint and constraint base, for defining the necessary and necessary and sufficient condition of class;
Value constraint includes All valuesSome valuesFor the codomain of limitation attribute;
Constraint base includes Max cardinality (≤), Min cardinality (>=) and Exact cardinality
(=), for the number of limitation attribute value;
2.5) knowledge encoding
Body formally expresses concept and relation by modeling language, and the tea-plant pests field determined using OWL language pair is known
Knowledge is encoded, and realizes the semantic formal expression of domain knowledge, generates user-friendly tea-plant pests ontology library;
Cataloged procedure is to carry out manual coding with reference to the OWL grammatical ruless issued by World Wide Web Consortium W3C, or using by this
The prot é g é softwares of Tan Fu universities exploitation are encoded by OO form.
2.6) ontology inference
FaCT++ inference machines in by being integrated in prot é g é softwares make inferences detection to tea-plant pests domain body;If inspection
Measure and there is logic or example conflict, then corrected according to prompting, and again reasoning inspection until errorless;
2.7) ontology evaluation
Ontology evaluation includes concept evaluation and the relationship of the concepts evaluation:
1. concept evaluation, including:
Conceptual integrity is evaluated, i.e., concept should as far as possible comprising the most of basic and important concepts in the field;
Concept evaluation of the accuracy, the concept defined in body is not only needed with integrity, and needs correctness;
Integrity detection and accuracy are carried out respectively to tea-plant pests domain body using the concept evaluation methodology based on corpus
Detection, measures accuracy rate and recall rate;
2. the relationship of the concepts evaluation, including:
Relation evaluation of the accuracy, that is, judge whether the classification of the relationship of the concepts is correct, meets objective fact;
Relationship consistency detect, refer to judge concept in body, assert and with the relation between other each conceptions of species, in front and back define be
It is no to have semantic conflict;
Relation terseness detects, refers to and judge whether the relationship of the concepts duplicates in body, redundancy error;
Using based on concordance, accuracy, letter of the Concept Semantic Similarity algorithm to tea-plant pests domain body the relationship of the concepts
Clean property etc. is evaluated respectively;
Evaluation methodology is:The semantic similarity of the concept and concept in corpus being associated with relation is evaluated first, if the semanteme
Similarity exceedes threshold value, then the concept associated by explanation relation is correct, so as to derive the relation for associating these concepts
It is correct;If the semantic similarity is less than threshold value, the concept associated by explanation relation is not correct, so as to derive
The relation of these concepts is associated nor correct;
It is judged as whether tea-plant pests Domain Ontology Modeling is good according to Ontological concept evaluation and assessed in relation acquired results, field
Whether the representation of knowledge reaches original target, if realizes the shared of knowledge and reuses;
3) knowledge sharing
Domain knowledge effectively expressing is realized by tea-plant pests body, and using domain knowledge as tea-plant pests domain knowledge
Storehouse, adds Prot é g é api interfaces, realizes that platform is shared and reused to the tea-plant pests domain knowledge based on ontology library;
During ontology library input knowledge services, new conceptual knowledge, form are modified and increased to old concept
Change ground definition and improve body logical framework, finally the knowledge to changing is encoded, and realizes tea-plant pests field ontology library
Adjustment, maintenance and the upgrading of corresponding persistence.
2. a kind of tea-plant pests representation of knowledge and shared system based on body, is characterized in that including that tea-plant pests domain knowledge is obtained
Delivery block, tea-plant pests domain knowledge expression module and tea-plant pests domain knowledge sharing module;
The step of knowledge acquisition module, comprises the steps one~tri-:
Step one:Demand analyses are carried out to market and user, the territory of knowledge representation is determined;
Step 2:By the cooperation with domain expert, from multiple sources such as expert's monograph, document website, knowhow neck is obtained
Domain knowledge (term), the concept and the relationship of the concepts in combing domain knowledge;
Step 3:According to the domain knowledge through arranging, build the logical framework that the domain knowledge represents body, i.e. O=(C, R,
I,A,F);Define class, example, object properties, data attribute and the axiom rule of domain knowledge;
The step of knowledge representation module, comprises the steps four~seven:
Step 4:Complete the establishment such as domain body class, attribute, example, axiom, build the relation of each class, each example
Axiom rule in object properties, data attribute and body;
Step 5:The domain knowledge determined using OWL language pair is encoded, and realizes the semantic formal table of domain knowledge
Reach;
Step 6:Represent that the domain knowledge for having built body carries out concordance and pushes away based on description logic and Tableau (x) algorithms
Reason, checks its knowledge representation with the presence or absence of logic conflict and example conflict;If there is conflict after reasoning, need according to miscue
Return to step four, completes body modification and reasoning inspection again, until body is errorless;
Step 7:Evaluated based on the body expression of corpus and Arithmetic of Semantic Similarity to domain knowledge;If evaluation result
It is bad, return to step two, step 3 is needed, redefine domain knowledge, the logical framework of amendment body;If result is good, i.e.,
Generate domain knowledge ontology library;
The step of knowledge sharing module, comprises the steps eight~nine:
Step 8:Domain knowledge ontology library is connected by Prot é g é api interfaces with WWW, you can for tea grower's production, knowledge
The knowledge services such as decision-making, Semantic Web field provides the representation of knowledge based on body and shared platform;
Step 9:During the representation of knowledge and shared platform carry out knowledge services, the feedback for collecting user side in time is needed
Information, the new domain knowledge for updating and demand for services, and send to knowledge acquisition module, i.e., into step 3, renolation sheet
Body logical framework, realizes that domain knowledge ontology library is routinely safeguarded and upgraded.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610972344.8A CN106570566B (en) | 2016-10-28 | 2016-10-28 | A kind of tea-plant pests representation of knowledge and sharing method based on ontology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610972344.8A CN106570566B (en) | 2016-10-28 | 2016-10-28 | A kind of tea-plant pests representation of knowledge and sharing method based on ontology |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106570566A true CN106570566A (en) | 2017-04-19 |
CN106570566B CN106570566B (en) | 2019-07-05 |
Family
ID=58539985
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610972344.8A Active CN106570566B (en) | 2016-10-28 | 2016-10-28 | A kind of tea-plant pests representation of knowledge and sharing method based on ontology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106570566B (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107133671A (en) * | 2017-05-26 | 2017-09-05 | 西南大学 | Citrus knowledge Modeling and extensive Ontology learning method based on agriculture Eight Terms Guidelines |
DE102017208143A1 (en) * | 2017-05-15 | 2018-11-15 | Siemens Aktiengesellschaft | Method for computer-assisted user assistance in the creation of a program for analyzing data of at least one technical system |
CN108959267A (en) * | 2018-07-18 | 2018-12-07 | 浙江大学 | A kind of expression of product function semantic modeling and application based on ontology |
CN109086347A (en) * | 2018-07-13 | 2018-12-25 | 武汉尼维智能科技有限公司 | A kind of construction method, device and the storage medium of international ocean shipping dangerous cargo knowledge mapping system |
CN109325070A (en) * | 2018-09-21 | 2019-02-12 | 中国科学院重庆绿色智能技术研究院 | A kind of culinary art knowledge architecture method and system based on ontology |
CN109933794A (en) * | 2019-03-18 | 2019-06-25 | 中科院合肥技术创新工程院 | A kind of Decision Ontology modeling method based on ontology OWL technology |
CN110866332A (en) * | 2019-10-29 | 2020-03-06 | 中国电子科技集团公司第三十八研究所 | Complex cable assembly assembling method and system |
CN113487032A (en) * | 2021-07-15 | 2021-10-08 | 上海海洋大学 | Aquatic product HACCP knowledge reasoning system and method |
WO2022095121A1 (en) * | 2020-11-05 | 2022-05-12 | 广东工业大学 | Kwon-glazed porcelain knowledge base construction method based on semantic ontology |
CN114626382A (en) * | 2022-05-12 | 2022-06-14 | 南京航空航天大学 | Automatic disease identification method for downward-facing water pipeline |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101067864A (en) * | 2007-06-13 | 2007-11-07 | 北京农业信息技术研究中心 | Long-distance controlling apparatus using for crops production expert diagnosing system and method thereof |
CN102831316A (en) * | 2012-08-24 | 2012-12-19 | 安徽农业大学 | Intelligent tea tree insect pest diagnosis prototype system based on cloud ontology |
CN106022948A (en) * | 2016-07-20 | 2016-10-12 | 安徽朗坤物联网有限公司 | Comprehensive service system of agricultural internet of things |
-
2016
- 2016-10-28 CN CN201610972344.8A patent/CN106570566B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101067864A (en) * | 2007-06-13 | 2007-11-07 | 北京农业信息技术研究中心 | Long-distance controlling apparatus using for crops production expert diagnosing system and method thereof |
CN102831316A (en) * | 2012-08-24 | 2012-12-19 | 安徽农业大学 | Intelligent tea tree insect pest diagnosis prototype system based on cloud ontology |
CN106022948A (en) * | 2016-07-20 | 2016-10-12 | 安徽朗坤物联网有限公司 | Comprehensive service system of agricultural internet of things |
Non-Patent Citations (1)
Title |
---|
耿凡凡 等: "描述逻辑在农业本体中的应用研究", 《重庆科技学院学报(》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102017208143A1 (en) * | 2017-05-15 | 2018-11-15 | Siemens Aktiengesellschaft | Method for computer-assisted user assistance in the creation of a program for analyzing data of at least one technical system |
CN107133671A (en) * | 2017-05-26 | 2017-09-05 | 西南大学 | Citrus knowledge Modeling and extensive Ontology learning method based on agriculture Eight Terms Guidelines |
CN109086347A (en) * | 2018-07-13 | 2018-12-25 | 武汉尼维智能科技有限公司 | A kind of construction method, device and the storage medium of international ocean shipping dangerous cargo knowledge mapping system |
CN108959267A (en) * | 2018-07-18 | 2018-12-07 | 浙江大学 | A kind of expression of product function semantic modeling and application based on ontology |
CN109325070A (en) * | 2018-09-21 | 2019-02-12 | 中国科学院重庆绿色智能技术研究院 | A kind of culinary art knowledge architecture method and system based on ontology |
CN109933794A (en) * | 2019-03-18 | 2019-06-25 | 中科院合肥技术创新工程院 | A kind of Decision Ontology modeling method based on ontology OWL technology |
CN110866332A (en) * | 2019-10-29 | 2020-03-06 | 中国电子科技集团公司第三十八研究所 | Complex cable assembly assembling method and system |
WO2022095121A1 (en) * | 2020-11-05 | 2022-05-12 | 广东工业大学 | Kwon-glazed porcelain knowledge base construction method based on semantic ontology |
CN113487032A (en) * | 2021-07-15 | 2021-10-08 | 上海海洋大学 | Aquatic product HACCP knowledge reasoning system and method |
CN114626382A (en) * | 2022-05-12 | 2022-06-14 | 南京航空航天大学 | Automatic disease identification method for downward-facing water pipeline |
CN114626382B (en) * | 2022-05-12 | 2022-10-11 | 南京航空航天大学 | Automatic disease identification method for downward-facing water pipeline |
Also Published As
Publication number | Publication date |
---|---|
CN106570566B (en) | 2019-07-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106570566B (en) | A kind of tea-plant pests representation of knowledge and sharing method based on ontology | |
Wang et al. | LogEvent2vec: LogEvent-to-vector based anomaly detection for large-scale logs in internet of things | |
US11615331B2 (en) | Explainable artificial intelligence | |
Muralidharan et al. | Sustainability, transformational leadership, and social entrepreneurship | |
CN101777042B (en) | Neural network and tag library-based statement similarity algorithm | |
Jabbar et al. | A methodology of real-time data fusion for localized big data analytics | |
da Silva et al. | Linguistic patterns and linguistic styles for requirements specification: Focus on data entities | |
Aqra et al. | Incremental algorithm for association rule mining under dynamic threshold | |
Rodríguez-García et al. | Knowledge-based system for crop pests and diseases recognition | |
Nnamoko et al. | Solid waste image classification using deep convolutional neural network | |
Sagar et al. | Comparing commit messages and source code metrics for the prediction refactoring activities | |
Keller et al. | Ten (mostly) simple rules to future‐proof trait data in ecological and evolutionary sciences | |
Fekete et al. | A survey of ledger technology-based databases | |
Aljofey et al. | A feature-based robust method for abnormal contracts detection in ethereum blockchain | |
Chen et al. | Logls: Research on system log anomaly detection method based on dual lstm | |
Lin et al. | VulEye: a novel graph neural network vulnerability detection approach for PHP application | |
Wu et al. | Smart contract vulnerability detection based on hybrid attention mechanism model | |
Park et al. | Code visualization for performance improvement of Java code for controlling smart traffic system in the smart city | |
Molnár et al. | The application of directed hyper-graphs for analysis of models of information systems | |
Ponnambalam et al. | Analysing the barriers involved in recycling the textile waste in India using fuzzy DEMATEL | |
Danciu et al. | Coverage fulfillment automation in hardware functional verification using genetic algorithms | |
Rasool | Customizable feature based design pattern recognition integrating multiple techniques | |
Zhu et al. | Integrating land-cover products based on ontologies and local accuracy | |
Zhao et al. | Organizational geosocial network: A graph machine learning approach integrating geographic and public policy information for studying the development of social organizations in China | |
Zhou et al. | A New Text-Mining–Bayesian Network Approach for Identifying Chemical Safety Risk Factors |
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 |