Ndinoda kupa kuruzhinji chidimbu chebhuku iri richangoburitswa:
Ontological modelling yebhizinesi: nzira uye matekinoroji [Chinyorwa]: monograph / [S. V. Gorshkov, S. S. Kralin, O. I. Mushtak nevamwe; mupepeti mukuru S.V. Gorshkov]. - Ekaterinburg: Ural University Publishing House, 2019. - 234 p.: kurwara, tafura; 20 cm - Munyori. inotaridzwa nechekumashure. With. — Bibliography pakupera kwech. - ISBN 978-5-7996-2580-1: 200 makopi.
Chinangwa chekutumira chidimbu ichi paHabré chakapetwa zvina:
- Hazvigoneki kuti chero munhu angakwanisa kubata bhuku iri mumaoko ake kana asiri mutengi weanoremekedzwa ; Haisi kutengeswa.
- Kugadziriswa kwakaitwa kune zvinyorwa (hazvina kujekeswa pazasi) uye kuwedzerwa kwakaitwa izvo zvisingaenderane zvakanyanya nechimiro cheakadhindwa monograph: zvinyorwa zvemusoro (pasi pevapambi) uye hyperlink.
- ndinoda ku unganidza mibvunzo nemhinduro, kuitira kuti tizvirangarire pakubatanidza rugwaro urwu mufomu rakadzokororwa mune mamwe mabhuku api naapi.
- Vazhinji Semantic Web uye Linked Data adherents vachiri kutenda kuti denderedzwa ravo rakamanikana, zvikuru nokuti ruzhinji haasati anyatsotsanangurirwa kuti zvikuru sei kuva anoomerera Semantic Web uye Linked Data. Munyori wechidimbu, kunyangwe ari wedenderedzwa iri, haabati pfungwa iyi, asi, zvisinei, anozviona seanosungirwa kuita kumwe kuedza.
Uye saka,
Semantic Webhu
Iko kushanduka kweInternet kunogona kumiririrwa sezvizvi (kana kutaura nezve zvikamu zvaro zvakaumbwa nenzira inoratidzwa pazasi):
- Zvinyorwa paInternet. Makiyi matekinoroji - Gopher, FTP, nezvimwe.
Indaneti network yepasi rose yekutsinhana kwezviwanikwa zvemuno. - Zvinyorwa zveInternet. Makiyi matekinoroji iHTML neHTTP.
Mamiriro ezviwanikwa zvakafumurwa zvinotarisa maitiro eiyo nzira yekutapurirana. - Internet data. Makiyi matekinoroji - REST uye SIPO API, XHR, nezvimwe.
Nguva yekushandiswa kweInternet, kwete vanhu chete vanova vatengi vezviwanikwa. - Internet data. Makiyi matekinoroji akabatanidzwa Data tekinoroji.
Iyi nhanho yechina, yakafanotaurwa naBerners-Lee, musiki wekiyi tekinoroji yechipiri uye mutungamiriri weW3C, inonzi Semantic Web; Yakabatanidzwa Data tekinoroji yakagadzirirwa kuita data pawebhu kwete chete muchina-kuverengeka, asiwo "muchina-unonzwisisika."
Kubva pane zvinotevera, muverengi achanzwisisa kunyorerana pakati peakakosha pfungwa dzechikamu chechipiri nechechina:
- URLs dzakafanana neURIs,
- iyo analogue yeHTML ndeye RDF,
- HTML hyperlinks akafanana neURI zvinoitika mumagwaro eRDF.
Iyo Semantic Webhu ndeye systemic chiono chenguva yemberi yeInternet pane chaiyo yakasarudzika kana yekumanikidza maitiro, kunyangwe ichigona kutora izvi zvekupedzisira kufunga. Semuenzaniso, hunhu hwakakosha hweinonzi Webhu 2.0 inoonekwa se "zvinogadzirwa nemushandisi." Kunyanya, kurudziro yeW3C inodaidzwa kuti ifunge nezvazvo ""uye basa rakadaro se .
Iyo Semantic Web Yakafa Here?
Kana ukaramba , mamiriro ezvinhu newebhu semantic anenge akafanana necommunism munguva dzekusimudzira socialism (uye kana kuvimbika kune Ilyich's conditional orders inocherechedzwa, regai munhu wose azvisarudzire). Tsvaga injini manikidza mawebhusaiti kushandisa RDFa neJSON-LD uye ivo pachavo vanoshandisa matekinoroji ane chekuita neakatsanangurwa pazasi (Google Knowledge Graph, Bing Knowledge Graph).
Mukutaura kwakawanda, munyori haakwanisi kutaura kuti chii chiri kudzivirira kupararira kukuru, asi anogona kutaura maererano neruzivo rwega. Pane matambudziko anogona kugadziriswa "kunze kwebhokisi" mumamiriro ezvinhu eSW anogumbura, kunyange zvazvo asina kunyanya kupararira. Nekuda kweizvozvo, avo vanotarisana nemabasa aya havana nzira yekumanikidza kune avo vanokwanisa kupa mhinduro, nepo iyo yekupedzisira yekuzvimiririra yekugadziriswa kwemhinduro inopesana nemhando dzebhizinesi ravo. Saka isu tinoenderera mberi nekupatsanura HTML uye kunamatira pamwechete akasiyana API, imwe neimwe shittier.
Zvisinei, Linked Data matekinoroji akapararira kupfuura mainstream Web; Iri bhuku, chokwadi, rakatsaurirwa kune aya maapplication. Parizvino, iyo Linked Data nharaunda inotarisira kuti matekinoroji aya atonyanya kupararira nekuda kwekurekodha kwaGartner (kana kuzivisa, sezvaunoda) zvemaitiro akadai. Zivo Girafu и Data Fabric. Ndinoda kutenda kuti haizove iyo "bhasikoro" kuita kweaya mazano anozobudirira, asi ayo ane hukama neiyo W3C zviyero zvinokurukurwa pazasi.
Yakabatanidzwa Data
Berners-Lee akatsanangura Yakabatanidzwa Dhata sewebhu semantic "yakaitwa zvakanaka": seti yemaitiro uye matekinoroji anoibvumira kuzadzisa zvinangwa zvayo zvekupedzisira. Nheyo dzekutanga dzeYakabatanidzwa Data Berners-Lee zvinotevera.
Musimboti 1. Kushandisa URIs kudoma masangano.
URIs zviziviso zvepasi rose zvinopesana nezvinotaridza tambo dzemuno dzezvinyorwa. Mushure meizvozvo, musimboti uyu wakanyatsoratidzwa muGoogle Knowledge Girafu sirogani "".
Musimboti 2. Kushandisa URIs muHTTP scheme kuitira kuti irege kutaurwa.
Nekureva URI, zvinofanirwa kuve zvichigoneka kuwana zvakacherechedzwa kuseri kwechiratidzo icho (chifananidziro chine zita remushandisi " chakajeka pano).*"muC); zvakanyanya, kuwana imwe inomiririra yeiyi inomiririrwa - zvichienderana nekukosha kweiyo HTTP musoro Accept:. Zvichida, nekuuya kweiyo AR / VR nguva, zvinokwanisika kuwana iyo sosi pachayo, asi parizvino, ingangoita gwaro reRDF, inova mhedzisiro yekuita SPARQL mubvunzo. DESCRIBE.
Musimboti 3. Kushandiswa kweW3C zviyero - kunyanya RDF(S) uye SPARQL - kunyanya kana uchibvisa maURI.
Aya ega "matanho" eiyo Linked Data tekinoroji stack, inozivikanwawo se , icharondedzerwa pasi apa.
Musimboti 4. Kushandisa mareferensi kune mamwe maURI kana uchitsanangura masangano.
RDF inokutendera kuti uzvimise kune tsananguro yeshoko yechishandiso mumutauro wechisikigo, uye musimboti wechina unodaidza kuti usaite izvi. Kana nheyo yekutanga ichicherechedzwa pasi rose, zvinogoneka pakutsanangura chitubu chekutaura kune vamwe, kusanganisira "vatorwa", ndosaka data ichinzi yakabatana. Muchokwadi, zvinenge zvisingadzivisiki kushandisa URIs yakatumidzwa mumutauro weRFS.
R.F.D.
(Resource Description Framework) inzira yekutsanangura masangano ane hukama.
Zvinyorwa zve "subject-predicate-object" mhando, inonzi triplets, inoitwa pamusoro pemasangano uye hukama hwavo. Munyaya yakapfava, nyaya, predicate, uye chinhu zvese maURIs. URI imwechete inogona kunge iri munzvimbo dzakasiyana mumatatu matatu akasiyana: kuva chidzidzo, chivakashure, uye chinhu; Saka, matatu matatu anoumba rudzi rwegirafu inonzi RDF graph.
Zvidzidzo uye zvinhu zvinogona kunge zvisiri zveURI chete, asiwo zvinonzi node dzisina chinhu, uye zvinhu zvinogonawo kuva literals. Literals mienzaniso yemhando dzekare dzinosanganisira tambo inomiririra uye chiratidzo chemhando.
Mienzaniso yekunyora zvinyorwa (muTurtle syntax, zvimwe pamusoro payo pazasi): "5.0"^^xsd:float и "five"^^xsd:string. Zvinyorwa zvine mhando rdf:langString inogona zvakare kugadzirirwa netag yemutauro; muTurtle zvakanyorwa seizvi: "five"@en и "пять"@ru.
Empty node "zvisingazivikanwe" zviwanikwa pasina zviziviso zvepasi rose, nezve izvo zvirevo zvinogona, zvisinei, kuitwa; mhando dzezvinhu zviripo.
Saka (izvi ndizvo, chokwadi, iyo pfungwa yese yeRDF):
- nyaya iURI kana node isina chinhu,
- chivakashure iURI,
- chinhu iURI, node isina chinhu, kana chaiyo.
Sei zvirevo zvisingagone kuva node dzisina chinhu?
Chikonzero chingangove chishuwo chekunzwisisa zvisina kurongwa uye kududzira katatu mumutauro wekutanga-predicate logic. s p o sechinhu chakadai 
kupi - chirevo, и - constants. Tsanangudzo dzekunzwisisa uku dziri mugwaro "", iyo ine chimiro cheW3C inoshanda yeboka noti. Nekunzwisisa uku, katatu s p []kupi [] - node isina chinhu, ichashandurwa se kupi - chinja, asi sei ipapo kushandura s [] o? Gwaro rine W3C Recommendation status "” inopa imwe nzira yekushandura, asi haifunge nezve mukana wezvinoreva kuva node dzisina chinhu.
Zvisinei, Manu Sporni .
RDF imhando yeabstract. RDF inogona kunyorwa (serialized) mune akasiyana syntaxes: , (inoverengwa nevanhu vakawanda), , (binary).
Iyo imwechete RDF inogona kuisirwa muRDF/XML nenzira dzakasiyana, saka, semuenzaniso, hazvina musoro kusimbisa iyo inoguma XML uchishandisa XSD kana kuedza kubvisa data uchishandisa XPath. Saizvozvo, JSON-LD haafanire kugutsa avhareji yeJavascript yekuvandudza chishuwo chekushanda neRDF uchishandisa Javascript's dot uye square-bracket notation (kunyangwe JSON-LD inofamba nenzira iyoyo nekupa michina. ).
Mazhinji masyntaxes anopa nzira dzekupfupisa URIs refu. Somuenzaniso, ad @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> muTurtle zvinozokubvumira kuti unyore panzvimbo <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> zvakarurama rdf:type.
RDFS
(RDF Schema) - a basic modeling mazwi, anosuma pfungwa dzezvivakwa uye kirasi uye zvivakwa zvakaita se. rdf:type, rdfs:subClassOf, rdfs:domain и rdfs:range. Uchishandisa duramazwi re RDFS, semuenzaniso, mazwi anotevera anoshanda anogona kunyorwa:
rdf:type rdf:type rdf:Property .
rdf:Property rdf:type rdfs:Class .
rdfs:Class rdfs:subClassOf rdfs:Resource .
rdfs:subClassOf rdfs:domain rdfs:Class .
rdfs:domain rdfs:domain rdf:Property .
rdfs:domain rdfs:range rdfs:Class .
rdfs:label rdfs:range rdfs:Literal .RDFS tsananguro uye yekuenzanisira mazwi, asi haisi mutauro unomanikidza (kunyangwe iyo yepamutemo yakatarwa uye mukana wekushandiswa kwakadaro). Izwi rekuti "Schema" harifanirwe kunzwisiswa nenzira imwechete seyekutaura "XML Schema". Semuyenzaniso, :author rdfs:range foaf:Person zvinoreva kuti rdf:type midziyo yose :author - foaf:Person, asi hazvirevi kuti izvi zvinofanira kutaurwa pachine nguva.
SPARQL
(SPARQL Protocol uye RDF Query Mutauro) - mutauro wekubvunza RDF data. Muchiitiko chakareruka, mubvunzo weSPARQL seti yemasamples ayo anofananidzwa nematatu egirafu ari kubvunzwa. Mapeteni anogona kuve nezvinosiyana-siyana mumusoro wenyaya, prediketi, uye zvinzvimbo.
Mubvunzo unozodzosa hunhu hwakasiyana-siyana hwekuti, kana hwatsiviwa mumasamples, hunogona kuguma nechikamu cheiyo yakabvunzwa RDF graph (chikamu chematatu ayo). Mabhii ezita rimwechete mumasampuli akasiyana ematatu anofanirwa kuve nehunhu hwakafanana.
Semuyenzaniso, kupihwa seti iri pamusoro peanomwe RDFS axioms, iyo inotevera mubvunzo ichadzoka rdfs:domain и rdfs:range sezvinokosha ?s и ?p zvichiteerana:
SELECT * WHERE {
?s ?p rdfs:Class .
?p ?p rdf:Property .
}Zvakakosha kucherechedza kuti SPARQL inodudza uye hausi mutauro wekutsanangura kutenderera kwegirafu (zvisinei, mamwe marekodhi eRDF anopa nzira dzekugadzirisa hurongwa hwekuita mubvunzo). Naizvozvo, mamwe matambudziko akajairwa egirafu, semuenzaniso, kutsvaga nzira ipfupi, haigone kugadziriswa muSPARQL, kusanganisira kushandisa (asi, zvakare, ega ega RDF repositori anopa akakosha ekuwedzera kugadzirisa aya matambudziko).
SPARQL haigoverane fungidziro yekuvhurika kwenyika uye inotevera nzira ye "kuramba sekukundikana", umo magadzirirwo akadai FILTER NOT EXISTS {…}. Kugoverwa kwedata kunotariswa uchishandisa michina .
Iyo SPARQL yekuwana nzvimbo - iyo RDF yekuchengetedza inokwanisa kugadzirisa SPARQL mibvunzo - haina akananga analogues kubva padanho rechipiri (ona kwekutanga ndima iyi). Inogona kufananidzwa nedhatabhesi, zvichienderana nezviri mukati mayo mapeji eHTML akagadzirwa, asi anowanikwa kune kunze. Iyo SPARQL yekuwana nzvimbo inofanana neiyo API yekuwana nzvimbo kubva padanho rechitatu, asi ine misiyano mikuru miviri. Chekutanga, zvinokwanisika kusanganisa akati wandei "atomic" mibvunzo kuita imwe (iyo inoonekwa seyakakosha hunhu hweGraphQL), uye chechipiri, API yakadai inongozvinyora yega (ndizvo zvakaedza HATEOAS kuzadzisa).
Polemical remark
RDF inzira yekuburitsa data pawebhu, saka kuchengetedza RDF kunofanirwa kutorwa segwaro DBMS. Ichokwadi, sezvo RDF iri girafu uye kwete muti, ivo zvakare vakave vari graph-based. Zvinoshamisa kuti zvakabudirira zvachose. Ndiani aifunga kuti paizova nevanhu vakangwara vaizoshandisa ma blank nodes. Codd iri pano .
Kune zvakare nzira dzisina kuzara-dzakazara dzekuronga kuwana kune RDF data, semuenzaniso, (LDF) uye (LDP).
OWL
(Webhu Ontology Mutauro) - formalism yekumiririra ruzivo, syntactic vhezheni yekutsanangura logic (pese pese pazasi pane chokwadi kutaura OWL 2, vhezheni yekutanga yeOWL yakavakirwa pa ).
Pfungwa dzezvinotsanangudza muOWL dzinoenderana nemakirasi, mabasa anoenderana nezvivakwa, vanhu vanochengeta zita ravo rekare. Axioms anonziwo axioms.
Semuenzaniso, mune izvo zvinonzi yeOWL notation axiom inotozivikanwa kwatiri zvichanyorwa sezvizvi:
Class: Human
Class: Parent
EquivalentClass: Human and (inverse hasParent) some Human
ObjectProperty: hasParentKune mamwe masitaxesi ekunyora OWL, sekuti , yakashandiswa muchirevo chepamutemo, uye . Uyezve, OWL inogona kuiswa serieri uyezve - mune chero yeakananga syntaxes.
OWL ine hukama huviri neRDF. Kune rimwe divi, inogona kutariswa semhando yeduramazwi rinowedzera RDFS. Nekune rimwe divi, iine simba rakawanda formalism iyo RDF ingori serialization fomati. Haasi ese ekutanga OWL ekuvaka anogona kunyorwa uchishandisa imwechete RDF katatu.
Zvichienderana nekuti ndeipi subset yeOWL constructs inotenderwa kushandiswa, vanotaura nezvezvinonzi . Iwo akamisikidzwa uye ane mukurumbira ndewe OWL EL, OWL RL uye OWL QL. Sarudzo yeprofile inokanganisa computational kuomarara kwezvakajairwa matambudziko. Seti yakakwana yeOWL inovaka inoenderana ne , inonzi OWL DL. Dzimwe nguva ivo vanotaurawo nezve OWL Yakazara, umo OWL inovaka inotenderwa kushandiswa nerusununguko rwakazara rwuri muRDF, pasina semantic uye computational zvirambidzo. . Somuenzaniso, chimwe chinhu chinogona kuva zvose kirasi uye pfuma. OWL Yakazara haigoneki.
Misimboti yakakosha yekubatanidza mhedzisiro muOWL kutorwa kwefungidziro yenyika yakavhurika. ) uye kurambwa kwekufungidzira kwemazita akasarudzika (rimwe zita rekufungidzira, ) Pazasi tinoona kuti aya misimboti anogona kutungamira uye kusuma mamwe maOWL anovaka.
Rega iyo ontology ive neinotevera chidimbu (muManchester syntax):
Class: manyChildren
EquivalentTo: Human that hasChild min 3
Individual: John
Types: Human
Facts: hasChild Alice, hasChild Bob, hasChild CarolZvichatevera here kubva pane zvataurwa kuti John ane vana vakawanda? Kuramba UNA kuchamanikidza injini yekufungidzira kuti ipindure mubvunzo uyu mune zvakaipa, sezvo Alice naBob vangangove munhu mumwechete. Kuti zvinotevera zviitike, zvinodikanwa kuwedzera axiom inotevera:
DifferentIndividuals: Alice, Bob, Carol, JohnRega ikozvino chidimbu cheontology chive neinotevera fomu (John anonzi ane vana vazhinji, asi ane vana vaviri chete):
Class: manyChildren
EquivalentTo: Human that hasChild min 3
Individual: John
Types: Human, manyChildren
Facts: hasChild Alice, hasChild Bob
DifferentIndividuals: Alice, Bob, Carol, JohnKo iyi ontology ichave isingaenderane here (iyo inogona kududzirwa seumbowo hwe data risiri iro)? Kugamuchira OWA kuchaita kuti injini yekufungidzira ipindure mune zvakaipa: "kumwe" kumwe (mune imwe ontology) zvinogona kutaurwa kuti Carol mwana waJohn zvakare.
Kubvisa mukana weizvi, ngatiwedzerei chokwadi chitsva pamusoro paJohn:
Individual: John
Facts: hasChild Alice, hasChild Bob, not hasChild CarolKuti tisabatanidze chitarisiko chevamwe vana, ngatitii kuti zvinhu zvose zvinokosha zveimba "kuva nemwana" vanhu, avo tine vana chete:
ObjectProperty: hasChild
Domain: Human
Сharacteristics: Irreflexive
Class: Human
EquivalentTo: { Alice, Bill, Carol, John }Ikozvino iyo ontology ichave inopokana, iyo injini yekufungidzira isingazotadza kuzivisa. Neyokupedzisira yezvirevo zvatakaita, neimwe nzira, “tavhara” nyika, uye ona kuti mukana wokuti Johane ave mwana wake unobviswa sei.
Kubatanidza Enterprise Data
Iyo Yakabatanidzwa Data seti yemaitiro uye matekinoroji yakatanga kuitirwa kuburitsa data paWebhu. Kushandiswa kwavo munharaunda yekambani yemukati kunotarisana nematambudziko akati wandei.
Semuenzaniso, munzvimbo yakavharwa yekambani, simba rekubvisa reOWL rinobva pakugamuchirwa kweOWA uye kurambwa kweUNA, zvisarudzo nekuda kwekuvhurika uye kugoverwa kweWebhu, hazvina kusimba. Uye pano zvinotevera zvinogadzirisa zvinogoneka.
- Kupa OWL ine semantics, zvichireva kusiiwa kweOWA uye kutorwa kweUNA, kuitiswa kweinjini yekubuda inoenderana. - Munzira iyi Stardog RDF kuchengetedza.
- Kusiya kugona kweOWL yekubvisa uchifarira mainjini ekutonga. - Stardog inotsigira ; Jena uye GraphDB inopa mitemo
- Kuramba kweiyo deductive kugona kweOWL, kushandiswa kweimwe kana imwe subset padyo neRFS yekuenzanisira. - Ona zvimwe pamusoro peizvi pazasi.
Imwe nyaya ndiyo yakanyanya kutariswa iyo nyika yemubatanidzwa ingangove nayo pazvinhu zvemhando yedata uye kushomeka kwematurusi ekusimbisa data muiyo Yakabatanidzwa Data stack. Zvabuda pano ndezvizvi.
- Zvekare, shandisa kusimbiswa kweOWL inovaka ine yakavharwa semantics yenyika uye akasarudzika mazita kana yakakodzera inferensi injini iripo.
- Shandisa , yakamisikidzwa mushure mekunyorwa kweSemantic Web Layer Cake layers yakagadziriswa (zvisinei, inogona kushandiswawo semitemo injini), kana .
- Kunzwisisa kuti zvese zvinozoitwa neSPARQL mibvunzo, uchigadzira yako yakapusa data yekusimbisa nzira uchiishandisa.
Nekudaro, kunyangwe kurambwa kwakazara kwekugona kwekubvisa uye maturusi ekusimbisa kunosiya iyo Yakabatanidzwa Data stack kunze kwemakwikwi mumabasa akafanana muchimiro kune yakavhurika uye yakagoverwa webhu - mumabasa ekubatanidza data.
Zvakadini neyakajairwa bhizinesi ruzivo system?
Izvi zvinogoneka, asi iwe unofanirwa, chokwadi, kuziva chaizvo izvo matambudziko anoenderana tekinoroji achagadzirisa. Ini ndichatsanangura pano maitiro akajairika evatori vechikamu mubudiriro kuratidza kuti iyi tekinoroji stack inotaridzika sei kubva pakuona kweyakajairika IT. Ndiyeuchidze zvishoma nezvemufananidzo wenzou.
- Business analyst: RDF chimwe chinhu chakafanana neyakachengetwa yakachengetwa modhi.
- System Analyst: RDF yakafanana , chete neboka remaindekisi uye mutauro wekubvunza uri nyore.
- yokuvaka: zvakanaka, izvi zvese zviri mumweya wepfungwa dzeakapfuma modhi uye yakaderera kodhi, munguva pfupi yapfuura pamusoro peizvi.
- Project Manager: ehe zvakangofanana !
Kudzidzira kunoratidza kuti iyo stack inonyanya kushandiswa mumabasa ane chekuita nekugovera uye heterogeneity yedata, semuenzaniso, pakuvaka MDM (Master Data Management) kana DWH (Data Warehouse) makirasi masisitimu. Matambudziko akadaro aripo mune chero indasitiri.
Panyaya yeindasitiri-yakananga maapplication, Linked Data tekinoroji parizvino inonyanya kufarirwa mune anotevera maindasitiri.
- biomedical technologies (apo mukurumbira wavo unoratidzika kunge une hukama nekuoma kwenzvimbo);
current
"Boiling Point" nguva pfupi yadarika yakaita musangano wakarongwa nesangano re "National Medical Knowledge Base"".
- kugadzirwa uye mashandiro ezvigadzirwa zvakaomarara (hombe mekiniki engineering, mafuta uye gasi kugadzirwa; kazhinji isu tiri kutaura nezve standard );
current
Pano, zvakare, chikonzero ndechekuoma kwenzvimbo yenyaya, apo, semuenzaniso, padanho rekumusoro, kana tikataura nezveindasitiri yemafuta uye gasi, kuverenga kuri nyore kunoda mamwe mabasa eCAD.
Muna 2008, chiitiko chekuisa mumiriri, chakarongwa neChevron, chakaitika .
ISO 15926, pakupedzisira, yaiita seyakarema kuindasitiri yemafuta negasi (uye yakawana pamwe yakanyanya kushanda muinjiniya yemagetsi). Chete Statoil (Equinor) yakanyatsobatanidzwa pairi; muNorway, yakazara . Vamwe vari kuedza kuita zvinhu zvavo. Semuenzaniso, maererano nerunyerekupe, Ministry of Energy yepamba inotarisira kugadzira "conceptual ontological modhi yemafuta uye simba rakaoma," yakafanana, sezviri pachena, .
- masangano emari (kunyangwe XBRL inogona kutorwa semhando yehybrid yeSDMX uye RDF Data Cube ontology);
current
Mukutanga kwegore, LinkedIn yakashingairira spammed munyori nezvinzvimbo kubva kunenge hofori yese yeindasitiri yezvemari, yaanoziva kubva kuTV yakatevedzana "Force Majeure": Goldman Sachs, JPMorgan Chase uye / kana Morgan Stanley, Wells Fargo, SWIFT/Visa/Mastercard, Bank of America, Citigroup, Fed, Deutsche Bank... Pamwe munhu wese aitsvaga waaigona kutumira kwaari. . Vashoma vakakwanisa kuwana: masangano ezvemari akatora zvese .
PaHeadHunter, Sberbank chete yakawana chimwe chinhu chinonakidza; yaive nezve "EAV chengetedzo ine RDF-senge data data."
Pamwe, mutsauko muchiyero cherudo kune anowirirana matekinoroji edzimba uye dzekuMadokero masangano emari imhaka yemamiriro ekunze ezviitiko zvekupedzisira. Sezviri pachena, kubatanidzwa kuyambuka miganhu yenyika kunoda zvemhando dzakasiyana siyana dzesangano uye dzehunyanzvi mhinduro.
- mibvunzo-mhinduro masisitimu ane maapplication ekutengesa (IBM Watson, Apple Siri, Google Knowledge Graph);
current
Nenzira, musiki weSiri, Thomas Gruber, ndiye munyori weiyo tsanangudzo yeontology (mupfungwa yeIT) se "chirevo chekufungidzira." Mukuona kwangu, kurongazve mazwi mutsanangudzo iyi hakuchinji zvarinoreva, izvo zvimwe zvinoratidza kuti hazvipo.
- kuburitswa kwedata rakarongeka (nechikonzero chakakura izvi zvinogona kuverengerwa kune Yakabatanidzwa Yakavhurika Dhata).
current
Mafeni mahombe eAkabatanidzwa Data ndiwo anonzi GLAM: Galleries, Libraries, Archives, uye Museums. Zvakwana kutaura kuti Raibhurari yeCongress iri kusimudzira kutsiva MARC21 , iyo inopa hwaro hweramangwana rerondedzero yebhaibheri uye, hongu, zvichibva paRDF.
Wikidata inowanzotaurwa semuenzaniso wepurojekiti yakabudirira mumunda we Linked Open Data - mhando yemuchina-inoverengeka vhezheni yeWikipedia, izvo zvirimo, mukusiyana neDBPedia, hazvigadziriswe neimport kubva kune zvinyorwa zvinyorwa zvemabhokisi, asi zvirimo. yakagadzirwa zvakanyanya kana zvishoma nemaoko (uye inozove sosi yeruzivo kune mamwe mabhokisi e info).
Isu tinokurudzirawo kuti uzviongorore vashandisi veStardog RDF yekuchengetedza pane iyo Stardog webhusaiti muchikamu che "Vatengi".
Iva sezvazvingaite, muGartner "Enterprise Taxonomy and Ontology Management" inoiswa pakati pekudzikira mumupata wekuodzwa mwoyo netarisiro yekusvika "panzvimbo yekubudirira" kwete pamberi pemakore gumi.
Kubatanidza Enterprise Data
Mafungiro, fungidziro, fungidziro...
Nekuda kwechido chenhoroondo, ndakatara pazasi mafambiro aGartner kwemakore akasiyana pamatekinoroji anotifarira.
| Gore | Technology | Chirevo | Chinzvimbo | Makore kuenda kubani |
|---|---|---|---|---|
| 2001 | Semantic Webhu | Emerging Technologies | Innovation Trigger | 5-10 |
| 2006 | Corporate Semantic Web | Emerging Technologies | Peak of Inflated Tarisiro | 5-10 |
| 2012 | Semantic Webhu | Big Data | Peak of Inflated Tarisiro | > 10 |
| 2015 | Yakabatanidzwa Data | Advanced Analytics uye Data Sayenzi | Mugwagwa Wokuodzwa mwoyo | 5-10 |
| 2016 | Enterprise Ontology Management | Emerging Technologies | Mugwagwa Wokuodzwa mwoyo | > 10 |
| 2018 | Zivo Girafu | Emerging Technologies | Innovation Trigger | 5-10 |
Zvisinei, kare mukati imwe nzira yekumusoro yakaonekwa - Knowledge Graphs. Kumwe kuzvarwa patsva kwakaitika: graph DBMSs, uko kutarisisa kwevashandisi uye kuedza kwevagadziri kwakashanduka kuchinjirwa, pasi pesimba rezvikumbiro zveaimbova uye maitiro ekupedzisira, akatanga kutora contours uye chinzvimbo. yevakavatangira vakwikwidzi.
Inenge yega yega girafu DBMS ikozvino inozvizivisa ipuratifomu yakakodzera yekuvaka yekambani "ruzivo girafu" ("yakabatanidzwa data" dzimwe nguva inotsiviwa ne "data rakabatana"), asi zvirevo zvakadaro zvinoruramiswa sei?
Girafu dhatabhesi ichiri semantic; iyo data mugirafu DBMS ichiri yakafanana data silo. Tambo identifiers pachinzvimbo cheURIs inoita kuti basa rekubatanidza madhirafu maviri eDBMS richiri basa rekubatanidza, nepo kubatanidza zvitoro zviviri zveRDF zvinowanzouya pasi pakungobatanidza magirafu maviri eRDF. Chimwe chikamu cheasemanticity ndeye kusadzokororwa kweLPG graph modhi, izvo zvinoita kuti zviome kubata metadata uchishandisa iyo imwechete chikuva.
Chekupedzisira, girafu maDBMS haana injini dzekufungidzira kana injini dzekutonga. Mhedzisiro yeinjini dzakadaro inogona kudzokororwa nekuomesera mibvunzo, asi izvi zvinogoneka kunyangwe muSQL.
Nekudaro, inotungamira RDF yekuchengetedza masisitimu haina kunetseka kutsigira iyo LPG modhi. Nzira yakasimba kwazvo ndiyo yakatsanangurwa panguva imwe muBlazegraph: iyo RDF * modhi, inosanganisa RDF neLPG.
More
Unogona kuverenga zvakawanda nezve RDF yekuchengetedza rutsigiro rweLPG modhi mune yapfuura chinyorwa paHabré: . Ndinovimba rimwe zuva chinyorwa chakasiyana chichanyorwa nezve Ruzivo Girafu uye Data Fabric. Chikamu chekupedzisira, sezvazviri nyore kunzwisisa, chakanyorwa nekukurumidza, zvisinei, kunyange mwedzi mitanhatu gare gare, zvinhu zvose hazvina kunyatsojeka nemafungiro aya.
Literature
- Halpin, H., Monnin, A. (eds.) (2014). Philosophical Engineering: Kuenda kune Philosophy yeWebhu
- Allemang, D., Hendler, J. (2011) Semantic Web yeWorking Ontologist (2nd ed.)
- Staab, S., Studer, R. (eds.) (2009) Handbook on Ontologies (2nd ed.)
- Wood, D. (ed.). (2011) Kubatanidza Enterprise Data
- Keet, M. (2018) Nhanganyaya yeOntology Engineering
Source: www.habr.com