{"id":137,"date":"2020-09-03T16:40:09","date_gmt":"2020-09-03T13:40:09","guid":{"rendered":"http:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-30-ogrenme-analitiginde-bagli-veriler-olasiliklar-ve-zorluklar\/"},"modified":"2020-09-03T16:40:09","modified_gmt":"2020-09-03T13:40:09","slug":"bolum-30-ogrenme-analitiginde-bagli-veriler-olasiliklar-ve-zorluklar","status":"publish","type":"chapter","link":"https:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-30-ogrenme-analitiginde-bagli-veriler-olasiliklar-ve-zorluklar\/","title":{"raw":"B\u00f6l\u00fcm 30 \u00d6\u011frenme Analiti\u011finde Ba\u011fl\u0131 Veriler: Olas\u0131l\u0131klar ve Zorluklar","rendered":"B\u00f6l\u00fcm 30 \u00d6\u011frenme Analiti\u011finde Ba\u011fl\u0131 Veriler: Olas\u0131l\u0131klar ve Zorluklar"},"content":{"raw":"\n

Amal Zouaq1<\/sup>, Jelena Jovanovi\u01072<\/sup>, Sre\u0107ko Joksimovi\u01073<\/sup>, Dragan Ga\u0161evi\u01074<\/sup><\/span><\/span><\/p>\n1<\/sup>Elektrik M\u00fchendisli\u011fi ve Bilgisayar Bilimleri Fak\u00fcltesi, Ottawa \u00dcniversitesi, Kanada<\/span><\/span>\n\n2<\/sup> Yaz\u0131l\u0131m M\u00fchendisli\u011fi B\u00f6l\u00fcm\u00fc, Belgrad \u00dcniversitesi, S\u0131rbistan<\/span><\/span>\n\n3<\/sup>Moray House E\u011fitim Okulu, Edinburgh \u00dcniversitesi, Birle\u015fik Krall\u0131k<\/span><\/span>\n\n4<\/sup> Enformatik Okulu, Edinburgh \u00dcniversitesi, Birle\u015fik Krall\u0131k<\/span><\/span>\n\nDOI: 10.18608\/hla17.030<\/span><\/span>\n

\u00d6Z<\/h2>\n\u00d6\u011frenme analiti\u011fi (\u00d6A); \u00f6\u011frenme ortamlar\u0131n\u0131n \u00e7e\u015fitlili\u011fi ve farkl\u0131l\u0131\u011f\u0131, kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i dersler (KA\u00c7D) gibi \u00f6l\u00e7eklenebilir \u00f6\u011frenme modellerinin ortaya \u00e7\u0131kmas\u0131 ve sosyal medya platformlar\u0131n\u0131n \u00f6\u011frenme s\u00fcre\u00e7leriyle b\u00fct\u00fcnle\u015ftirilmesi nedeniyle veri \u00fcretiminde art\u0131\u015fa tan\u0131kl\u0131k eder. Bu \u00e7e\u015fitlilik, \u00f6\u011frenme platformlar\u0131n\u0131n birlikte \u00e7al\u0131\u015fabilirli\u011fini, \u00e7oklu bilgi kaynaklar\u0131ndan gelen heterojen verilerin b\u00fct\u00fcnle\u015ftirilmesini ve \u00f6\u011frenme kaynaklar\u0131 ile \u00f6\u011frenme izlerinin i\u00e7erik analizini ilgilendiren bir\u00e7ok sorunu olu\u015fturur. Bu b\u00f6l\u00fcm, veri b\u00fct\u00fcnle\u015ftirme ve analizinin olas\u0131 bir \u00e7er\u00e7evesi olan ba\u011fl\u0131 veri (BV) kullan\u0131m\u0131n\u0131 ele al\u0131r. \u00d6A'daki BV giri\u015fimlerinin ve e\u011fitsel veri madencili\u011finin (EVM) literat\u00fcr taramas\u0131n\u0131 sa\u011flar ve bu alanlarda BV'nin kullan\u0131m\u0131yla ilgili imk\u00e2n ve zorluklardan baz\u0131lar\u0131n\u0131 ele al\u0131r.<\/span>\n\nAnahtar Kelimeler<\/span>: Ba\u011fl\u0131 veri (BV), veri b\u00fct\u00fcnle\u015ftirme, i\u00e7erik analizi, e\u011fitsel veri madencili\u011fi (EVM)<\/span>\n

Kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i derslerin (KA\u00c7D) ortaya \u00e7\u0131k\u0131\u015f\u0131 ve a\u00e7\u0131k veri giri\u015fimi, \u00fcniversite merkezli bir modelden, \u00e7ok platformlu ve \u00e7ok kaynakl\u0131 bir modele ge\u00e7erek e\u011fitimde yeni f\u0131rsatlar\u0131n ortaya \u00e7\u0131kmas\u0131n\u0131 sa\u011flam\u0131\u015ft\u0131r. Asl\u0131nda, g\u00fcn\u00fcm\u00fcz\u00fcn \u00f6\u011frenme ortamlar\u0131 yaln\u0131zca \u00e7e\u015fitli \u00e7evrimi\u00e7i \u00f6\u011frenme platformlar\u0131n\u0131 de\u011fil, ayn\u0131 zamanda \u00f6\u011frenenlerin ba\u011fland\u0131\u011f\u0131, ileti\u015fim kurdu\u011fu, veri ve kaynaklar\u0131 de\u011fi\u015f toku\u015f ettikleri sosyal medya uygulamalar\u0131n\u0131 da (\u00f6r. SlideShare, YouTube, Facebook, Twitter veya Linkedin) kapsar. Hem formal (\u00fcniversite dersleri) hem de informal (sosyal medya, KA\u00c7D) d\u00fczeylerde olan \u00f6\u011frenme art\u0131k \u00e7e\u015fitli bi\u00e7imlerde ve ortamlarda ger\u00e7ekle\u015fmektedir. Bu \u00f6\u011frenen verilerinin \u00e7e\u015fitli platform ve ara\u00e7larda da\u011f\u0131lmas\u0131n\u0131 sa\u011flam\u0131\u015ft\u0131r ve \u00f6\u011frenen verilerini \u00f6\u011frenme ortam\u0131na kapsaml\u0131 bir bak\u0131\u015f a\u00e7\u0131s\u0131 i\u00e7in \u00e7e\u015fitli ortamlarda ba\u011flamak i\u00e7in etkili ara\u00e7lara ihtiya\u00e7 duyulmas\u0131na neden olmu\u015ftur. Platformlar aras\u0131nda veri al\u0131\u015fveri\u015fi ihtiyac\u0131n\u0131n belirgin bir \u00f6rne\u011fi, ba\u011flant\u0131c\u01311<\/sup><\/a> bKA\u00c7D'dir. bKA\u00c7D'lerde \u00f6\u011frenme, tan\u0131m gere\u011fi, tek bir platformda ger\u00e7ekle\u015fmez ancak \u00f6\u011frenenler aras\u0131nda bilgi ve kaynak payla\u015f\u0131m\u0131 i\u00e7in bir dizi \u00f6zel \u00e7evrimi\u00e7i \u00f6\u011frenme uygulamas\u0131na ve sosyal medya ve a\u011f olu\u015fturma uygulamalar\u0131na dayan\u0131r (Siemens, 2005). Bu geli\u015fmeler yeni ihtiya\u00e7lara yol a\u00e7arak hem veri toplamada hem de verinin kullan\u0131m\u0131nda yeni zorluklar\u0131 beraberinde getirmi\u015ftir.<\/p>\n

Veri toplama a\u00e7\u0131s\u0131ndan bak\u0131ld\u0131\u011f\u0131nda, bulut hizmetlerinin ortaya \u00e7\u0131k\u0131\u015f\u0131 ve \u00f6l\u00e7eklenebilir web mimarilerinin h\u0131zl\u0131 geli\u015fimi, \u00e7e\u015fitli \u00e7evrimi\u00e7i uygulamalardan veri \u00e7ekilmesine ve kullan\u0131lmas\u0131na izin verir. Bu Facebook, Linkedin veya Twitter gibi b\u00fcy\u00fck web payda\u015flar\u0131 ve Coursera ve Udacity gibi KA\u00c7D sa\u011flay\u0131c\u0131lar\u0131 taraf\u0131ndan b\u00fcy\u00fck \u00f6l\u00e7ekli aray\u00fczlerin (API) geli\u015ftirilmesiyle desteklenmektedir. Veri kullan\u0131m\u0131 a\u00e7\u0131s\u0131ndan bak\u0131ld\u0131\u011f\u0131nda, e\u011fitim platformlar\u0131nda ortaya \u00e7\u0131kan kaynaklar\u0131n ve etkile\u015fimlerin \u00e7oklu\u011fu, farkl\u0131 t\u00fcrde verilerin ele al\u0131nmas\u0131n\u0131 da kapsayan analitik yetenekleri gerektirir. \u00dcretilen \u00e7e\u015fitli veri t\u00fcrlerinden baz\u0131lar\u0131 \u00f6\u011frenenlerin \u00f6\u011frenme etkile\u015fimleri ve sosyal platformlarda ger\u00e7ekle\u015ftirdikleri etkile\u015fimlerini (\u00f6\u011frenenlerin g\u00fcnl\u00fckleri\/izleri) tutarken di\u011fer veri t\u00fcrleri ders i\u00e7eriklerinden, \u00f6\u011frenenlerin bloglar\u0131ndan ve tart\u0131\u015fma forumlar\u0131na kadar uzanan yap\u0131land\u0131r\u0131lmam\u0131\u015f i\u00e7erik bi\u00e7imlerini de tutar. Bu \u00e7ok \u00e7e\u015fitli t\u00fcr ve veri kaynaklar\u0131, \u00f6\u011frenenleri ve \u00f6\u011frenme s\u00fcrecini daha iyi anlamak, zaman\u0131nda, bilgilendirici ve uyarlamal\u0131 geri bildirim sa\u011flamak ve hayat boyu \u00f6\u011frenmeyi te\u015fvik etmek i\u00e7in \u00f6\u011frenme analiti\u011fi alan\u0131 ve genel hedefleri i\u00e7in verimli bir zemin sa\u011flar (Gaesvic, Dawson ve Siemens, 2015).<\/p>\n

Heterojen kaynaklardan gelen verilerin toplanmas\u0131, entegrasyonu ve kullan\u0131m\u0131yla ilgili zorluklar, genellikle e\u011fitim toplulu\u011funda, heterojen verilerin geli\u015ftirilmesine ve onlardan yararlan\u0131lmas\u0131na izin veren standart bir veri modeli geli\u015ftirilerek ele al\u0131nmaktad\u0131r (Dietze vd., 2013). Bu b\u00f6l\u00fcm hem formal hem de informal \u00e7evrimi\u00e7i \u00f6\u011frenme ortamlar\u0131nda b\u00f6yle bir veri modelinin geli\u015ftirilmesine ve kullan\u0131lmas\u0131na y\u00f6nelik potansiyel bir yakla\u015f\u0131m olan ba\u011fl\u0131 veriye (BV) odaklanmaktad\u0131r. \u00d6zellikle, BV ilkelerinin kullan\u0131lmas\u0131 (Bizer, Heath ve Berners \u2013 Lee, 2009), \u00f6\u011frenme ortamlar\u0131nda k\u00fcresel olarak kullan\u0131labilir bir bilgi a\u011f\u0131 kurulmas\u0131na izin vererek (d'Aquin, Adamou ve Dietze, 2013), k\u00fcresel bir e\u011fitim grafi\u011fine \u00f6nc\u00fcl\u00fck eder. Her bir \u00f6\u011frenen i\u00e7in, \u00f6\u011frenme etkinlikleriyle ilgili t\u00fcm veri ve kaynaklar\u0131 birbirine ba\u011flayan, bireysel d\u00fczeyde benzer grafikler olu\u015fturulabilir. Bu t\u00fcr grafiklerin e\u011fitim potansiyelleri ve yararlar\u0131 \u00f6nceden incelenmi\u015f ve ele al\u0131nm\u0131\u015ft\u0131r. \u00d6rne\u011fin, Heath ve Bizer (2011), \u0130ngiliz \u00fcniversitelerinde, \u00f6\u011frenme kaynaklar\u0131n\u0131n i\u00e7eriklerinden elde edilen bilgileri kapsayan bir e\u011fitim grafi\u011fi<\/i><\/span> \u00f6nerirler. Bilgi \u00e7izgelerinin2<\/sup><\/a>, Google, Microsoft ve Facebook gibi artan say\u0131da \u00f6nemli firmalar taraf\u0131ndan geli\u015ftirilmesi ve kullan\u0131lmas\u0131 dikkate al\u0131nd\u0131\u011f\u0131nda bu t\u00fcr \u00e7izgelerin \u00f6\u011frenmeye sa\u011flad\u0131\u011f\u0131 potansiyel ve imk\u00e2nlar incelenmelidir. (Zablith, 2015).<\/p>\n

Bu b\u00f6l\u00fcm, \u00f6ncelikle \u00f6\u011frenme analiti\u011fi (\u00d6A) \/ e\u011fitsel veri madencili\u011fi (EVM) alan\u0131ndaki mevcut ve potansiyel uygulamalara odaklanarak e\u011fitimde BV kullan\u0131m\u0131n\u0131n mevcut durumunu incelemektedir. Bir sonraki b\u00f6l\u00fcmde BV ilkelerine k\u0131sa bir giri\u015ften sonra, BV potansiyeli iki \u00f6zel boyutta analiz edilir: 1) veri birle\u015ftirmenin boyutu ve 2) veri analizi ve yorumlama boyutu. Son olarak, \u00d6A \/ EVM'de BV kullan\u0131m\u0131 ile ilgili baz\u0131 olas\u0131l\u0131klar\u0131 ve zorluklar\u0131 ele almaktay\u0131z.<\/p>\n\n

E\u011e\u0130T\u0130MDE BA\u011eLI VER\u0130LER<\/h2>\n

Ba\u011fl\u0131 veri, internet \u00fczerinde kaynaklar\u0131 payla\u015fmak i\u00e7in fiil\u00ee bir standart olma potansiyeline sahiptir (Kessler, d'Aquin ve Dietze, 2013). Ba\u011fl\u0131 veri, varl\u0131klar\u0131n kendine \u00f6zg\u00fc tan\u0131mlamalar\u0131n\u0131 yapmak i\u00e7in GBT'leri ve varl\u0131klar\u0131 tan\u0131mlamak i\u00e7in KT\u00c7 veri modeli1<\/span><\/sup> kullanarak bu tan\u0131mlamalar\u0131 ba\u011flant\u0131lar (links) arac\u0131l\u0131\u011f\u0131yla a\u00e7\u0131k bir \u015fekilde tan\u0131mlanm\u0131\u015f anlamlarla ili\u015fkilendirir. Esasen, BV d\u00f6rt ilkeye dayanmaktad\u0131r:<\/p>\n\n

    \n \t
  1. \n

    GBT'leri durum isimleri gibi \u00f6rne\u011fin \"Paris\" adl\u0131 bir tarihi roman\u0131n kendine \u00f6zg\u00fc tan\u0131mlamas\u0131n\u0131 ISBN (bir t\u00fcr GBT) kullanarak yapal\u0131m: 0385535309<\/p>\n<\/li>\n \t

  2. \n

    HTTP GBT'leri arac\u0131l\u0131\u011f\u0131yla adlar\u0131 arama yetene\u011fi sa\u011flar; bir ISBN benzersiz bir kitab\u0131 tan\u0131mlarken, web \u00fczerinde do\u011frudan eri\u015fim sa\u011flamak i\u00e7in kullan\u0131lamaz, bu y\u00fczden bunun yerine HTTP GBT'leri kullan\u0131lmal\u0131d\u0131r; \u00d6rne\u011fimizde yer alan kitab\u0131m\u0131za \u015fu HTTP GBT'sinden bak\u0131labilir: <http:\/\/www.worldcat.org\/oclc\/827951628><\/span><\/p>\n<\/li>\n \t

  3. \n

    GBT'ye bakt\u0131ktan sonra, KT\u00c7 ve SPARQL3<\/sup><\/a><\/span> standartlar\u0131n\u0131 kullanarak faydal\u0131 bilgiler d\u00f6nd\u00fcr\u00fcn; \u00f6rne\u011fin, <http:\/\/www.worldcat.org\/oclc\/827951628> GBT'si taraf\u0131ndan tan\u0131mlanan kayna\u011f\u0131n kitap t\u00fcr\u00fcnde<\/i><\/span> oldu\u011funu ve tarihsel kurgu kategorisinde ait oldu\u011funu makineyle i\u015flenebilir bir \u015fekilde ifade edebiliriz: <http:\/\/www.worldcat.org\/oclc\/827951628> rdf:t\u00fcr \u015fema:Kitap; \u015fema:t\u00fcr \"Tarihsel kurgu\".<\/i><\/span><\/p>\n<\/li>\n \t

  4. \n

    Varl\u0131klar\u0131n GBT'leri arac\u0131l\u0131\u011f\u0131yla kendine \u00f6zg\u00fc tan\u0131mlanan di\u011fer varl\u0131klara ili\u015fkin ba\u011flant\u0131lar\u0131 (linkler) d\u00e2hil edin; \u00f6rnek olarak ald\u0131\u011f\u0131m\u0131z kitab\u0131m\u0131z\u0131n yazar\u0131yla ba\u011flant\u0131 kurabiliriz: <http: \/\/ www.worldcat.org\/oclc\/827951628> \u015fema: yazar <http:\/\/viaf.org\/viaf\/34666><\/i><\/span> buradaki GBT yazar Edward Rutherfurd'\u0131 benzersiz olarak tan\u0131mlar.<\/p>\n<\/li>\n<\/ol>\n

    Bu ilkelerin sadeli\u011fi sayesinde, BV m\u00fckemmel bir modelleme \u00e7er\u00e7evesini ve k\u00fcresel boyutta veri sorgulamay\u0131 sunar. \u00c7e\u015fitli uygulamalarda ve alanlarda kullan\u0131labilir ve e\u011fitim toplulu\u011fuyla uzun s\u00fcredir kar\u015f\u0131la\u015f\u0131lan birlikte \u00e7al\u0131\u015fabilirlik ve veri y\u00f6netimi zorluklar\u0131na bir cevap olu\u015fturabilir (Dietze vd., 2013).<\/p>\n

    ]Web \u00fczerinde ba\u011fl\u0131 veri olarak yay\u0131nlanan milyarlarca veri \u00f6gesi; devlete ait veri, bilimsel bilgi ve \u00e7e\u015fitli \u00e7evrimi\u00e7i topluluklar\u0131 i\u00e7eren veriler gibi birka\u00e7 alan ad\u0131ndan gelen bir a\u00e7\u0131k ba\u011fl\u0131 veri bulutunun (BAV)4<\/sup><\/a> yer ald\u0131\u011f\u0131 k\u00fcresel bir a\u00e7\u0131k veri uzay\u0131n\u0131 olu\u015fturur. DBpedia5<\/sup><\/a>, Yago6<\/sup><\/a> ve Wikidata7<\/sup><\/a> gibi BAV'lerde b\u00fcy\u00fck etki alanlar\u0131 aras\u0131 bilgi tabanlar\u0131 da ortaya \u00e7\u0131km\u0131\u015ft\u0131r. Bu nedenle BV, web sa\u011flay\u0131c\u0131lar\u0131 ve web API'leri de d\u00e2hil olmak \u00fczere binbir \u00e7e\u015fit veri eri\u015fim noktalar\u0131 arac\u0131l\u0131\u011f\u0131yla veriye nas\u0131l ula\u015f\u0131laca\u011f\u0131 ve kullan\u0131laca\u011f\u0131 konusunda k\u00fcresel bir de\u011fi\u015fimi m\u00fcmk\u00fcn k\u0131lan potansiyele sahip olup dinamik veri birle\u015ftirmelerinin sorunsuz bir \u015fekilde olu\u015fturulmas\u0131n\u0131 sa\u011flar. (Bizer vd., 2009). Asl\u0131nda, BV'nin g\u00f6ze \u00e7arpan bir \u00f6zelli\u011fi, farkl\u0131 veri kaynaklar\u0131ndan gelen \u00f6geler aras\u0131nda anlam bak\u0131m\u0131ndan zengin ba\u011flant\u0131lar kurmas\u0131 ve b\u00f6ylece daha sorunsuz veri entegrasyonu ve yeniden kullan\u0131m\u0131 i\u00e7in veri silolar\u0131 (\u00f6r. geleneksel veritabanlar\u0131) a\u00e7mas\u0131d\u0131r.<\/p>\n

    T\u00fcm bu potansiyel faydalara ra\u011fmen, BV bi\u00e7imcili\u011fi ve teknolojileri, teknoloji destekli \u00f6\u011frenme alan\u0131nda yava\u015f yava\u015f benimsenmekte olup; BV teknolojilerini kullanan giri\u015fimler son zamanlarda ortaya \u00e7\u0131kmaktad\u0131r (Dietze vd., 2013). \u00d6A \/ EVM alan\u0131nda, 1) kaynak ke\u015ffi (\u00f6r. Ayr\u0131nt\u0131l\u0131 arama) ve i\u00e7erik zenginle\u015ftirme (\u00f6r. BAV veri k\u00fcmelerindeki verilerle i\u00e7eri\u011fi art\u0131rma), (Maturana, Alvarado, Lopez \u2013Sola, Ibanez ve Elosegui, 2013); 2) anlamsal a\u00e7\u0131klamalara dayal\u0131 i\u00e7erik analizi (Joksimovi vd., 2015); 3) kaynak ve hizmet entegrasyonu (Dietze vd., 2012); 4) ki\u015fiselle\u015ftirme (Dietze, Drachsler ve Giordano, 2014) ve 5) EVM sonu\u00e7lar\u0131n\u0131n yorumlanmas\u0131 (d'Aquin vd., 2013) d\u00e2hil olmak \u00fczere BAV'den fayda sa\u011flayacak \u00e7e\u015fitli uygulama senaryolar\u0131 belirleyebiliriz.<\/p>\n\n

    BA\u011eLI VER\u0130LER\u0130 KULLANARAK VER\u0130 ENTEGRASYONU<\/h2>\n

    BV'nin en belirgin faydalar\u0131ndan biri, veri b\u00fct\u00fcnle\u015ftirme potansiyelinde yatmaktad\u0131r. Bu \u00d6A \/ EVM alan\u0131 i\u00e7in \u00f6zellikle \u00f6nemlidir, \u00e7\u00fcnk\u00fc informal ve hayat boyu \u00f6\u011frenmede kullan\u0131lan \u00e7e\u015fitli kaynaklardan (uygulamalar ve servisler) \u00f6\u011frenen ve i\u00e7erik verilerinin toplanmas\u0131n\u0131 ve y\u00f6netilmesini gerektirir (Santos vd., 2015). \u00d6zellikle, kapsaml\u0131 bir \u00f6\u011frenen modeli olu\u015fturmak i\u00e7in, \u00f6\u011frenenin etkile\u015fime girdi\u011fi farkl\u0131 \u00f6\u011frenme platformlar\u0131na \/ ara\u00e7lar\u0131na kaydedilen \u00f6\u011frenen verilerini birle\u015ftirmesi gerekir (Desmarais ve Baker, 2012). Bu nedenle, \u00e7oklu veri formatlar\u0131n\u0131n ele al\u0131nmas\u0131 ve genel olarak birlikte \u00e7al\u0131\u015fabilirlik eksikli\u011fi ile ilgili zorluklar kilit bir sorun haline gelmektedir (Chatti, Dyckhoff, Schroeder ve Thus, 2012; Duval, 2011). Daha genel olarak, \u00f6\u011frenme platformlar\u0131 aras\u0131nda anlam kayb\u0131 olmadan veri aktar\u0131m\u0131, \u00f6n i\u015fleme, kullan\u0131m, birle\u015ftirme ve analiz kolayl\u0131\u011f\u0131 \u00d6A \/ EVM'nin etkinli\u011fi i\u00e7in \u00f6nemli fakt\u00f6rler haline gelmektedir (Cooper, 2013).<\/p>\n

    Biyomedikal (Belleau, Nolin, Tourigny, Rigault ve Morissette, 2008), farmakoloji (Groth vd., 2014) ve \u00e7evre bilimleri (Lausch, Schmidt ve Tischendorf, 2015) gibi bir\u00e7ok alan veri b\u00fct\u00fcnle\u015ftirme sorunlar\u0131n\u0131 BV' den faydalanarak \u00e7\u00f6z\u00fcmlemi\u015ftir. T\u00fcm bunlar, BV teknolojilerinin \u00d6A \/ EVM'nin gerektirdi\u011fi sa\u011flam veri b\u00fct\u00fcnle\u015ftirme katman\u0131na imk\u00e2n verebilece\u011fini ortaya koymaktad\u0131r.<\/p>\n\n

    E\u011fitsel Topluluklarda Veri B\u00fct\u00fcnle\u015ftirmenin \u00d6nceki Giri\u015fimleri<\/h3>\n

    Teknoloji destekli \u00f6\u011frenme ara\u015ft\u0131rma toplulu\u011fu uzun zamand\u0131r \u00e7oklu standartla\u015ft\u0131rma \u00e7abalar\u0131yla sonu\u00e7lanan veri entegrasyonunun \u00f6nemini kabul etmi\u015ftir. Cooper (2013), \u00f6\u011frenmeyle ilgili \u00e7e\u015fitli standartlara de\u011ferli bir genel bak\u0131\u015f a\u00e7\u0131s\u0131 sunar. Temel olarak bu standartlar, e\u011fitim i\u00e7eriklerinin ve sa\u011flay\u0131c\u0131lar\u0131n\u0131n yan\u0131 s\u0131ra \u00f6\u011frenenlerin ve onlar\u0131n etkinliklerini i\u00e7eren verilerin temsili ile ilgilidir.<\/p>\n

    \u00d6\u011frenci d\u00fczeyinde, standartlar bireyler ve ge\u00e7mi\u015fleri ile ilgili ger\u00e7eklere, bunlar\u0131n di\u011fer insanlarla olan ba\u011flant\u0131lar\u0131na ve etkile\u015fimlerine ve \u00f6\u011frenme ortamlar\u0131 taraf\u0131ndan sunulan kaynaklarla etkile\u015fime odaklan\u0131r (ki\u015fi ve \u00f6\u011frenme etkinlikleri boyutlar\u0131). Model \u00f6\u011frenenler (\u00f6r. FOAF8<\/sup><\/a>) ve \u00f6\u011frenenlerin etkinlikleri ve etkile\u015fimleri i\u00e7in \u00e7e\u015fitli belirtimler mevcuttur (\u00f6r. Ba\u011flamsalla\u015ft\u0131r\u0131lm\u0131\u015f \u0130lgi \u00dcst Verisi [Schmitz, Wolpers, Kirschenmann ve Niemann, 2011], Etkinlik Ak\u0131\u015flar\u01319<\/sup><\/a> veya ADL xAPI10<\/sup><\/a>).<\/p>\n

    \u0130\u00e7erik d\u00fczeyinde, IEEE \u00d6\u011frenme Nesnesi \u00dcst Verisi (LO\u00dcV11<\/sup><\/a>) ve ADL SCORM12<\/sup><\/a> gibi \u00f6nceki giri\u015fimler, \u00e7evrimi\u00e7i e\u011fitim kaynaklar\u0131n\u0131n tan\u0131m\u0131n\u0131 veya bilgisayar destekli de\u011ferlendirmenin belirtimini (\u00f6r. IMS QTI13<\/sup><\/a>) birle\u015ftiren kelimeler ve standartlar olu\u015fturmaya \u00e7al\u0131\u015ft\u0131. Di\u011fer \u00e7abalar, sosyal ve etkile\u015fim verileri k\u00fcmelerini bir araya getirmeyi ama\u00e7layan Niemann, Wolpers, Stoitsis, Chinis ve Manouselis (2013) \u00e7al\u0131\u015fmalar\u0131 gibi \u00e7e\u015fitli veri modelleri aras\u0131ndaki haritalamay\u0131 hedeflemi\u015ftir. Son olarak, bu standartlarla tutarl\u0131 olan hizmetlerin uygulanmas\u0131na y\u00f6nelik k\u0131lavuz sa\u011flamak i\u00e7in \u00e7e\u015fitli aray\u00fczler \u00f6nerilmi\u015ftir (Dietze vd., 2013).<\/p>\n

    Farkl\u0131 bak\u0131\u015f a\u00e7\u0131lar\u0131na dayanan bu \u00e7abalar birbirleriyle rekabet eden bir\u00e7ok proje ortaya \u00e7\u0131karm\u0131\u015ft\u0131r ve dolay\u0131s\u0131yla aralar\u0131nda pek birlikte \u00e7al\u0131\u015fabilirlik olmayan \u00e7e\u015fitli teknolojilere, dillere ve modellere y\u00f6nelik alt topluluklar olu\u015fturmu\u015ftur. BV felsefesi, birlikte \u00e7al\u0131\u015fabilirlik sorunlar\u0131na webde \u00e7ok say\u0131da modelin izin vermesi ve bu modellerin webde eri\u015filebilir semantik ba\u011flant\u0131lar kullanarak k\u00f6pr\u00fclenmesi yoluyla bir \u00e7\u00f6z\u00fcm sa\u011flar. Bu nedenle, farkl\u0131 olarak temsil edilen anlamsal olarak benzer modeller, farkl\u0131 modellerden kaynaklanan kavramlar aras\u0131nda anlaml\u0131 ba\u011flant\u0131lar kuran yaz\u0131l\u0131 linkler kullan\u0131larak ayn\u0131 hizada olabilir; \u00f6rne\u011fin, e\u015fitlik ba\u011flant\u0131lar\u0131 (owl:benzerOlan)<\/i><\/span> veya hiyerar\u015fik ba\u011flant\u0131lar (rdfs:ninAltgrubu veya skos:broader).<\/i><\/span><\/p>\n\n

    Ba\u011fl\u0131 Veri Kullanarak Mevcut Veri B\u00fct\u00fcnle\u015ftirme Giri\u015fimleri<\/h3>\n

    BV'ye dayal\u0131 entegrasyon, belirli konu alanlar\u0131ndaki varl\u0131k t\u00fcrlerini, varl\u0131klar\u0131n \u00f6zniteliklerini ve varl\u0131klar aras\u0131ndaki ba\u011flant\u0131 t\u00fcrlerini tan\u0131mlayan web eri\u015fimli s\u00f6zl\u00fcklerinin kullan\u0131labilirli\u011fini gerektirir. Ayr\u0131ca, belirli bir g\u00f6rev i\u00e7in birden fazla veri k\u00fcmesinden yararlan\u0131lmas\u0131na izin veren hizmetlerin kullan\u0131labilirli\u011finin yan\u0131 s\u0131ra verileri BV olarak a\u00e7\u0131\u011fa \u00e7\u0131karan hizmetlere de ba\u011fl\u0131d\u0131r. Bu b\u00f6l\u00fcm, e\u011fitim alan\u0131ndaki mevcut baz\u0131 s\u00f6z varl\u0131\u011f\u0131n\u0131 ve e\u011fitsel verileri \u00f6\u011frenme verisi olarak g\u00f6sterme \u00e7abalar\u0131n\u0131 tan\u0131tmaktad\u0131r. E\u011fitimle ilgili kelimelerin daha kapsaml\u0131 bir incelemesi Dietze vd. (2014). Bu b\u00f6l\u00fcmde ayr\u0131ca birden fazla BV veri k\u00fcmesinin entegrasyonunu kullanan hizmetlere \u00f6rnekler verilmi\u015ftir.<\/p>\n

    Gittik\u00e7e artan say\u0131da e\u011fitim kurumu, Birle\u015fik Krall\u0131k'taki A\u00e7\u0131k \u00dcniversitedeki gibi BV ilkelerini izleyerek verilerini yay\u0131mlamaktad\u0131r ve BV'yi, dersleri ayr\u0131nt\u0131l\u0131 ve yeniden kullan\u0131labilir bir \u015fekilde birbirine ba\u011flamak i\u00e7in y\u00fcksek\u00f6\u011frenim programlar\u0131 etraf\u0131nda kavramsal bir katman olarak kullanmay\u0131 \u00f6nermi\u015ftir. Ba\u015fka bir \u00e7al\u0131\u015fma da ABYNM14<\/sup><\/a> tabanl\u0131 becerileri zenginle\u015ftirilmi\u015f \u00f6z ge\u00e7mi\u015fler olu\u015fturmak i\u00e7in KA\u00c7D kurs a\u00e7\u0131klamalar\u0131na ba\u011flamaktad\u0131r (Zotou, Papantoniou, Kremer, Peristeras ve Tambouris, 2014). \u0130lgin\u00e7 bir \u015fekilde, yazarlar Coursera ve Udacity KA\u00c7D platformlar\u0131nda \u00f6\u011fretilen benzer becerileri tan\u0131mlayabiliyorlar, b\u00f6ylece iki farkl\u0131 KA\u00c7D platformunun kurslar\u0131 aras\u0131nda \u00f6rt\u00fcl\u00fc ba\u011flant\u0131lar sa\u011fl\u0131yorlar. Birisi, platformlar aras\u0131 bir KA\u00c7D kurs \u00f6neri servisine dayanan hayat boyu \u00f6\u011frenme i\u00e7in heyecan verici f\u0131rsatlar \u00f6ng\u00f6rebilir.<\/p>\n

    Genel olarak e\u011fitim alan\u0131nda BV'nin ve \u00f6zellikle de \u00d6A \/ EVM'nin artan \u00f6neminin bir ba\u015fka g\u00f6stergesi, BV kavramlar\u0131n\u0131n ve teknolojilerinin xAPI belirtimine15<\/sup><\/a> uyarlanmas\u0131d\u0131r. XAPI ile, geli\u015ftiriciler \u00f6nceden tan\u0131mlanm\u0131\u015f bir arabirim ve bir dizi depolama ve alma kurallar\u0131 arac\u0131l\u0131\u011f\u0131yla bir \u00f6\u011frenme deneyimi izleme hizmeti olu\u015fturabilir. De Nies, Salliau, Verborgh, Mannens ve Van de Walle (2015), xAPI belirtimini BV olarak kullanarak olu\u015fturulan veri modellerini g\u00f6stermeyi \u00f6nermektedir. Bu \u00f6neri, birlikte \u00e7al\u0131\u015fabilir bir \u00f6\u011frenme izleri verisi modeli sa\u011flar ve \u00f6\u011frenenlerin izlerinin semantik olarak birlikte \u00e7al\u0131\u015fabilir BV olarak kusursuz bir \u015fekilde ortaya \u00e7\u0131kmas\u0131n\u0131 sa\u011flar. Benzer \u015fekilde, Softic vd. (2014), Semantik Web teknolojilerinin (KT\u00c7, SPARQL) ki\u015fisel \u00f6\u011frenme ortamlar\u0131ndaki \u00f6\u011frenen g\u00fcnl\u00fcklerini modellemek i\u00e7in kullan\u0131m\u0131n\u0131 rapor etmektedir.<\/p>\n

    Temel altyap\u0131 olarak webin \u00f6l\u00e7eklenebilirli\u011fine ve W3C standartlar\u0131 KT\u00c7 ve SPARQL'in birlikte \u00e7al\u0131\u015fabilirli\u011fine dayanarak, benzer giri\u015fimlerin ademi merkeziyet\u00e7i ve uyarlanabilir \u00f6\u011frenme hizmetlerinin geli\u015ftirilmesine daha fazla katk\u0131da bulunabilece\u011fine inan\u0131yoruz.<\/p>\n\n

    BA\u011eLI VER\u0130 KULLANARAK VER\u0130LER\u0130N ANAL\u0130Z ED\u0130LMES\u0130 VE ANLAMLANDIRILMASI<\/h2>\n

    KA\u00c7D'lerde giderek artan miktardaki \u00f6\u011frenme i\u00e7eriklerinin yan\u0131 s\u0131ra \u00e7e\u015fitli \u00e7evrimi\u00e7i sosyal medya ve ileti\u015fim kanallar\u0131nda yer alan yap\u0131land\u0131r\u0131lmam\u0131\u015f i\u00e7eriklerin h\u0131zl\u0131 art\u0131\u015f\u0131 da dikkate al\u0131nmal\u0131d\u0131r. Birka\u00e7 isim vermek gerekirse ba\u015fl\u0131klar (konular), e\u011filimler (trendler) ve d\u00fc\u015f\u00fcnceler gibi uzaktan \u00f6\u011frenimle ilgili \u00f6gelerin ortaya \u00e7\u0131kar\u0131lmas\u0131n\u0131n otomatikle\u015ftirilmesine ihtiya\u00e7 duyulmaktad\u0131r. E\u011fer d\u00fczenli girdi verileri (\u00f6r. \u00f6\u011frenenlerin g\u00fcnl\u00fckleri), BAV veri k\u00fcmelerinden gelen (\u00f6r. kursla ili\u015fkilendirilen konularla ilgili veriler) gerekli bilgilerle zenginle\u015ftirilirse analitik i\u00e7in ihtiya\u00e7 duyulan ilgili \u00f6gelerin ortaya \u00e7\u0131kar\u0131lmas\u0131 ve \/ veya \u00f6nerilmesi iyile\u015ftirilebilir (d'Aquin ve Jay, 2013). DBpedia ve Yago gibi BAV etki alanlar\u0131 aras\u0131 bilgi tabanlar\u0131n\u0131n tek veya geleneksel i\u00e7erik analizi teknikleriyle (\u00f6r. sosyal a\u011f analizi, metin madencili\u011fi, \u00f6rt\u00fck anlamsal dizinleme) ile birlikte kullan\u0131lmas\u0131 halinde, \u00e7ekilen konular\u0131n daha kolay yorumlanmas\u0131na imk\u00e2n veren i\u00e7erik analizi bilgi tabanlar\u0131n\u0131n (\u00f6r. DBpedia) geli\u015ftirilmesi i\u00e7in umut verici bir yol sunar.<\/p>\n\n

    \u00d6A \/ EVM Alan\u0131ndaki Bilimsel Yay\u0131nlar\u0131n Analizi<\/h3>\n

    BV taraf\u0131ndan desteklenen ve e\u011fitim ba\u011flam\u0131yla ilgili bir di\u011fer uygulama alan\u0131, semantik yay\u0131nlama (\u00f6r. kitap kataloglar\u0131n\u0131 BV olarak yay\u0131nlama) ve bilimsel yay\u0131nlar\u0131n \u00fcst analizidir. Asl\u0131nda, BV teknolojilerinin temel ba\u015far\u0131lar\u0131ndan biri, BNF16<\/sup><\/a> gibi \u00e7e\u015fitli i\u00e7erik yay\u0131nc\u0131lar\u0131 ve DBLP17<\/sup><\/a> gibi bilimsel tabanl\u0131 yay\u0131nc\u0131l\u0131k giri\u015fimleri taraf\u0131ndan erken benimsenmi\u015f olmas\u0131d\u0131r. Bu bilimsel yay\u0131nlarla ilgili BAV kelimelerinin ve veri k\u00fcmelerinin \u00e7oklu\u011funa yol a\u00e7m\u0131\u015ft\u0131r. Bu veri k\u00fcmeleri; belirli bir alana ait konular\u0131 belirleyen, ara\u015ft\u0131rmac\u0131lar\u0131 etkileyen ve ara\u015ft\u0131rma toplulu\u011funu ayr\u0131nt\u0131l\u0131 olarak tan\u0131mlayan \u00e7e\u015fitli bilimetrik hesaplamalar i\u00e7in temel olu\u015fturur (Mirriahi, Ga\u0161evi\u0107, Dawson ve Long, 2014; Ochoa, Suthers, Verbert ve Duval, 2014). Ayr\u0131ca, ilgili bilgileri bulmak i\u00e7in e\u011fitim sekt\u00f6r\u00fcnden gelen uzmanlara (ara\u015ft\u0131rmac\u0131lar, \u00f6\u011frenciler, k\u00fct\u00fcphaneciler, kurs \u00fcreticileri) an\u0131nda yard\u0131mc\u0131 olurlar.<\/p>\n

    \u00d6A \/ EVM alan\u0131nda, \u00d6\u011frenme Analiti\u011fi ve Bilgi (\u00d6AB) Veri K\u00fcmesi (Taibi ve Dietze, 2013), \u00d6A \/ EVM topluluklar\u0131ndan bir yay\u0131n derlemi temsil eder. \u00d6AB Veri K\u00fcmesi hem yay\u0131nlar\u0131n i\u00e7eri\u011fini hem de \u00fcst verilerini (\u00f6r. anahtar kelimeler, yazarlar, konferans) i\u00e7erir. \u00d6AB Veri K\u00fcmesi, var olan \u00e7e\u015fitli BAV kelime bilgilerine dayand\u0131\u011f\u0131 ve BV teknolojilerinin ba\u015far\u0131l\u0131 bir uygulamas\u0131n\u0131 kapsad\u0131\u011f\u0131ndan veri b\u00fct\u00fcnle\u015ftirme giri\u015fimini ifade eder. \u00d6AB Veri K\u00fcmesinin analizi, 2013'ten bu yana, \u00d6A \/ EVM yay\u0131nlar\u0131yla ilgili ara\u015ft\u0131rma ve analitik \u00e7al\u0131\u015fmalar\u0131 te\u015fvik eden y\u0131ll\u0131k \u00d6AB Veri M\u00fccadelesi ile te\u015fvik edilmi\u015ftir. Bu veri k\u00fcmesi, veri analiti\u011fi ve i\u00e7erik analizi uygulamalar\u0131n\u0131n geli\u015ftirilmesi i\u00e7in daha fazla kullan\u0131lm\u0131\u015ft\u0131r. \u00d6zellikle, konular\u0131n ve veri k\u00fcmesindeki konular aras\u0131ndaki ili\u015fkilerin y\u0131ll\u0131k olarak, topluluk ba\u015f\u0131na (\u00d6A-EVM), yay\u0131n ba\u015f\u0131na ve ayr\u0131nt\u0131l\u0131 olarak saptanmas\u0131 de\u011ferli bir uygulamad\u0131r. \u00d6rne\u011fin, Zouaq, Joksimovic ve Ga\u0161evi\u0107'in (2013) \u00e7al\u0131\u015fmas\u0131, g\u00f6ze \u00e7arpan konular\u0131 ve aralar\u0131ndaki ili\u015fkileri belirlemek i\u00e7in \u00d6AB Veri K\u00fcmesinde ontoloji \u00f6\u011frenme tekniklerini kullanm\u0131\u015ft\u0131r. Konular\u0131 ke\u015ffetmek i\u00e7in uygulanan di\u011fer teknikler aras\u0131nda gizli Dirichlet tahsisi (GDT; Sharkey ve Ansari, 2014) ve k\u00fcmelemesi (Scheffel, Niemann, Leon Rojas, Drachsler ve Specht, 2014) say\u0131labilir. Bu yakla\u015f\u0131mlar metne dayal\u0131 bir i\u00e7erik analizi sunarken, di\u011fer konular da veri entegrasyon \u00e7abalar\u0131nda BAV bilgi tabanlar\u0131na (\u00f6r. DBpedia) ve ilgi alanlar\u0131n\u0131 tan\u0131mlamak i\u00e7in semantik a\u00e7\u0131klamalara dayanarak daha da ileri gitmi\u015ftir. \u00d6rne\u011fin; Miliki\u0107, Krcadinac, Jovanovi\u0107, Brankov, ve Keca (2013) ve Nunes, Fetahu ve Casanova (2013), yay\u0131nlarda konu ba\u015fl\u0131klar\u0131n\u0131 ve adland\u0131r\u0131lm\u0131\u015f varl\u0131klar\u0131 tan\u0131mlamak i\u00e7in s\u0131ras\u0131yla TagMe ve DBpedia Spotlight hizmetlerine g\u00fcvendi. Bu durumda BV'nin yarar\u0131 1) veri k\u00fcmesini BAV kavramlar\u0131, anahtar kelimeler ve temalarla zenginle\u015ftirme kabiliyeti ve 2) potansiyel i\u015fbirlikli tespiti (Hu vd., 2014), veri k\u00fcmesi \u00f6nerileri veya daha genel anlamsal aramalar gibi geli\u015fmi\u015f hizmetler geli\u015ftirme kabiliyeti taraf\u0131ndan vurgulanm\u0131\u015ft\u0131r (Nunes vd., 2013).<\/p>\n\n

    Veri Madencili\u011fi Sonu\u00e7lar\u0131n\u0131n Yorumlanmas\u0131<\/h3>\n

    Bir\u00e7ok ara\u015ft\u0131rma, \u00f6\u011frenenlerin etkile\u015fimlerini ve g\u00f6r\u00fc\u015fme verilerini (\u00f6r. \u00f6\u011frenenlerin konu\u015fmalar\u0131, fikirleri ve akademik performanslar\u0131 aras\u0131ndaki ba\u011flant\u0131y\u0131 belirleyerek (Dowell vd., 2015) veya kurs kay\u0131t verilerini (d'Aquin ve Jay, 2013) analiz ederek anla\u015f\u0131lmas\u0131n\u0131, ba\u011flant\u0131lar\u0131 ve kestirimci modelleri sunmu\u015ftur. Bununla birlikte, bu analizlerin \u00e7o\u011fu kapal\u0131 ya da silo veri k\u00fcmesiyle s\u0131n\u0131rl\u0131 kalmaktad\u0131r ve b\u00fcy\u00fck veri k\u00fcmelerinde yorumlanmalar\u0131 genellikle zordur.<\/p>\n

    Genel olarak, \u00d6A \/ EVM'deki model ke\u015ffi, sonu\u00e7lar\u0131n birka\u00e7 boyuta (\u00f6r. konular, \u00f6\u011frenci \u00f6zellikleri, \u00f6\u011frenme ortamlar\u0131, vb.) g\u00f6re anlaml\u0131 bir \u015fekilde yorumlanmas\u0131 i\u00e7in bir model ve insan analisti gerektirir (d'Aquin ve Jay, 2013). D'Aquin ve Jay (2013) taraf\u0131ndan yap\u0131lan \u00e7al\u0131\u015fma, veri madencili\u011fi s\u00fcrecinde ke\u015ffedilen kal\u0131plar\u0131 zenginle\u015ftirmek ve ba\u011flamla\u015ft\u0131rmak i\u00e7in BV'nin faydas\u0131 hakk\u0131nda yeni bilgiler sunmaktad\u0131r. \u00d6zellikle, ke\u015ffedilen \u00f6r\u00fcnt\u00fcleri BV GBT'leri ile birlikte a\u00e7\u0131klamay\u0131 teklif ederler, b\u00f6ylece bu \u00f6r\u00fcnt\u00fclerin yorumlamay\u0131 kolayla\u015ft\u0131rmak i\u00e7in mevcut veri k\u00fcmeleriyle daha da zenginle\u015ftirilebilir. Yazarlar, fikri zamanla ders mod\u00fcllerine \u00f6\u011frenci kayd\u0131yla ilgili bir \u00f6rnek olay incelemesi ile g\u00f6stermektedir. S\u0131k s\u0131k kurs dizilimlerini \u00e7\u0131kar\u0131rlar ve bunlar\u0131 kurs GBT'leri vas\u0131tas\u0131yla kurs a\u00e7\u0131klamalar\u0131yla, yani kursu tan\u0131mlayan bir dizi \u00f6zellikle ili\u015fkilendirerek zenginle\u015ftirirler. (Zincir) \u00f6zellikleri, \u00e7apraz temelli bir s\u0131n\u0131fland\u0131rmada (\u00f6r. kurs s\u0131ralar\u0131n\u0131n s\u0131kl\u0131kla tercih edilen ortak konular\u0131) ve navigasyon tabanl\u0131 bir yap\u0131 olarak kullan\u0131lan analitik boyutlar\u0131 sa\u011flar. Bu vaka \u00e7al\u0131\u015fmas\u0131nda g\u00f6sterildi\u011fi gibi, BV ke\u015ffedilen kal\u0131plar\u0131 d\u0131\u015f bilgi tabanlar\u0131na ba\u011flayarak ve yeni bilgiyi ortaya \u00e7\u0131karmak i\u00e7in BAV anlamsal ba\u011flant\u0131lar\u0131n\u0131 kullanarak yeni analitik boyutlar\u0131 ke\u015ffetmeye yard\u0131mc\u0131 olabilir. Bu \u00f6zellikle \u00e7e\u015fitli fakt\u00f6rlerin bir kal\u0131ba veya olguya katk\u0131da bulunabilece\u011fi \u00e7ok disiplinli ara\u015ft\u0131rmalarda \u00f6nemlidir. \u00d6\u011frenme davran\u0131\u015flar\u0131n\u0131n karma\u015f\u0131kl\u0131\u011f\u0131 g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, bu deste\u011fin \u00d6A \/ EVM sonu\u00e7lar\u0131n\u0131n yorumlanmas\u0131nda faydas\u0131 oldu\u011fu d\u00fc\u015f\u00fcn\u00fclebilir.<\/p>\n\n

    TARTI\u015eMA ve BAKI\u015e A\u00c7ISI<\/h2>\n

    \u00d6\u011frenme deneyimine y\u00f6nelik genel analitik yakla\u015f\u0131m, \u00f6\u011frenme verilerinin toplanmas\u0131, y\u00f6netimi, sorgulanmas\u0131, birle\u015ftirilmesi ve zenginle\u015ftirilmesi i\u00e7in son teknoloji veri y\u00f6netimi teknikleri gerektirir. BV kavram\u0131 ve teknolojisi, W3C standartlar\u0131na (KT\u00c7, SPARQL) dayanmaktad\u0131r ve veri y\u00f6netiminin t\u00fcm bu y\u00f6nlerine katk\u0131da bulunma potansiyeline sahiptir. \u0130lk olarak, BV teknolojilerinin ard\u0131ndaki temel ama\u00e7lardan biri, verilerin anlam bilimini koruyarak ve kald\u0131rarak, \u00e7e\u015fitli ama\u00e7lar i\u00e7in verileri kolayca i\u015flenebilir ve tekrar kullan\u0131labilir hale getirmektir. \u0130kincisi, BV, \u00e7e\u015fitli veri k\u00fcmelerinin sorunsuz bir \u015fekilde birle\u015ftirilmesini ve sorgulanmas\u0131n\u0131 sa\u011flayarak veri y\u00f6netimine merkezi olmayan bir yakla\u015f\u0131m sa\u011flar. \u00dc\u00e7\u00fcnc\u00fcs\u00fc, web \u00fczerindeki ba\u011flant\u0131l\u0131 a\u00e7\u0131k veriler olarak mevcut b\u00fcy\u00fck \u00f6l\u00e7ekli bilgi tabanlar\u0131, analitik s\u00fcre\u00e7le ilgili \u00e7e\u015fitli hizmetler i\u00e7in zemin sa\u011flar; \u00f6rne\u011fin, i\u00e7erik analizi ve zenginle\u015ftirme i\u00e7in semantik ek a\u00e7\u0131klamalar. D\u00f6rd\u00fcnc\u00fcs\u00fc, webde BV olarak a\u00e7\u0131\u011fa \u00e7\u0131kar\u0131lan veriler, analitik s\u00fcrecin farkl\u0131 a\u015famalar\u0131nda gerekli olan talep \u00fczerine (tam zaman\u0131nda) veri \/ bilgi giri\u015fi sa\u011flayabilir, \u00e7\u00fcnk\u00fc bu bilgi daima \u00f6nceden tam olarak tahmin edilemez. Potansiyel faydalar ayr\u0131ca sonu\u00e7 analizini anlamsal a\u00e7\u0131dan zengin bir formatta temsil etmeyi de i\u00e7erir, b\u00f6ylece sonu\u00e7lar uygulamalar aras\u0131nda payla\u015f\u0131labilir ve ilgili taraflara (\u00f6\u011fretmenler, \u00f6\u011frenciler) ihtiya\u00e7 ve tercihlere ba\u011fl\u0131 olarak (\u00f6r. farkl\u0131 g\u00f6rsel veya anlat\u0131m bi\u00e7imlerine ba\u011fl\u0131 olarak) farkl\u0131 \u015fekillerde iletilebilir. Di\u011fer taraftan, veri maddelerinin anlamsal olarak zengin temsilinden ve kar\u015f\u0131l\u0131kl\u0131 ili\u015fkilerinden kaynaklanan \u00e7oklu veri kaynaklar\u0131 \u00fczerindeki \u00e7\u0131kar\u0131m yetenekleri sayesinde, BV tabanl\u0131 y\u00f6ntemler, metin i\u00e7eri\u011findeki temalar\u0131 ve konular\u0131 ke\u015ffetmek i\u00e7in mevcut analitik y\u00f6ntemlere amaca uygun bir katk\u0131 olabilir. Ayr\u0131ca, \u00e7oklu veri kaynaklar\u0131 \u00fczerindeki \u00e7\u0131kar\u0131m yetenekleri sayesinde, veri \u00f6gelerinin anlamsal olarak zengin temsilinden ve kar\u015f\u0131l\u0131kl\u0131 ili\u015fkilerinden kaynaklanan BV tabanl\u0131 y\u00f6ntemler, metin i\u00e7eri\u011findeki temalar\u0131 ve konular\u0131 ke\u015ffetmek i\u00e7in mevcut analitik y\u00f6ntemlere alakal\u0131 bir ek olabilir. Daha genel olarak, \u00d6A \/ EVM toplulu\u011funda istatistiksel ve makine \u00f6\u011frenme y\u00f6ntemleri yayg\u0131n olmakla birlikte, verilerin a\u00e7\u0131k\u00e7a tan\u0131mlanm\u0131\u015f anlam bilimine ve a\u00e7\u0131k bilgi kaynaklar\u0131na (\u00f6zellikle a\u00e7\u0131k, web tabanl\u0131 bilgiye dayal\u0131) dayanan di\u011fer veri analiz y\u00f6ntemleri ve teknikleri geleneksel analitik yakla\u015f\u0131mlar\u0131 daha da g\u00fc\u00e7l\u00fc k\u0131lar.<\/p>\n

    Son olarak<\/span>, BV teknolojileri \u00f6\u011frenme ortamlar\u0131n\u0131n ve sosyal medya platformlar\u0131n\u0131n heterojenli\u011fi ile ba\u015fa \u00e7\u0131kmada yararl\u0131 olabilir. \u00d6zellikle, ortak bir \u015fema payla\u015fmayan \u00e7e\u015fitli veri k\u00fcmelerini sorgulayabilir ve birle\u015ftirebilirsiniz. Bu y\u00f6n\u00fcn kendisi, ortak bir modele \/ \u015femaya uyumu gerektiren \u00f6nceki yakla\u015f\u0131mlardan daha esnek ve pratik bir yakla\u015f\u0131m\u0131 temsil etmektedir.<\/p>\n

    Ancak BV Kullan\u0131m\u0131 ile ilgili olarak a\u015fa\u011f\u0131da zikretti\u011fimiz baz\u0131 zorluklardan bahsetmek m\u00fcmk\u00fcnd\u00fcr:<\/p>\n\n

      \n \t
    1. \n

      Kalite<\/span>: BAV veri k\u00fcmelerinin kalitesi bir endi\u015fe kayna\u011f\u0131d\u0131r (Kontokostas vd., 2014) ve \u00f6\u011frenme kaynaklar\u0131n\u0131 ve izlerini d\u0131\u015f veri k\u00fcmelerine ve bilgi tabanlar\u0131na ba\u011flamak karma\u015f\u0131k veriye yol a\u00e7abilir. Veri temizli\u011fi i\u00e7in baz\u0131 giri\u015fimler olmas\u0131na ra\u011fmen, bu sorun hala tam olarak \u00e7\u00f6z\u00fclememi\u015ftir.<\/p>\n<\/li>\n \t

    2. \n

      Hizalama<\/span>: \u015eemalar aras\u0131nda ortak web GBT'lerinin kullan\u0131m\u0131n\u0131n yan\u0131 s\u0131ra, zorlu bir g\u00f6rev olan kelime ve modelleri anlamsal olarak hizalamaya ihtiya\u00e7 duyulur. Mevcut uyum yakla\u015f\u0131mlar\u0131 genellikle belirsizlikler ile ba\u015fa \u00e7\u0131kmayan s\u00f6zdizimsel e\u015fle\u015fmeye dayanmaktad\u0131r. Uyum sorununu hafifletmenin bir yolu, m\u00fcmk\u00fcn oldu\u011funda ana BV kelimelerinin18<\/sup><\/a><\/span> fark\u0131nda olmak ve yeniden kullanmakt\u0131r (\u00f6r. foaf:isim<\/i><\/span>, FOAF belirtimindeki bir ki\u015finin ad\u0131n\u0131 g\u00f6steren ve yeni bir \u00f6zellik olu\u015fturmak yerine kullan\u0131labilir) ;<\/p>\n<\/li>\n \t

    3. \n

      Gizlilik<\/span>: KA\u00c7D'lerdeki ve \u00f6\u011frenme platformlar\u0131ndaki veriler \u00e7o\u011fu zaman gizlilik nedeniyle depolanmaktad\u0131r. Bilgiyi \u00f6\u011frenme ile sosyal platformlar aras\u0131nda birle\u015ftirmek, \u00f6\u011frenenlere verilere eri\u015fim izni ve \u00f6\u011frenmede kulland\u0131klar\u0131 farkl\u0131 hizmetler i\u00e7in giri\u015f bilgileri vermelerini gerektirir.<\/p>\n<\/li>\n<\/ol>\n

      Yukar\u0131da belirtilen ve BAV veri k\u00fcmelerinin ve bilgi tabanlar\u0131n\u0131n Google bilgi \u00e7izgesi19<\/sup><\/a> veya Facebook bilgi \u00e7izgesi aramas\u0131 gibi baz\u0131 b\u00fcy\u00fck firmalar\u0131n kullan\u0131lmas\u0131 ve bunlar\u0131n e\u011fitim kurumlar\u0131nda benimsenmesi artan BV'de kullan\u0131lmas\u0131na ra\u011fmen, bug\u00fcn\u00fcn \u00f6\u011frenme platformlar\u0131 i\u00e7in umut verici bir teknolojik belkemi\u011fidir. Ayr\u0131ca, ham veri toplama ve depolanmas\u0131ndan veri kullan\u0131m\u0131na ve zenginle\u015ftirmeye, analitik sonu\u00e7lar\u0131n yorumlanmas\u0131na kadar genel \u00f6\u011frenme analiti\u011fi s\u00fcrecini kolayla\u015ft\u0131rmak i\u00e7in yararl\u0131 bir bi\u00e7imcilik sa\u011flar.<\/p>\n\n

      KAYNAK\u00c7A<\/h2>\nBelleau, F., Nolin, M.-A., Tourigny, N., Rigault, P., & Morissette, J. (2008). Bio2RDF: Towards a mashup to build bioinformatics knowledge systems. Journal of Biomedical Informatics, 41<\/i>(5), 706\u2013716. <\/span>\n\nBizer, C., Heath, T., & Berners-Lee, T. (2009). Linked data - the story so far. International Journal on Semantic Web and Information Systems, 5<\/i>(3), 1\u201322. Preprint retrieved from http:\/\/tomheath.com\/papers\/bizer-heath-berners-lee-ijswis-linked-data.pdf <\/span>\n\nChatti, M. A., Dyckhoff, A. L., Schroeder, U., & Th\u00fcs, H. (2012). A reference model for learning analytics. International Journal of Technology Enhanced Learning, 4<\/i>(5\u20136), 318\u2013331. <\/span>\n\nCooper, A. R. (2013). Learning analytics interoperability: A survey of current literature and candidate standards. http:\/\/blogs.cetis.ac.uk\/adam\/2013\/05\/03\/learning-analytics-interoperability <\/span>\n\nd'Aquin, M., & Jay, N. (2013). Interpreting data mining results with linked data for learning analytics: Motivation, case study and directions. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 155\u2013164). New York: ACM. <\/span>\n\nd'Aquin, M., Adamou, A., & Dietze, S. (2013, May). Assessing the educational linked data landscape. Proceedings of the 5th Annual ACM Web Science Conference <\/i>(WebSci\u201913), 2\u20134 May 2013, Paris, France (pp. 43\u201346). New York: ACM. <\/span>\n\nDe Nies, T., Salliau, F., Verborgh, R., Mannens, E., & Van de Walle, R. (2015, May). TinCan2PROV: Exposing interoperable provenance of learning processes through experience API logs. Proceedings of the 24th International Conference on World Wide Web <\/i>(WWW\u201915), 18\u201322 May 2015, Florence, Italy (pp. 689\u2013694). New York: ACM. <\/span>\n\nDesmarais, M. C., & Baker, R. S. (2012). A review of recent advances in learner and skill modeling in intelligent learning environments. User Modeling and User-Adapted Interaction, 22<\/i>(1\u20132), 9\u201338. <\/span>\n\nDietze, S., Drachsler, H., & Giordano, D. (2014). A survey on linked data and the social web as facilitators for TEL recommender systems. In N. Manouselis, K. Verbert, H. Drachsler, & O. C. Santos (Eds.), Recommender Systems for Technology Enhanced Learning <\/i>(pp. 47\u201375). New York: Springer. <\/span>\n\nDietze, S., Sanchez-Alonso, S., Ebner, H., Qing Yu, H., Giordano, D., Marenzi, I., & Nunes, B. P. (2013). Interlinking educational resources and the web of data: A survey of challenges and approaches. Program, 47<\/i>(1), 60\u201391. <\/span>\n\nDietze, S., Yu, H. Q., Giordano, D., Kaldoudi, E., Dovrolis, N., & Taibi, D. (2012). Linked education: Interlinking educational resources and the web of data. Proceedings of the 27th Annual ACM Symposium on Applied Computing <\/i>(SAC 2012), 26\u201330 March 2012, Riva (Trento), Italy (pp. 366\u2013371). New York: ACM. Dowell, N. M., Skrypnyk, O., Joksimovi\u0107, S., Graesser, A. C., Dawson, S., Ga\u0161evi\u0107, D., Hennis, T. A., de Vries, P., & Kovanovi\u0107, V. (2015). Modeling learners' social centrality and performance through language and discourse. In O. C. Santos, J. G. Boticario, C. Romero, M. Pechenizkiy, A. Merceron, P. Mitros, J. M. Luna, C. Mihaescu, P. Moreno, A. Hershkovitz, S. Ventura, & M. Desmarais (Eds.), Proceedings of the 8th International Conference on Education Data Mining <\/i>(EDM2015), 26\u201329 June 2015, Madrid, Spain (pp. 250\u2013257). International Educational Data Mining Society. http:\/\/files.eric.ed.gov\/fulltext\/ED560532.pdf <\/span>\n\nDuval, E. (2011). Attention please! Learning analytics for visualization and recommendation. Proceedings of the 1st International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201911), 27 February\u20131 March 2011, Banff, AB, Canada (pp. 9\u201317). New York: ACM. <\/span>\n\nGa\u0161evi\u0107, D., Dawson, S., & Siemens, G. (2015). Let's not forget: Learning analytics are about learning. TechTrends, 59<\/i>(1), 64\u201371. <\/span>\n\nGroth, P., Loizou, A., Gray, A. J., Goble, C., Harland, L., & Pettifer, S. (2014). API-centric linked data integration: The open PHACTS discovery platform case study. Web Semantics: Science, Services and Agents on the World Wide Web, 29<\/i>, 12\u201318. <\/span>\n\nHeath, T., & Bizer, C. (2011). Linked data: Evolving the web into a global data space. Synthesis lectures on the semantic web: Theory and technology, 1<\/i>(1), 1\u2013136. Morgan & Claypool. <\/span>\n\nHu, Y., McKenzie, G., Yang, J. A., Gao, S., Abdalla, A., & Janowicz, K. (2014). A linked-data-driven web portal for learning analytics: Data enrichment, interactive visualization, and knowledge discovery. In LAK Workshops. http:\/\/geog.ucsb.edu\/~jano\/LAK2014.pdf <\/span>\n\nJoksimovi\u0107, S., Kovanovi\u0107, V., Jovanovi\u0107, J., Zouaq, A., Ga\u0161evi\u0107, D., & Hatala, M. (2015). What do cMOOC participants talk about in social media? A topic analysis of discourse in a cMOOC. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March, Poughkeepsie, NY, USA (pp. 156\u2013165). New York: ACM. <\/span>\n\nJovanovi\u0107, J., Bagheri, E., Cuzzola, J., Ga\u0161evi\u0107, D., Jeremic, Z., & Bashash, R. (2014). Automated semantic annotation of textual content. IEEE IT Professional, 16<\/i>(6), 38\u201346. <\/span>\n\nKagemann, S., & Bansal, S. (2015). MOOCLink: Building and utilizing linked data from massive open online courses. Proceedings of the 9th IEEE International Conference on Semantic Computing <\/i>(IEEE ICSC 2015), 7\u20139 February 2015, Anaheim, California, USA (pp. 373\u2013380). IEEE. <\/span>\n\nKessler, C., d\u2019Aquin, M., & Dietze, S. (2013). Linked data for science and education. Journal of Semantic Web, 4<\/i>(1), 1\u20132. <\/span>\n\nKontokostas, D., Westphal, P., Auer, S., Hellmann, S., Lehmann, J., Cornelissen, R., & Zaveri, A. (2014). Test-driven evaluation of linked data quality. Proceedings of the 23rd International Conference on World Wide Web <\/i>(WWW\u201914), 7\u201311 April 2014, Seoul, Republic of Korea (pp. 747\u2013758). New York: ACM. <\/span>\n\nLausch, A., Schmidt, A., & Tischendorf, L. (2015). Data mining and linked open data: New perspectives for data analysis in environmental research. Ecological Modelling, 295<\/i>, 5\u201317. <\/span>\n\nMaturana, R. A., Alvarado, M. E., L\u00f3pez-Sola, S., Ib\u00e1\u00f1ez, M. J., & El\u00f3segui, L. R. (2013). Linked data based applications for learning analytics research: Faceted searches, enriched contexts, graph browsing and dynamic graphic visualisation of data. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_03.pdf <\/span>\n\nMiliki\u0107, N., Krcadinac, U., Jovanovi\u0107, J., Brankov, B., & Keca, S. (2013). Paperista: Visual exploration of semantically annotated research papers. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_04.pdf <\/span>\n\nMirriahi, N., Ga\u0161evi\u0107, D., Dawson, S., & Long, P. D. (2014). Scientometrics as an important tool for the growth of the field of learning analytics. Journal of Learning Analytics, 1<\/i>(2), 1\u20134. <\/span>\n\nMu\u00f1oz-Merino, P. J., Pardo, A., Kloos, C. D., Mu\u00f1oz-Organero, M., Wolpers, M., Katja, K., & Friedrich, M. (2010). CAM in the semantic web world. In A. Paschke, N. Henze, & T. Pellegrini (Eds.), Proceedings of the 6th International Conference on Semantic Systems <\/i>(I-Semantics\u201910), 1\u20133 September 2010, Graz, Austria. New York: ACM. doi:10.1145\/1839707.1839737Newman, M. E. (2006). Modularity and community structure in networks, Proceedings of the National Academy of Sciences, 103<\/i>(23), 8577\u20138582. <\/span>\n\nNiemann, K., Wolpers, M., Stoitsis, G., Chinis, G., & Manouselis, N. (2013). Aggregating social and usage datasets for learning analytics: Data-oriented challenges. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 245\u2013249). New York: ACM. <\/span>\n\nNunes, B. P., Fetahu, B., & Casanova, M. A. (2013). Cite4Me: Semantic retrieval and analysis of scientific publications. LAK Data Challenge, 974. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_06.pdf <\/span>\n\nOchoa, X., Suthers, D., Verbert, K., & Duval, E. (2014). Analysis and reflections on the third learning analytics and knowledge conference (LAK 2013). Journal of Learning Analytics, 1<\/i>(2), 5\u201322. <\/span>\n\nPiedra, N., Chicaiza, J. A., L\u00f3pez, J., & Tovar, E. (2014). An architecture based on linked data technologies for the integration and reuse of OER in MOOCs context. Open Praxis, 6<\/i>(2), 171\u2013187. <\/span>\n\nSantos, J. L., Verbert, K., Klerkx, J., Duval, E., Charleer, S., & Ternier, S. (2015). Tracking data in open learning environments. Journal of Universal Computer Science, 21<\/i>(7), 976\u2013996. <\/span>\n\nScheffel, M., Niemann, K., Leon Rojas, S., Drachsler, H., & Specht, M. (2014). Spiral me to the core: Getting a visual grasp on text corpora through clusters and keywords. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol- 1137\/lakdatachallenge2014_submission_3.pdf <\/span>\n\nSchmitz, H. C., Wolpers, M., Kirschenmann, U., & Niemann, K. (2011). Contextualized attention metadata. In C. Roda (Ed.), Human attention in digital environments <\/i>(pp. 186\u2013209). New York: Cambridge University Press. <\/span>\n\nSharkey, M., & Ansari, M. (2014). Deconstruct and reconstruct: Using topic modeling on an analytics corpus. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-1137\/lakdatachallenge2014_submission_1.pdf <\/span>\n\nSiemens, G. (2005). Connectivism: A learning theory for the digital age. International Journal of Instructional Technology and Distance Learning, 2<\/i>(1), 3\u201310. http:\/\/itdl.org\/Journal\/Jan_05\/article01.htm <\/span>\n\nSoftic, S., De Vocht, L., Taraghi, B., Ebner, M., Mannens, E., & De Walle, R. V. (2014). Leveraging learning analytics in a personal learning environment using linked data. Bulletin of the IEEE Technical Committee on Learning Technology, 16<\/i>(4), 10\u201313. <\/span>\n\nTaibi, D., & Dietze, S. (2013), Fostering analytics on learning analytics research: The LAK dataset. LAK Data Challenge, 974. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_preface.pdf <\/span>\n\nZablith, F. (2015). Interconnecting and enriching higher education programs using linked data. Proceedings of the 24th International Conference on World Wide Web <\/i>(WWW\u201915), 18\u201322 May 2015, Florence, Italy (pp. 711\u2013716). New York: ACM. <\/span>\n\nZotou, M., Papantoniou, A., Kremer, K., Peristeras, V., & Tambouris, E. (2014). Implementing \u201crethinking education\u201d: Matching skills profiles with open courses through linked open data technologies. Bulletin of the IEEE Technical Committee on Learning Technology, 16<\/i>(4), 18\u201321. <\/span>\n\nZouaq, A., Joksimovi\u0107, S., & Ga\u0161evi\u0107, D. (2013). Ontology learning to analyze research trends in learning analytics publications. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_08.pdf<\/span><\/a><\/span><\/span>\n\n
      \n\n
      \n\n1<\/a> connectivist<\/span>\n\n<\/div>\n
      \n\n2<\/a> knowledge graph<\/span>\n\n<\/div>\n
      \n\n3<\/a> https:\/\/www.w3.org\/TR\/sparql11\u2013query\/<\/span>\n\n<\/div>\n
      \n\n4<\/a> http:\/\/lod-cloud.net\/<\/span>\n\n<\/div>\n
      \n\n5<\/a> http:\/\/lod-cloud.net\/<\/span>\n\n<\/div>\n
      \n\n6<\/a> http:\/\/bit.ly\/yago-naga<\/span>\n\n<\/div>\n
      \n\n7<\/a> http:\/\/bit.ly\/wikidata-main<\/span>\n\n<\/div>\n
      \n\n8<\/a> http:\/\/www.foaf-project.org\/<\/span>\n\n<\/div>\n
      \n\n9<\/a> http:\/\/activitystrea.ms\/<\/span>\n\n<\/div>\n
      \n\n10<\/a> http:\/\/www.adlnet.gov\/tla\/experience-api<\/span>\n\n<\/div>\n
      \n\n11<\/a> http:\/\/ieeeltsc.org\/wg12LOM\/<\/span>\n\n<\/div>\n
      \n\n12<\/a> http:\/\/www.adlnet.org\/<\/span>\n\n<\/div>\n
      \n\n13<\/a> http:\/\/www.imsglobal.org\/question\/<\/span>\n\n<\/div>\n
      \n\n14<\/a> Avrupa Komisyonu, \"ABYNM: Avrupa Beceriler, Yeterlilikler, Nitelikler ve Meslekler\", https:\/\/ec.europa.eu\/esco<\/span>\n\n<\/div>\n
      \n\n15<\/a> https:\/\/github.com\/adlnet\/xAPI\u2013Spec<\/span>\n\n<\/div>\n
      \n\n16<\/a> http:\/\/www.bnf.fr\/en\/tools\/a.welcome_to_the_bnf.html<\/span>\n\n<\/div>\n
      \n\n17<\/a> http:\/\/datahub.io\/dataset\/l3s-dblp<\/span>\n\n<\/div>\n
      \n\n18<\/a> http:\/\/lov.okfn.org\/dataset\/lov\/<\/span>\n\n<\/div>\n
      \n\n19<\/a> \u00c7evirenin notu: Google\u2019un arama esnas\u0131nda sa\u011f tarafta konu ile ilgili ayr\u0131 bir kutu i\u00e7erisinde getirdi\u011fi bilgi paneli<\/span>\n\n<\/div>\n","rendered":"

      Amal Zouaq1<\/sup>, Jelena Jovanovi\u01072<\/sup>, Sre\u0107ko Joksimovi\u01073<\/sup>, Dragan Ga\u0161evi\u01074<\/sup><\/span><\/span><\/p>\n

      1<\/sup>Elektrik M\u00fchendisli\u011fi ve Bilgisayar Bilimleri Fak\u00fcltesi, Ottawa \u00dcniversitesi, Kanada<\/span><\/span><\/p>\n

      2<\/sup> Yaz\u0131l\u0131m M\u00fchendisli\u011fi B\u00f6l\u00fcm\u00fc, Belgrad \u00dcniversitesi, S\u0131rbistan<\/span><\/span><\/p>\n

      3<\/sup>Moray House E\u011fitim Okulu, Edinburgh \u00dcniversitesi, Birle\u015fik Krall\u0131k<\/span><\/span><\/p>\n

      4<\/sup> Enformatik Okulu, Edinburgh \u00dcniversitesi, Birle\u015fik Krall\u0131k<\/span><\/span><\/p>\n

      DOI: 10.18608\/hla17.030<\/span><\/span><\/p>\n

      \u00d6Z<\/h2>\n

      \u00d6\u011frenme analiti\u011fi (\u00d6A); \u00f6\u011frenme ortamlar\u0131n\u0131n \u00e7e\u015fitlili\u011fi ve farkl\u0131l\u0131\u011f\u0131, kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i dersler (KA\u00c7D) gibi \u00f6l\u00e7eklenebilir \u00f6\u011frenme modellerinin ortaya \u00e7\u0131kmas\u0131 ve sosyal medya platformlar\u0131n\u0131n \u00f6\u011frenme s\u00fcre\u00e7leriyle b\u00fct\u00fcnle\u015ftirilmesi nedeniyle veri \u00fcretiminde art\u0131\u015fa tan\u0131kl\u0131k eder. Bu \u00e7e\u015fitlilik, \u00f6\u011frenme platformlar\u0131n\u0131n birlikte \u00e7al\u0131\u015fabilirli\u011fini, \u00e7oklu bilgi kaynaklar\u0131ndan gelen heterojen verilerin b\u00fct\u00fcnle\u015ftirilmesini ve \u00f6\u011frenme kaynaklar\u0131 ile \u00f6\u011frenme izlerinin i\u00e7erik analizini ilgilendiren bir\u00e7ok sorunu olu\u015fturur. Bu b\u00f6l\u00fcm, veri b\u00fct\u00fcnle\u015ftirme ve analizinin olas\u0131 bir \u00e7er\u00e7evesi olan ba\u011fl\u0131 veri (BV) kullan\u0131m\u0131n\u0131 ele al\u0131r. \u00d6A’daki BV giri\u015fimlerinin ve e\u011fitsel veri madencili\u011finin (EVM) literat\u00fcr taramas\u0131n\u0131 sa\u011flar ve bu alanlarda BV’nin kullan\u0131m\u0131yla ilgili imk\u00e2n ve zorluklardan baz\u0131lar\u0131n\u0131 ele al\u0131r.<\/span><\/p>\n

      Anahtar Kelimeler<\/span>: Ba\u011fl\u0131 veri (BV), veri b\u00fct\u00fcnle\u015ftirme, i\u00e7erik analizi, e\u011fitsel veri madencili\u011fi (EVM)<\/span><\/p>\n

      Kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i derslerin (KA\u00c7D) ortaya \u00e7\u0131k\u0131\u015f\u0131 ve a\u00e7\u0131k veri giri\u015fimi, \u00fcniversite merkezli bir modelden, \u00e7ok platformlu ve \u00e7ok kaynakl\u0131 bir modele ge\u00e7erek e\u011fitimde yeni f\u0131rsatlar\u0131n ortaya \u00e7\u0131kmas\u0131n\u0131 sa\u011flam\u0131\u015ft\u0131r. Asl\u0131nda, g\u00fcn\u00fcm\u00fcz\u00fcn \u00f6\u011frenme ortamlar\u0131 yaln\u0131zca \u00e7e\u015fitli \u00e7evrimi\u00e7i \u00f6\u011frenme platformlar\u0131n\u0131 de\u011fil, ayn\u0131 zamanda \u00f6\u011frenenlerin ba\u011fland\u0131\u011f\u0131, ileti\u015fim kurdu\u011fu, veri ve kaynaklar\u0131 de\u011fi\u015f toku\u015f ettikleri sosyal medya uygulamalar\u0131n\u0131 da (\u00f6r. SlideShare, YouTube, Facebook, Twitter veya Linkedin) kapsar. Hem formal (\u00fcniversite dersleri) hem de informal (sosyal medya, KA\u00c7D) d\u00fczeylerde olan \u00f6\u011frenme art\u0131k \u00e7e\u015fitli bi\u00e7imlerde ve ortamlarda ger\u00e7ekle\u015fmektedir. Bu \u00f6\u011frenen verilerinin \u00e7e\u015fitli platform ve ara\u00e7larda da\u011f\u0131lmas\u0131n\u0131 sa\u011flam\u0131\u015ft\u0131r ve \u00f6\u011frenen verilerini \u00f6\u011frenme ortam\u0131na kapsaml\u0131 bir bak\u0131\u015f a\u00e7\u0131s\u0131 i\u00e7in \u00e7e\u015fitli ortamlarda ba\u011flamak i\u00e7in etkili ara\u00e7lara ihtiya\u00e7 duyulmas\u0131na neden olmu\u015ftur. Platformlar aras\u0131nda veri al\u0131\u015fveri\u015fi ihtiyac\u0131n\u0131n belirgin bir \u00f6rne\u011fi, ba\u011flant\u0131c\u01311<\/sup><\/a> bKA\u00c7D’dir. bKA\u00c7D’lerde \u00f6\u011frenme, tan\u0131m gere\u011fi, tek bir platformda ger\u00e7ekle\u015fmez ancak \u00f6\u011frenenler aras\u0131nda bilgi ve kaynak payla\u015f\u0131m\u0131 i\u00e7in bir dizi \u00f6zel \u00e7evrimi\u00e7i \u00f6\u011frenme uygulamas\u0131na ve sosyal medya ve a\u011f olu\u015fturma uygulamalar\u0131na dayan\u0131r (Siemens, 2005). Bu geli\u015fmeler yeni ihtiya\u00e7lara yol a\u00e7arak hem veri toplamada hem de verinin kullan\u0131m\u0131nda yeni zorluklar\u0131 beraberinde getirmi\u015ftir.<\/p>\n

      Veri toplama a\u00e7\u0131s\u0131ndan bak\u0131ld\u0131\u011f\u0131nda, bulut hizmetlerinin ortaya \u00e7\u0131k\u0131\u015f\u0131 ve \u00f6l\u00e7eklenebilir web mimarilerinin h\u0131zl\u0131 geli\u015fimi, \u00e7e\u015fitli \u00e7evrimi\u00e7i uygulamalardan veri \u00e7ekilmesine ve kullan\u0131lmas\u0131na izin verir. Bu Facebook, Linkedin veya Twitter gibi b\u00fcy\u00fck web payda\u015flar\u0131 ve Coursera ve Udacity gibi KA\u00c7D sa\u011flay\u0131c\u0131lar\u0131 taraf\u0131ndan b\u00fcy\u00fck \u00f6l\u00e7ekli aray\u00fczlerin (API) geli\u015ftirilmesiyle desteklenmektedir. Veri kullan\u0131m\u0131 a\u00e7\u0131s\u0131ndan bak\u0131ld\u0131\u011f\u0131nda, e\u011fitim platformlar\u0131nda ortaya \u00e7\u0131kan kaynaklar\u0131n ve etkile\u015fimlerin \u00e7oklu\u011fu, farkl\u0131 t\u00fcrde verilerin ele al\u0131nmas\u0131n\u0131 da kapsayan analitik yetenekleri gerektirir. \u00dcretilen \u00e7e\u015fitli veri t\u00fcrlerinden baz\u0131lar\u0131 \u00f6\u011frenenlerin \u00f6\u011frenme etkile\u015fimleri ve sosyal platformlarda ger\u00e7ekle\u015ftirdikleri etkile\u015fimlerini (\u00f6\u011frenenlerin g\u00fcnl\u00fckleri\/izleri) tutarken di\u011fer veri t\u00fcrleri ders i\u00e7eriklerinden, \u00f6\u011frenenlerin bloglar\u0131ndan ve tart\u0131\u015fma forumlar\u0131na kadar uzanan yap\u0131land\u0131r\u0131lmam\u0131\u015f i\u00e7erik bi\u00e7imlerini de tutar. Bu \u00e7ok \u00e7e\u015fitli t\u00fcr ve veri kaynaklar\u0131, \u00f6\u011frenenleri ve \u00f6\u011frenme s\u00fcrecini daha iyi anlamak, zaman\u0131nda, bilgilendirici ve uyarlamal\u0131 geri bildirim sa\u011flamak ve hayat boyu \u00f6\u011frenmeyi te\u015fvik etmek i\u00e7in \u00f6\u011frenme analiti\u011fi alan\u0131 ve genel hedefleri i\u00e7in verimli bir zemin sa\u011flar (Gaesvic, Dawson ve Siemens, 2015).<\/p>\n

      Heterojen kaynaklardan gelen verilerin toplanmas\u0131, entegrasyonu ve kullan\u0131m\u0131yla ilgili zorluklar, genellikle e\u011fitim toplulu\u011funda, heterojen verilerin geli\u015ftirilmesine ve onlardan yararlan\u0131lmas\u0131na izin veren standart bir veri modeli geli\u015ftirilerek ele al\u0131nmaktad\u0131r (Dietze vd., 2013). Bu b\u00f6l\u00fcm hem formal hem de informal \u00e7evrimi\u00e7i \u00f6\u011frenme ortamlar\u0131nda b\u00f6yle bir veri modelinin geli\u015ftirilmesine ve kullan\u0131lmas\u0131na y\u00f6nelik potansiyel bir yakla\u015f\u0131m olan ba\u011fl\u0131 veriye (BV) odaklanmaktad\u0131r. \u00d6zellikle, BV ilkelerinin kullan\u0131lmas\u0131 (Bizer, Heath ve Berners \u2013 Lee, 2009), \u00f6\u011frenme ortamlar\u0131nda k\u00fcresel olarak kullan\u0131labilir bir bilgi a\u011f\u0131 kurulmas\u0131na izin vererek (d’Aquin, Adamou ve Dietze, 2013), k\u00fcresel bir e\u011fitim grafi\u011fine \u00f6nc\u00fcl\u00fck eder. Her bir \u00f6\u011frenen i\u00e7in, \u00f6\u011frenme etkinlikleriyle ilgili t\u00fcm veri ve kaynaklar\u0131 birbirine ba\u011flayan, bireysel d\u00fczeyde benzer grafikler olu\u015fturulabilir. Bu t\u00fcr grafiklerin e\u011fitim potansiyelleri ve yararlar\u0131 \u00f6nceden incelenmi\u015f ve ele al\u0131nm\u0131\u015ft\u0131r. \u00d6rne\u011fin, Heath ve Bizer (2011), \u0130ngiliz \u00fcniversitelerinde, \u00f6\u011frenme kaynaklar\u0131n\u0131n i\u00e7eriklerinden elde edilen bilgileri kapsayan bir e\u011fitim grafi\u011fi<\/i><\/span> \u00f6nerirler. Bilgi \u00e7izgelerinin2<\/sup><\/a>, Google, Microsoft ve Facebook gibi artan say\u0131da \u00f6nemli firmalar taraf\u0131ndan geli\u015ftirilmesi ve kullan\u0131lmas\u0131 dikkate al\u0131nd\u0131\u011f\u0131nda bu t\u00fcr \u00e7izgelerin \u00f6\u011frenmeye sa\u011flad\u0131\u011f\u0131 potansiyel ve imk\u00e2nlar incelenmelidir. (Zablith, 2015).<\/p>\n

      Bu b\u00f6l\u00fcm, \u00f6ncelikle \u00f6\u011frenme analiti\u011fi (\u00d6A) \/ e\u011fitsel veri madencili\u011fi (EVM) alan\u0131ndaki mevcut ve potansiyel uygulamalara odaklanarak e\u011fitimde BV kullan\u0131m\u0131n\u0131n mevcut durumunu incelemektedir. Bir sonraki b\u00f6l\u00fcmde BV ilkelerine k\u0131sa bir giri\u015ften sonra, BV potansiyeli iki \u00f6zel boyutta analiz edilir: 1) veri birle\u015ftirmenin boyutu ve 2) veri analizi ve yorumlama boyutu. Son olarak, \u00d6A \/ EVM’de BV kullan\u0131m\u0131 ile ilgili baz\u0131 olas\u0131l\u0131klar\u0131 ve zorluklar\u0131 ele almaktay\u0131z.<\/p>\n

      E\u011e\u0130T\u0130MDE BA\u011eLI VER\u0130LER<\/h2>\n

      Ba\u011fl\u0131 veri, internet \u00fczerinde kaynaklar\u0131 payla\u015fmak i\u00e7in fiil\u00ee bir standart olma potansiyeline sahiptir (Kessler, d’Aquin ve Dietze, 2013). Ba\u011fl\u0131 veri, varl\u0131klar\u0131n kendine \u00f6zg\u00fc tan\u0131mlamalar\u0131n\u0131 yapmak i\u00e7in GBT’leri ve varl\u0131klar\u0131 tan\u0131mlamak i\u00e7in KT\u00c7 veri modeli1<\/span><\/sup> kullanarak bu tan\u0131mlamalar\u0131 ba\u011flant\u0131lar (links) arac\u0131l\u0131\u011f\u0131yla a\u00e7\u0131k bir \u015fekilde tan\u0131mlanm\u0131\u015f anlamlarla ili\u015fkilendirir. Esasen, BV d\u00f6rt ilkeye dayanmaktad\u0131r:<\/p>\n

        \n
      1. \n

        GBT’leri durum isimleri gibi \u00f6rne\u011fin “Paris” adl\u0131 bir tarihi roman\u0131n kendine \u00f6zg\u00fc tan\u0131mlamas\u0131n\u0131 ISBN (bir t\u00fcr GBT) kullanarak yapal\u0131m: 0385535309<\/p>\n<\/li>\n

      2. \n

        HTTP GBT’leri arac\u0131l\u0131\u011f\u0131yla adlar\u0131 arama yetene\u011fi sa\u011flar; bir ISBN benzersiz bir kitab\u0131 tan\u0131mlarken, web \u00fczerinde do\u011frudan eri\u015fim sa\u011flamak i\u00e7in kullan\u0131lamaz, bu y\u00fczden bunun yerine HTTP GBT’leri kullan\u0131lmal\u0131d\u0131r; \u00d6rne\u011fimizde yer alan kitab\u0131m\u0131za \u015fu HTTP GBT’sinden bak\u0131labilir: <http:\/\/www.worldcat.org\/oclc\/827951628><\/span><\/p>\n<\/li>\n

      3. \n

        GBT’ye bakt\u0131ktan sonra, KT\u00c7 ve SPARQL3<\/sup><\/a><\/span> standartlar\u0131n\u0131 kullanarak faydal\u0131 bilgiler d\u00f6nd\u00fcr\u00fcn; \u00f6rne\u011fin, <http:\/\/www.worldcat.org\/oclc\/827951628> GBT’si taraf\u0131ndan tan\u0131mlanan kayna\u011f\u0131n kitap t\u00fcr\u00fcnde<\/i><\/span> oldu\u011funu ve tarihsel kurgu kategorisinde ait oldu\u011funu makineyle i\u015flenebilir bir \u015fekilde ifade edebiliriz: <http:\/\/www.worldcat.org\/oclc\/827951628> rdf:t\u00fcr \u015fema:Kitap; \u015fema:t\u00fcr “Tarihsel kurgu”.<\/i><\/span><\/p>\n<\/li>\n

      4. \n

        Varl\u0131klar\u0131n GBT’leri arac\u0131l\u0131\u011f\u0131yla kendine \u00f6zg\u00fc tan\u0131mlanan di\u011fer varl\u0131klara ili\u015fkin ba\u011flant\u0131lar\u0131 (linkler) d\u00e2hil edin; \u00f6rnek olarak ald\u0131\u011f\u0131m\u0131z kitab\u0131m\u0131z\u0131n yazar\u0131yla ba\u011flant\u0131 kurabiliriz: <http: \/\/ www.worldcat.org\/oclc\/827951628> \u015fema: yazar <http:\/\/viaf.org\/viaf\/34666><\/i><\/span> buradaki GBT yazar Edward Rutherfurd’\u0131 benzersiz olarak tan\u0131mlar.<\/p>\n<\/li>\n<\/ol>\n

        Bu ilkelerin sadeli\u011fi sayesinde, BV m\u00fckemmel bir modelleme \u00e7er\u00e7evesini ve k\u00fcresel boyutta veri sorgulamay\u0131 sunar. \u00c7e\u015fitli uygulamalarda ve alanlarda kullan\u0131labilir ve e\u011fitim toplulu\u011fuyla uzun s\u00fcredir kar\u015f\u0131la\u015f\u0131lan birlikte \u00e7al\u0131\u015fabilirlik ve veri y\u00f6netimi zorluklar\u0131na bir cevap olu\u015fturabilir (Dietze vd., 2013).<\/p>\n

        ]Web \u00fczerinde ba\u011fl\u0131 veri olarak yay\u0131nlanan milyarlarca veri \u00f6gesi; devlete ait veri, bilimsel bilgi ve \u00e7e\u015fitli \u00e7evrimi\u00e7i topluluklar\u0131 i\u00e7eren veriler gibi birka\u00e7 alan ad\u0131ndan gelen bir a\u00e7\u0131k ba\u011fl\u0131 veri bulutunun (BAV)4<\/sup><\/a> yer ald\u0131\u011f\u0131 k\u00fcresel bir a\u00e7\u0131k veri uzay\u0131n\u0131 olu\u015fturur. DBpedia5<\/sup><\/a>, Yago6<\/sup><\/a> ve Wikidata7<\/sup><\/a> gibi BAV’lerde b\u00fcy\u00fck etki alanlar\u0131 aras\u0131 bilgi tabanlar\u0131 da ortaya \u00e7\u0131km\u0131\u015ft\u0131r. Bu nedenle BV, web sa\u011flay\u0131c\u0131lar\u0131 ve web API’leri de d\u00e2hil olmak \u00fczere binbir \u00e7e\u015fit veri eri\u015fim noktalar\u0131 arac\u0131l\u0131\u011f\u0131yla veriye nas\u0131l ula\u015f\u0131laca\u011f\u0131 ve kullan\u0131laca\u011f\u0131 konusunda k\u00fcresel bir de\u011fi\u015fimi m\u00fcmk\u00fcn k\u0131lan potansiyele sahip olup dinamik veri birle\u015ftirmelerinin sorunsuz bir \u015fekilde olu\u015fturulmas\u0131n\u0131 sa\u011flar. (Bizer vd., 2009). Asl\u0131nda, BV’nin g\u00f6ze \u00e7arpan bir \u00f6zelli\u011fi, farkl\u0131 veri kaynaklar\u0131ndan gelen \u00f6geler aras\u0131nda anlam bak\u0131m\u0131ndan zengin ba\u011flant\u0131lar kurmas\u0131 ve b\u00f6ylece daha sorunsuz veri entegrasyonu ve yeniden kullan\u0131m\u0131 i\u00e7in veri silolar\u0131 (\u00f6r. geleneksel veritabanlar\u0131) a\u00e7mas\u0131d\u0131r.<\/p>\n

        T\u00fcm bu potansiyel faydalara ra\u011fmen, BV bi\u00e7imcili\u011fi ve teknolojileri, teknoloji destekli \u00f6\u011frenme alan\u0131nda yava\u015f yava\u015f benimsenmekte olup; BV teknolojilerini kullanan giri\u015fimler son zamanlarda ortaya \u00e7\u0131kmaktad\u0131r (Dietze vd., 2013). \u00d6A \/ EVM alan\u0131nda, 1) kaynak ke\u015ffi (\u00f6r. Ayr\u0131nt\u0131l\u0131 arama) ve i\u00e7erik zenginle\u015ftirme (\u00f6r. BAV veri k\u00fcmelerindeki verilerle i\u00e7eri\u011fi art\u0131rma), (Maturana, Alvarado, Lopez \u2013Sola, Ibanez ve Elosegui, 2013); 2) anlamsal a\u00e7\u0131klamalara dayal\u0131 i\u00e7erik analizi (Joksimovi vd., 2015); 3) kaynak ve hizmet entegrasyonu (Dietze vd., 2012); 4) ki\u015fiselle\u015ftirme (Dietze, Drachsler ve Giordano, 2014) ve 5) EVM sonu\u00e7lar\u0131n\u0131n yorumlanmas\u0131 (d’Aquin vd., 2013) d\u00e2hil olmak \u00fczere BAV’den fayda sa\u011flayacak \u00e7e\u015fitli uygulama senaryolar\u0131 belirleyebiliriz.<\/p>\n

        BA\u011eLI VER\u0130LER\u0130 KULLANARAK VER\u0130 ENTEGRASYONU<\/h2>\n

        BV’nin en belirgin faydalar\u0131ndan biri, veri b\u00fct\u00fcnle\u015ftirme potansiyelinde yatmaktad\u0131r. Bu \u00d6A \/ EVM alan\u0131 i\u00e7in \u00f6zellikle \u00f6nemlidir, \u00e7\u00fcnk\u00fc informal ve hayat boyu \u00f6\u011frenmede kullan\u0131lan \u00e7e\u015fitli kaynaklardan (uygulamalar ve servisler) \u00f6\u011frenen ve i\u00e7erik verilerinin toplanmas\u0131n\u0131 ve y\u00f6netilmesini gerektirir (Santos vd., 2015). \u00d6zellikle, kapsaml\u0131 bir \u00f6\u011frenen modeli olu\u015fturmak i\u00e7in, \u00f6\u011frenenin etkile\u015fime girdi\u011fi farkl\u0131 \u00f6\u011frenme platformlar\u0131na \/ ara\u00e7lar\u0131na kaydedilen \u00f6\u011frenen verilerini birle\u015ftirmesi gerekir (Desmarais ve Baker, 2012). Bu nedenle, \u00e7oklu veri formatlar\u0131n\u0131n ele al\u0131nmas\u0131 ve genel olarak birlikte \u00e7al\u0131\u015fabilirlik eksikli\u011fi ile ilgili zorluklar kilit bir sorun haline gelmektedir (Chatti, Dyckhoff, Schroeder ve Thus, 2012; Duval, 2011). Daha genel olarak, \u00f6\u011frenme platformlar\u0131 aras\u0131nda anlam kayb\u0131 olmadan veri aktar\u0131m\u0131, \u00f6n i\u015fleme, kullan\u0131m, birle\u015ftirme ve analiz kolayl\u0131\u011f\u0131 \u00d6A \/ EVM’nin etkinli\u011fi i\u00e7in \u00f6nemli fakt\u00f6rler haline gelmektedir (Cooper, 2013).<\/p>\n

        Biyomedikal (Belleau, Nolin, Tourigny, Rigault ve Morissette, 2008), farmakoloji (Groth vd., 2014) ve \u00e7evre bilimleri (Lausch, Schmidt ve Tischendorf, 2015) gibi bir\u00e7ok alan veri b\u00fct\u00fcnle\u015ftirme sorunlar\u0131n\u0131 BV’ den faydalanarak \u00e7\u00f6z\u00fcmlemi\u015ftir. T\u00fcm bunlar, BV teknolojilerinin \u00d6A \/ EVM’nin gerektirdi\u011fi sa\u011flam veri b\u00fct\u00fcnle\u015ftirme katman\u0131na imk\u00e2n verebilece\u011fini ortaya koymaktad\u0131r.<\/p>\n

        E\u011fitsel Topluluklarda Veri B\u00fct\u00fcnle\u015ftirmenin \u00d6nceki Giri\u015fimleri<\/h3>\n

        Teknoloji destekli \u00f6\u011frenme ara\u015ft\u0131rma toplulu\u011fu uzun zamand\u0131r \u00e7oklu standartla\u015ft\u0131rma \u00e7abalar\u0131yla sonu\u00e7lanan veri entegrasyonunun \u00f6nemini kabul etmi\u015ftir. Cooper (2013), \u00f6\u011frenmeyle ilgili \u00e7e\u015fitli standartlara de\u011ferli bir genel bak\u0131\u015f a\u00e7\u0131s\u0131 sunar. Temel olarak bu standartlar, e\u011fitim i\u00e7eriklerinin ve sa\u011flay\u0131c\u0131lar\u0131n\u0131n yan\u0131 s\u0131ra \u00f6\u011frenenlerin ve onlar\u0131n etkinliklerini i\u00e7eren verilerin temsili ile ilgilidir.<\/p>\n

        \u00d6\u011frenci d\u00fczeyinde, standartlar bireyler ve ge\u00e7mi\u015fleri ile ilgili ger\u00e7eklere, bunlar\u0131n di\u011fer insanlarla olan ba\u011flant\u0131lar\u0131na ve etkile\u015fimlerine ve \u00f6\u011frenme ortamlar\u0131 taraf\u0131ndan sunulan kaynaklarla etkile\u015fime odaklan\u0131r (ki\u015fi ve \u00f6\u011frenme etkinlikleri boyutlar\u0131). Model \u00f6\u011frenenler (\u00f6r. FOAF8<\/sup><\/a>) ve \u00f6\u011frenenlerin etkinlikleri ve etkile\u015fimleri i\u00e7in \u00e7e\u015fitli belirtimler mevcuttur (\u00f6r. Ba\u011flamsalla\u015ft\u0131r\u0131lm\u0131\u015f \u0130lgi \u00dcst Verisi [Schmitz, Wolpers, Kirschenmann ve Niemann, 2011], Etkinlik Ak\u0131\u015flar\u01319<\/sup><\/a> veya ADL xAPI10<\/sup><\/a>).<\/p>\n

        \u0130\u00e7erik d\u00fczeyinde, IEEE \u00d6\u011frenme Nesnesi \u00dcst Verisi (LO\u00dcV11<\/sup><\/a>) ve ADL SCORM12<\/sup><\/a> gibi \u00f6nceki giri\u015fimler, \u00e7evrimi\u00e7i e\u011fitim kaynaklar\u0131n\u0131n tan\u0131m\u0131n\u0131 veya bilgisayar destekli de\u011ferlendirmenin belirtimini (\u00f6r. IMS QTI13<\/sup><\/a>) birle\u015ftiren kelimeler ve standartlar olu\u015fturmaya \u00e7al\u0131\u015ft\u0131. Di\u011fer \u00e7abalar, sosyal ve etkile\u015fim verileri k\u00fcmelerini bir araya getirmeyi ama\u00e7layan Niemann, Wolpers, Stoitsis, Chinis ve Manouselis (2013) \u00e7al\u0131\u015fmalar\u0131 gibi \u00e7e\u015fitli veri modelleri aras\u0131ndaki haritalamay\u0131 hedeflemi\u015ftir. Son olarak, bu standartlarla tutarl\u0131 olan hizmetlerin uygulanmas\u0131na y\u00f6nelik k\u0131lavuz sa\u011flamak i\u00e7in \u00e7e\u015fitli aray\u00fczler \u00f6nerilmi\u015ftir (Dietze vd., 2013).<\/p>\n

        Farkl\u0131 bak\u0131\u015f a\u00e7\u0131lar\u0131na dayanan bu \u00e7abalar birbirleriyle rekabet eden bir\u00e7ok proje ortaya \u00e7\u0131karm\u0131\u015ft\u0131r ve dolay\u0131s\u0131yla aralar\u0131nda pek birlikte \u00e7al\u0131\u015fabilirlik olmayan \u00e7e\u015fitli teknolojilere, dillere ve modellere y\u00f6nelik alt topluluklar olu\u015fturmu\u015ftur. BV felsefesi, birlikte \u00e7al\u0131\u015fabilirlik sorunlar\u0131na webde \u00e7ok say\u0131da modelin izin vermesi ve bu modellerin webde eri\u015filebilir semantik ba\u011flant\u0131lar kullanarak k\u00f6pr\u00fclenmesi yoluyla bir \u00e7\u00f6z\u00fcm sa\u011flar. Bu nedenle, farkl\u0131 olarak temsil edilen anlamsal olarak benzer modeller, farkl\u0131 modellerden kaynaklanan kavramlar aras\u0131nda anlaml\u0131 ba\u011flant\u0131lar kuran yaz\u0131l\u0131 linkler kullan\u0131larak ayn\u0131 hizada olabilir; \u00f6rne\u011fin, e\u015fitlik ba\u011flant\u0131lar\u0131 (owl:benzerOlan)<\/i><\/span> veya hiyerar\u015fik ba\u011flant\u0131lar (rdfs:ninAltgrubu veya skos:broader).<\/i><\/span><\/p>\n

        Ba\u011fl\u0131 Veri Kullanarak Mevcut Veri B\u00fct\u00fcnle\u015ftirme Giri\u015fimleri<\/h3>\n

        BV’ye dayal\u0131 entegrasyon, belirli konu alanlar\u0131ndaki varl\u0131k t\u00fcrlerini, varl\u0131klar\u0131n \u00f6zniteliklerini ve varl\u0131klar aras\u0131ndaki ba\u011flant\u0131 t\u00fcrlerini tan\u0131mlayan web eri\u015fimli s\u00f6zl\u00fcklerinin kullan\u0131labilirli\u011fini gerektirir. Ayr\u0131ca, belirli bir g\u00f6rev i\u00e7in birden fazla veri k\u00fcmesinden yararlan\u0131lmas\u0131na izin veren hizmetlerin kullan\u0131labilirli\u011finin yan\u0131 s\u0131ra verileri BV olarak a\u00e7\u0131\u011fa \u00e7\u0131karan hizmetlere de ba\u011fl\u0131d\u0131r. Bu b\u00f6l\u00fcm, e\u011fitim alan\u0131ndaki mevcut baz\u0131 s\u00f6z varl\u0131\u011f\u0131n\u0131 ve e\u011fitsel verileri \u00f6\u011frenme verisi olarak g\u00f6sterme \u00e7abalar\u0131n\u0131 tan\u0131tmaktad\u0131r. E\u011fitimle ilgili kelimelerin daha kapsaml\u0131 bir incelemesi Dietze vd. (2014). Bu b\u00f6l\u00fcmde ayr\u0131ca birden fazla BV veri k\u00fcmesinin entegrasyonunu kullanan hizmetlere \u00f6rnekler verilmi\u015ftir.<\/p>\n

        Gittik\u00e7e artan say\u0131da e\u011fitim kurumu, Birle\u015fik Krall\u0131k’taki A\u00e7\u0131k \u00dcniversitedeki gibi BV ilkelerini izleyerek verilerini yay\u0131mlamaktad\u0131r ve BV’yi, dersleri ayr\u0131nt\u0131l\u0131 ve yeniden kullan\u0131labilir bir \u015fekilde birbirine ba\u011flamak i\u00e7in y\u00fcksek\u00f6\u011frenim programlar\u0131 etraf\u0131nda kavramsal bir katman olarak kullanmay\u0131 \u00f6nermi\u015ftir. Ba\u015fka bir \u00e7al\u0131\u015fma da ABYNM14<\/sup><\/a> tabanl\u0131 becerileri zenginle\u015ftirilmi\u015f \u00f6z ge\u00e7mi\u015fler olu\u015fturmak i\u00e7in KA\u00c7D kurs a\u00e7\u0131klamalar\u0131na ba\u011flamaktad\u0131r (Zotou, Papantoniou, Kremer, Peristeras ve Tambouris, 2014). \u0130lgin\u00e7 bir \u015fekilde, yazarlar Coursera ve Udacity KA\u00c7D platformlar\u0131nda \u00f6\u011fretilen benzer becerileri tan\u0131mlayabiliyorlar, b\u00f6ylece iki farkl\u0131 KA\u00c7D platformunun kurslar\u0131 aras\u0131nda \u00f6rt\u00fcl\u00fc ba\u011flant\u0131lar sa\u011fl\u0131yorlar. Birisi, platformlar aras\u0131 bir KA\u00c7D kurs \u00f6neri servisine dayanan hayat boyu \u00f6\u011frenme i\u00e7in heyecan verici f\u0131rsatlar \u00f6ng\u00f6rebilir.<\/p>\n

        Genel olarak e\u011fitim alan\u0131nda BV’nin ve \u00f6zellikle de \u00d6A \/ EVM’nin artan \u00f6neminin bir ba\u015fka g\u00f6stergesi, BV kavramlar\u0131n\u0131n ve teknolojilerinin xAPI belirtimine15<\/sup><\/a> uyarlanmas\u0131d\u0131r. XAPI ile, geli\u015ftiriciler \u00f6nceden tan\u0131mlanm\u0131\u015f bir arabirim ve bir dizi depolama ve alma kurallar\u0131 arac\u0131l\u0131\u011f\u0131yla bir \u00f6\u011frenme deneyimi izleme hizmeti olu\u015fturabilir. De Nies, Salliau, Verborgh, Mannens ve Van de Walle (2015), xAPI belirtimini BV olarak kullanarak olu\u015fturulan veri modellerini g\u00f6stermeyi \u00f6nermektedir. Bu \u00f6neri, birlikte \u00e7al\u0131\u015fabilir bir \u00f6\u011frenme izleri verisi modeli sa\u011flar ve \u00f6\u011frenenlerin izlerinin semantik olarak birlikte \u00e7al\u0131\u015fabilir BV olarak kusursuz bir \u015fekilde ortaya \u00e7\u0131kmas\u0131n\u0131 sa\u011flar. Benzer \u015fekilde, Softic vd. (2014), Semantik Web teknolojilerinin (KT\u00c7, SPARQL) ki\u015fisel \u00f6\u011frenme ortamlar\u0131ndaki \u00f6\u011frenen g\u00fcnl\u00fcklerini modellemek i\u00e7in kullan\u0131m\u0131n\u0131 rapor etmektedir.<\/p>\n

        Temel altyap\u0131 olarak webin \u00f6l\u00e7eklenebilirli\u011fine ve W3C standartlar\u0131 KT\u00c7 ve SPARQL’in birlikte \u00e7al\u0131\u015fabilirli\u011fine dayanarak, benzer giri\u015fimlerin ademi merkeziyet\u00e7i ve uyarlanabilir \u00f6\u011frenme hizmetlerinin geli\u015ftirilmesine daha fazla katk\u0131da bulunabilece\u011fine inan\u0131yoruz.<\/p>\n

        BA\u011eLI VER\u0130 KULLANARAK VER\u0130LER\u0130N ANAL\u0130Z ED\u0130LMES\u0130 VE ANLAMLANDIRILMASI<\/h2>\n

        KA\u00c7D’lerde giderek artan miktardaki \u00f6\u011frenme i\u00e7eriklerinin yan\u0131 s\u0131ra \u00e7e\u015fitli \u00e7evrimi\u00e7i sosyal medya ve ileti\u015fim kanallar\u0131nda yer alan yap\u0131land\u0131r\u0131lmam\u0131\u015f i\u00e7eriklerin h\u0131zl\u0131 art\u0131\u015f\u0131 da dikkate al\u0131nmal\u0131d\u0131r. Birka\u00e7 isim vermek gerekirse ba\u015fl\u0131klar (konular), e\u011filimler (trendler) ve d\u00fc\u015f\u00fcnceler gibi uzaktan \u00f6\u011frenimle ilgili \u00f6gelerin ortaya \u00e7\u0131kar\u0131lmas\u0131n\u0131n otomatikle\u015ftirilmesine ihtiya\u00e7 duyulmaktad\u0131r. E\u011fer d\u00fczenli girdi verileri (\u00f6r. \u00f6\u011frenenlerin g\u00fcnl\u00fckleri), BAV veri k\u00fcmelerinden gelen (\u00f6r. kursla ili\u015fkilendirilen konularla ilgili veriler) gerekli bilgilerle zenginle\u015ftirilirse analitik i\u00e7in ihtiya\u00e7 duyulan ilgili \u00f6gelerin ortaya \u00e7\u0131kar\u0131lmas\u0131 ve \/ veya \u00f6nerilmesi iyile\u015ftirilebilir (d’Aquin ve Jay, 2013). DBpedia ve Yago gibi BAV etki alanlar\u0131 aras\u0131 bilgi tabanlar\u0131n\u0131n tek veya geleneksel i\u00e7erik analizi teknikleriyle (\u00f6r. sosyal a\u011f analizi, metin madencili\u011fi, \u00f6rt\u00fck anlamsal dizinleme) ile birlikte kullan\u0131lmas\u0131 halinde, \u00e7ekilen konular\u0131n daha kolay yorumlanmas\u0131na imk\u00e2n veren i\u00e7erik analizi bilgi tabanlar\u0131n\u0131n (\u00f6r. DBpedia) geli\u015ftirilmesi i\u00e7in umut verici bir yol sunar.<\/p>\n

        \u00d6A \/ EVM Alan\u0131ndaki Bilimsel Yay\u0131nlar\u0131n Analizi<\/h3>\n

        BV taraf\u0131ndan desteklenen ve e\u011fitim ba\u011flam\u0131yla ilgili bir di\u011fer uygulama alan\u0131, semantik yay\u0131nlama (\u00f6r. kitap kataloglar\u0131n\u0131 BV olarak yay\u0131nlama) ve bilimsel yay\u0131nlar\u0131n \u00fcst analizidir. Asl\u0131nda, BV teknolojilerinin temel ba\u015far\u0131lar\u0131ndan biri, BNF16<\/sup><\/a> gibi \u00e7e\u015fitli i\u00e7erik yay\u0131nc\u0131lar\u0131 ve DBLP17<\/sup><\/a> gibi bilimsel tabanl\u0131 yay\u0131nc\u0131l\u0131k giri\u015fimleri taraf\u0131ndan erken benimsenmi\u015f olmas\u0131d\u0131r. Bu bilimsel yay\u0131nlarla ilgili BAV kelimelerinin ve veri k\u00fcmelerinin \u00e7oklu\u011funa yol a\u00e7m\u0131\u015ft\u0131r. Bu veri k\u00fcmeleri; belirli bir alana ait konular\u0131 belirleyen, ara\u015ft\u0131rmac\u0131lar\u0131 etkileyen ve ara\u015ft\u0131rma toplulu\u011funu ayr\u0131nt\u0131l\u0131 olarak tan\u0131mlayan \u00e7e\u015fitli bilimetrik hesaplamalar i\u00e7in temel olu\u015fturur (Mirriahi, Ga\u0161evi\u0107, Dawson ve Long, 2014; Ochoa, Suthers, Verbert ve Duval, 2014). Ayr\u0131ca, ilgili bilgileri bulmak i\u00e7in e\u011fitim sekt\u00f6r\u00fcnden gelen uzmanlara (ara\u015ft\u0131rmac\u0131lar, \u00f6\u011frenciler, k\u00fct\u00fcphaneciler, kurs \u00fcreticileri) an\u0131nda yard\u0131mc\u0131 olurlar.<\/p>\n

        \u00d6A \/ EVM alan\u0131nda, \u00d6\u011frenme Analiti\u011fi ve Bilgi (\u00d6AB) Veri K\u00fcmesi (Taibi ve Dietze, 2013), \u00d6A \/ EVM topluluklar\u0131ndan bir yay\u0131n derlemi temsil eder. \u00d6AB Veri K\u00fcmesi hem yay\u0131nlar\u0131n i\u00e7eri\u011fini hem de \u00fcst verilerini (\u00f6r. anahtar kelimeler, yazarlar, konferans) i\u00e7erir. \u00d6AB Veri K\u00fcmesi, var olan \u00e7e\u015fitli BAV kelime bilgilerine dayand\u0131\u011f\u0131 ve BV teknolojilerinin ba\u015far\u0131l\u0131 bir uygulamas\u0131n\u0131 kapsad\u0131\u011f\u0131ndan veri b\u00fct\u00fcnle\u015ftirme giri\u015fimini ifade eder. \u00d6AB Veri K\u00fcmesinin analizi, 2013’ten bu yana, \u00d6A \/ EVM yay\u0131nlar\u0131yla ilgili ara\u015ft\u0131rma ve analitik \u00e7al\u0131\u015fmalar\u0131 te\u015fvik eden y\u0131ll\u0131k \u00d6AB Veri M\u00fccadelesi ile te\u015fvik edilmi\u015ftir. Bu veri k\u00fcmesi, veri analiti\u011fi ve i\u00e7erik analizi uygulamalar\u0131n\u0131n geli\u015ftirilmesi i\u00e7in daha fazla kullan\u0131lm\u0131\u015ft\u0131r. \u00d6zellikle, konular\u0131n ve veri k\u00fcmesindeki konular aras\u0131ndaki ili\u015fkilerin y\u0131ll\u0131k olarak, topluluk ba\u015f\u0131na (\u00d6A-EVM), yay\u0131n ba\u015f\u0131na ve ayr\u0131nt\u0131l\u0131 olarak saptanmas\u0131 de\u011ferli bir uygulamad\u0131r. \u00d6rne\u011fin, Zouaq, Joksimovic ve Ga\u0161evi\u0107’in (2013) \u00e7al\u0131\u015fmas\u0131, g\u00f6ze \u00e7arpan konular\u0131 ve aralar\u0131ndaki ili\u015fkileri belirlemek i\u00e7in \u00d6AB Veri K\u00fcmesinde ontoloji \u00f6\u011frenme tekniklerini kullanm\u0131\u015ft\u0131r. Konular\u0131 ke\u015ffetmek i\u00e7in uygulanan di\u011fer teknikler aras\u0131nda gizli Dirichlet tahsisi (GDT; Sharkey ve Ansari, 2014) ve k\u00fcmelemesi (Scheffel, Niemann, Leon Rojas, Drachsler ve Specht, 2014) say\u0131labilir. Bu yakla\u015f\u0131mlar metne dayal\u0131 bir i\u00e7erik analizi sunarken, di\u011fer konular da veri entegrasyon \u00e7abalar\u0131nda BAV bilgi tabanlar\u0131na (\u00f6r. DBpedia) ve ilgi alanlar\u0131n\u0131 tan\u0131mlamak i\u00e7in semantik a\u00e7\u0131klamalara dayanarak daha da ileri gitmi\u015ftir. \u00d6rne\u011fin; Miliki\u0107, Krcadinac, Jovanovi\u0107, Brankov, ve Keca (2013) ve Nunes, Fetahu ve Casanova (2013), yay\u0131nlarda konu ba\u015fl\u0131klar\u0131n\u0131 ve adland\u0131r\u0131lm\u0131\u015f varl\u0131klar\u0131 tan\u0131mlamak i\u00e7in s\u0131ras\u0131yla TagMe ve DBpedia Spotlight hizmetlerine g\u00fcvendi. Bu durumda BV’nin yarar\u0131 1) veri k\u00fcmesini BAV kavramlar\u0131, anahtar kelimeler ve temalarla zenginle\u015ftirme kabiliyeti ve 2) potansiyel i\u015fbirlikli tespiti (Hu vd., 2014), veri k\u00fcmesi \u00f6nerileri veya daha genel anlamsal aramalar gibi geli\u015fmi\u015f hizmetler geli\u015ftirme kabiliyeti taraf\u0131ndan vurgulanm\u0131\u015ft\u0131r (Nunes vd., 2013).<\/p>\n

        Veri Madencili\u011fi Sonu\u00e7lar\u0131n\u0131n Yorumlanmas\u0131<\/h3>\n

        Bir\u00e7ok ara\u015ft\u0131rma, \u00f6\u011frenenlerin etkile\u015fimlerini ve g\u00f6r\u00fc\u015fme verilerini (\u00f6r. \u00f6\u011frenenlerin konu\u015fmalar\u0131, fikirleri ve akademik performanslar\u0131 aras\u0131ndaki ba\u011flant\u0131y\u0131 belirleyerek (Dowell vd., 2015) veya kurs kay\u0131t verilerini (d’Aquin ve Jay, 2013) analiz ederek anla\u015f\u0131lmas\u0131n\u0131, ba\u011flant\u0131lar\u0131 ve kestirimci modelleri sunmu\u015ftur. Bununla birlikte, bu analizlerin \u00e7o\u011fu kapal\u0131 ya da silo veri k\u00fcmesiyle s\u0131n\u0131rl\u0131 kalmaktad\u0131r ve b\u00fcy\u00fck veri k\u00fcmelerinde yorumlanmalar\u0131 genellikle zordur.<\/p>\n

        Genel olarak, \u00d6A \/ EVM’deki model ke\u015ffi, sonu\u00e7lar\u0131n birka\u00e7 boyuta (\u00f6r. konular, \u00f6\u011frenci \u00f6zellikleri, \u00f6\u011frenme ortamlar\u0131, vb.) g\u00f6re anlaml\u0131 bir \u015fekilde yorumlanmas\u0131 i\u00e7in bir model ve insan analisti gerektirir (d’Aquin ve Jay, 2013). D’Aquin ve Jay (2013) taraf\u0131ndan yap\u0131lan \u00e7al\u0131\u015fma, veri madencili\u011fi s\u00fcrecinde ke\u015ffedilen kal\u0131plar\u0131 zenginle\u015ftirmek ve ba\u011flamla\u015ft\u0131rmak i\u00e7in BV’nin faydas\u0131 hakk\u0131nda yeni bilgiler sunmaktad\u0131r. \u00d6zellikle, ke\u015ffedilen \u00f6r\u00fcnt\u00fcleri BV GBT’leri ile birlikte a\u00e7\u0131klamay\u0131 teklif ederler, b\u00f6ylece bu \u00f6r\u00fcnt\u00fclerin yorumlamay\u0131 kolayla\u015ft\u0131rmak i\u00e7in mevcut veri k\u00fcmeleriyle daha da zenginle\u015ftirilebilir. Yazarlar, fikri zamanla ders mod\u00fcllerine \u00f6\u011frenci kayd\u0131yla ilgili bir \u00f6rnek olay incelemesi ile g\u00f6stermektedir. S\u0131k s\u0131k kurs dizilimlerini \u00e7\u0131kar\u0131rlar ve bunlar\u0131 kurs GBT’leri vas\u0131tas\u0131yla kurs a\u00e7\u0131klamalar\u0131yla, yani kursu tan\u0131mlayan bir dizi \u00f6zellikle ili\u015fkilendirerek zenginle\u015ftirirler. (Zincir) \u00f6zellikleri, \u00e7apraz temelli bir s\u0131n\u0131fland\u0131rmada (\u00f6r. kurs s\u0131ralar\u0131n\u0131n s\u0131kl\u0131kla tercih edilen ortak konular\u0131) ve navigasyon tabanl\u0131 bir yap\u0131 olarak kullan\u0131lan analitik boyutlar\u0131 sa\u011flar. Bu vaka \u00e7al\u0131\u015fmas\u0131nda g\u00f6sterildi\u011fi gibi, BV ke\u015ffedilen kal\u0131plar\u0131 d\u0131\u015f bilgi tabanlar\u0131na ba\u011flayarak ve yeni bilgiyi ortaya \u00e7\u0131karmak i\u00e7in BAV anlamsal ba\u011flant\u0131lar\u0131n\u0131 kullanarak yeni analitik boyutlar\u0131 ke\u015ffetmeye yard\u0131mc\u0131 olabilir. Bu \u00f6zellikle \u00e7e\u015fitli fakt\u00f6rlerin bir kal\u0131ba veya olguya katk\u0131da bulunabilece\u011fi \u00e7ok disiplinli ara\u015ft\u0131rmalarda \u00f6nemlidir. \u00d6\u011frenme davran\u0131\u015flar\u0131n\u0131n karma\u015f\u0131kl\u0131\u011f\u0131 g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, bu deste\u011fin \u00d6A \/ EVM sonu\u00e7lar\u0131n\u0131n yorumlanmas\u0131nda faydas\u0131 oldu\u011fu d\u00fc\u015f\u00fcn\u00fclebilir.<\/p>\n

        TARTI\u015eMA ve BAKI\u015e A\u00c7ISI<\/h2>\n

        \u00d6\u011frenme deneyimine y\u00f6nelik genel analitik yakla\u015f\u0131m, \u00f6\u011frenme verilerinin toplanmas\u0131, y\u00f6netimi, sorgulanmas\u0131, birle\u015ftirilmesi ve zenginle\u015ftirilmesi i\u00e7in son teknoloji veri y\u00f6netimi teknikleri gerektirir. BV kavram\u0131 ve teknolojisi, W3C standartlar\u0131na (KT\u00c7, SPARQL) dayanmaktad\u0131r ve veri y\u00f6netiminin t\u00fcm bu y\u00f6nlerine katk\u0131da bulunma potansiyeline sahiptir. \u0130lk olarak, BV teknolojilerinin ard\u0131ndaki temel ama\u00e7lardan biri, verilerin anlam bilimini koruyarak ve kald\u0131rarak, \u00e7e\u015fitli ama\u00e7lar i\u00e7in verileri kolayca i\u015flenebilir ve tekrar kullan\u0131labilir hale getirmektir. \u0130kincisi, BV, \u00e7e\u015fitli veri k\u00fcmelerinin sorunsuz bir \u015fekilde birle\u015ftirilmesini ve sorgulanmas\u0131n\u0131 sa\u011flayarak veri y\u00f6netimine merkezi olmayan bir yakla\u015f\u0131m sa\u011flar. \u00dc\u00e7\u00fcnc\u00fcs\u00fc, web \u00fczerindeki ba\u011flant\u0131l\u0131 a\u00e7\u0131k veriler olarak mevcut b\u00fcy\u00fck \u00f6l\u00e7ekli bilgi tabanlar\u0131, analitik s\u00fcre\u00e7le ilgili \u00e7e\u015fitli hizmetler i\u00e7in zemin sa\u011flar; \u00f6rne\u011fin, i\u00e7erik analizi ve zenginle\u015ftirme i\u00e7in semantik ek a\u00e7\u0131klamalar. D\u00f6rd\u00fcnc\u00fcs\u00fc, webde BV olarak a\u00e7\u0131\u011fa \u00e7\u0131kar\u0131lan veriler, analitik s\u00fcrecin farkl\u0131 a\u015famalar\u0131nda gerekli olan talep \u00fczerine (tam zaman\u0131nda) veri \/ bilgi giri\u015fi sa\u011flayabilir, \u00e7\u00fcnk\u00fc bu bilgi daima \u00f6nceden tam olarak tahmin edilemez. Potansiyel faydalar ayr\u0131ca sonu\u00e7 analizini anlamsal a\u00e7\u0131dan zengin bir formatta temsil etmeyi de i\u00e7erir, b\u00f6ylece sonu\u00e7lar uygulamalar aras\u0131nda payla\u015f\u0131labilir ve ilgili taraflara (\u00f6\u011fretmenler, \u00f6\u011frenciler) ihtiya\u00e7 ve tercihlere ba\u011fl\u0131 olarak (\u00f6r. farkl\u0131 g\u00f6rsel veya anlat\u0131m bi\u00e7imlerine ba\u011fl\u0131 olarak) farkl\u0131 \u015fekillerde iletilebilir. Di\u011fer taraftan, veri maddelerinin anlamsal olarak zengin temsilinden ve kar\u015f\u0131l\u0131kl\u0131 ili\u015fkilerinden kaynaklanan \u00e7oklu veri kaynaklar\u0131 \u00fczerindeki \u00e7\u0131kar\u0131m yetenekleri sayesinde, BV tabanl\u0131 y\u00f6ntemler, metin i\u00e7eri\u011findeki temalar\u0131 ve konular\u0131 ke\u015ffetmek i\u00e7in mevcut analitik y\u00f6ntemlere amaca uygun bir katk\u0131 olabilir. Ayr\u0131ca, \u00e7oklu veri kaynaklar\u0131 \u00fczerindeki \u00e7\u0131kar\u0131m yetenekleri sayesinde, veri \u00f6gelerinin anlamsal olarak zengin temsilinden ve kar\u015f\u0131l\u0131kl\u0131 ili\u015fkilerinden kaynaklanan BV tabanl\u0131 y\u00f6ntemler, metin i\u00e7eri\u011findeki temalar\u0131 ve konular\u0131 ke\u015ffetmek i\u00e7in mevcut analitik y\u00f6ntemlere alakal\u0131 bir ek olabilir. Daha genel olarak, \u00d6A \/ EVM toplulu\u011funda istatistiksel ve makine \u00f6\u011frenme y\u00f6ntemleri yayg\u0131n olmakla birlikte, verilerin a\u00e7\u0131k\u00e7a tan\u0131mlanm\u0131\u015f anlam bilimine ve a\u00e7\u0131k bilgi kaynaklar\u0131na (\u00f6zellikle a\u00e7\u0131k, web tabanl\u0131 bilgiye dayal\u0131) dayanan di\u011fer veri analiz y\u00f6ntemleri ve teknikleri geleneksel analitik yakla\u015f\u0131mlar\u0131 daha da g\u00fc\u00e7l\u00fc k\u0131lar.<\/p>\n

        Son olarak<\/span>, BV teknolojileri \u00f6\u011frenme ortamlar\u0131n\u0131n ve sosyal medya platformlar\u0131n\u0131n heterojenli\u011fi ile ba\u015fa \u00e7\u0131kmada yararl\u0131 olabilir. \u00d6zellikle, ortak bir \u015fema payla\u015fmayan \u00e7e\u015fitli veri k\u00fcmelerini sorgulayabilir ve birle\u015ftirebilirsiniz. Bu y\u00f6n\u00fcn kendisi, ortak bir modele \/ \u015femaya uyumu gerektiren \u00f6nceki yakla\u015f\u0131mlardan daha esnek ve pratik bir yakla\u015f\u0131m\u0131 temsil etmektedir.<\/p>\n

        Ancak BV Kullan\u0131m\u0131 ile ilgili olarak a\u015fa\u011f\u0131da zikretti\u011fimiz baz\u0131 zorluklardan bahsetmek m\u00fcmk\u00fcnd\u00fcr:<\/p>\n

          \n
        1. \n

          Kalite<\/span>: BAV veri k\u00fcmelerinin kalitesi bir endi\u015fe kayna\u011f\u0131d\u0131r (Kontokostas vd., 2014) ve \u00f6\u011frenme kaynaklar\u0131n\u0131 ve izlerini d\u0131\u015f veri k\u00fcmelerine ve bilgi tabanlar\u0131na ba\u011flamak karma\u015f\u0131k veriye yol a\u00e7abilir. Veri temizli\u011fi i\u00e7in baz\u0131 giri\u015fimler olmas\u0131na ra\u011fmen, bu sorun hala tam olarak \u00e7\u00f6z\u00fclememi\u015ftir.<\/p>\n<\/li>\n

        2. \n

          Hizalama<\/span>: \u015eemalar aras\u0131nda ortak web GBT’lerinin kullan\u0131m\u0131n\u0131n yan\u0131 s\u0131ra, zorlu bir g\u00f6rev olan kelime ve modelleri anlamsal olarak hizalamaya ihtiya\u00e7 duyulur. Mevcut uyum yakla\u015f\u0131mlar\u0131 genellikle belirsizlikler ile ba\u015fa \u00e7\u0131kmayan s\u00f6zdizimsel e\u015fle\u015fmeye dayanmaktad\u0131r. Uyum sorununu hafifletmenin bir yolu, m\u00fcmk\u00fcn oldu\u011funda ana BV kelimelerinin18<\/sup><\/a><\/span> fark\u0131nda olmak ve yeniden kullanmakt\u0131r (\u00f6r. foaf:isim<\/i><\/span>, FOAF belirtimindeki bir ki\u015finin ad\u0131n\u0131 g\u00f6steren ve yeni bir \u00f6zellik olu\u015fturmak yerine kullan\u0131labilir) ;<\/p>\n<\/li>\n

        3. \n

          Gizlilik<\/span>: KA\u00c7D’lerdeki ve \u00f6\u011frenme platformlar\u0131ndaki veriler \u00e7o\u011fu zaman gizlilik nedeniyle depolanmaktad\u0131r. Bilgiyi \u00f6\u011frenme ile sosyal platformlar aras\u0131nda birle\u015ftirmek, \u00f6\u011frenenlere verilere eri\u015fim izni ve \u00f6\u011frenmede kulland\u0131klar\u0131 farkl\u0131 hizmetler i\u00e7in giri\u015f bilgileri vermelerini gerektirir.<\/p>\n<\/li>\n<\/ol>\n

          Yukar\u0131da belirtilen ve BAV veri k\u00fcmelerinin ve bilgi tabanlar\u0131n\u0131n Google bilgi \u00e7izgesi19<\/sup><\/a> veya Facebook bilgi \u00e7izgesi aramas\u0131 gibi baz\u0131 b\u00fcy\u00fck firmalar\u0131n kullan\u0131lmas\u0131 ve bunlar\u0131n e\u011fitim kurumlar\u0131nda benimsenmesi artan BV’de kullan\u0131lmas\u0131na ra\u011fmen, bug\u00fcn\u00fcn \u00f6\u011frenme platformlar\u0131 i\u00e7in umut verici bir teknolojik belkemi\u011fidir. Ayr\u0131ca, ham veri toplama ve depolanmas\u0131ndan veri kullan\u0131m\u0131na ve zenginle\u015ftirmeye, analitik sonu\u00e7lar\u0131n yorumlanmas\u0131na kadar genel \u00f6\u011frenme analiti\u011fi s\u00fcrecini kolayla\u015ft\u0131rmak i\u00e7in yararl\u0131 bir bi\u00e7imcilik sa\u011flar.<\/p>\n

          KAYNAK\u00c7A<\/h2>\n

          Belleau, F., Nolin, M.-A., Tourigny, N., Rigault, P., & Morissette, J. (2008). Bio2RDF: Towards a mashup to build bioinformatics knowledge systems. Journal of Biomedical Informatics, 41<\/i>(5), 706\u2013716. <\/span><\/p>\n

          Bizer, C., Heath, T., & Berners-Lee, T. (2009). Linked data – the story so far. International Journal on Semantic Web and Information Systems, 5<\/i>(3), 1\u201322. Preprint retrieved from http:\/\/tomheath.com\/papers\/bizer-heath-berners-lee-ijswis-linked-data.pdf <\/span><\/p>\n

          Chatti, M. A., Dyckhoff, A. L., Schroeder, U., & Th\u00fcs, H. (2012). A reference model for learning analytics. International Journal of Technology Enhanced Learning, 4<\/i>(5\u20136), 318\u2013331. <\/span><\/p>\n

          Cooper, A. R. (2013). Learning analytics interoperability: A survey of current literature and candidate standards. http:\/\/blogs.cetis.ac.uk\/adam\/2013\/05\/03\/learning-analytics-interoperability <\/span><\/p>\n

          d’Aquin, M., & Jay, N. (2013). Interpreting data mining results with linked data for learning analytics: Motivation, case study and directions. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 155\u2013164). New York: ACM. <\/span><\/p>\n

          d’Aquin, M., Adamou, A., & Dietze, S. (2013, May). Assessing the educational linked data landscape. Proceedings of the 5th Annual ACM Web Science Conference <\/i>(WebSci\u201913), 2\u20134 May 2013, Paris, France (pp. 43\u201346). New York: ACM. <\/span><\/p>\n

          De Nies, T., Salliau, F., Verborgh, R., Mannens, E., & Van de Walle, R. (2015, May). TinCan2PROV: Exposing interoperable provenance of learning processes through experience API logs. Proceedings of the 24th International Conference on World Wide Web <\/i>(WWW\u201915), 18\u201322 May 2015, Florence, Italy (pp. 689\u2013694). New York: ACM. <\/span><\/p>\n

          Desmarais, M. C., & Baker, R. S. (2012). A review of recent advances in learner and skill modeling in intelligent learning environments. User Modeling and User-Adapted Interaction, 22<\/i>(1\u20132), 9\u201338. <\/span><\/p>\n

          Dietze, S., Drachsler, H., & Giordano, D. (2014). A survey on linked data and the social web as facilitators for TEL recommender systems. In N. Manouselis, K. Verbert, H. Drachsler, & O. C. Santos (Eds.), Recommender Systems for Technology Enhanced Learning <\/i>(pp. 47\u201375). New York: Springer. <\/span><\/p>\n

          Dietze, S., Sanchez-Alonso, S., Ebner, H., Qing Yu, H., Giordano, D., Marenzi, I., & Nunes, B. P. (2013). Interlinking educational resources and the web of data: A survey of challenges and approaches. Program, 47<\/i>(1), 60\u201391. <\/span><\/p>\n

          Dietze, S., Yu, H. Q., Giordano, D., Kaldoudi, E., Dovrolis, N., & Taibi, D. (2012). Linked education: Interlinking educational resources and the web of data. Proceedings of the 27th Annual ACM Symposium on Applied Computing <\/i>(SAC 2012), 26\u201330 March 2012, Riva (Trento), Italy (pp. 366\u2013371). New York: ACM. Dowell, N. M., Skrypnyk, O., Joksimovi\u0107, S., Graesser, A. C., Dawson, S., Ga\u0161evi\u0107, D., Hennis, T. A., de Vries, P., & Kovanovi\u0107, V. (2015). Modeling learners’ social centrality and performance through language and discourse. In O. C. Santos, J. G. Boticario, C. Romero, M. Pechenizkiy, A. Merceron, P. Mitros, J. M. Luna, C. Mihaescu, P. Moreno, A. Hershkovitz, S. Ventura, & M. Desmarais (Eds.), Proceedings of the 8th International Conference on Education Data Mining <\/i>(EDM2015), 26\u201329 June 2015, Madrid, Spain (pp. 250\u2013257). International Educational Data Mining Society. http:\/\/files.eric.ed.gov\/fulltext\/ED560532.pdf <\/span><\/p>\n

          Duval, E. (2011). Attention please! Learning analytics for visualization and recommendation. Proceedings of the 1st International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201911), 27 February\u20131 March 2011, Banff, AB, Canada (pp. 9\u201317). New York: ACM. <\/span><\/p>\n

          Ga\u0161evi\u0107, D., Dawson, S., & Siemens, G. (2015). Let’s not forget: Learning analytics are about learning. TechTrends, 59<\/i>(1), 64\u201371. <\/span><\/p>\n

          Groth, P., Loizou, A., Gray, A. J., Goble, C., Harland, L., & Pettifer, S. (2014). API-centric linked data integration: The open PHACTS discovery platform case study. Web Semantics: Science, Services and Agents on the World Wide Web, 29<\/i>, 12\u201318. <\/span><\/p>\n

          Heath, T., & Bizer, C. (2011). Linked data: Evolving the web into a global data space. Synthesis lectures on the semantic web: Theory and technology, 1<\/i>(1), 1\u2013136. Morgan & Claypool. <\/span><\/p>\n

          Hu, Y., McKenzie, G., Yang, J. A., Gao, S., Abdalla, A., & Janowicz, K. (2014). A linked-data-driven web portal for learning analytics: Data enrichment, interactive visualization, and knowledge discovery. In LAK Workshops. http:\/\/geog.ucsb.edu\/~jano\/LAK2014.pdf <\/span><\/p>\n

          Joksimovi\u0107, S., Kovanovi\u0107, V., Jovanovi\u0107, J., Zouaq, A., Ga\u0161evi\u0107, D., & Hatala, M. (2015). What do cMOOC participants talk about in social media? A topic analysis of discourse in a cMOOC. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March, Poughkeepsie, NY, USA (pp. 156\u2013165). New York: ACM. <\/span><\/p>\n

          Jovanovi\u0107, J., Bagheri, E., Cuzzola, J., Ga\u0161evi\u0107, D., Jeremic, Z., & Bashash, R. (2014). Automated semantic annotation of textual content. IEEE IT Professional, 16<\/i>(6), 38\u201346. <\/span><\/p>\n

          Kagemann, S., & Bansal, S. (2015). MOOCLink: Building and utilizing linked data from massive open online courses. Proceedings of the 9th IEEE International Conference on Semantic Computing <\/i>(IEEE ICSC 2015), 7\u20139 February 2015, Anaheim, California, USA (pp. 373\u2013380). IEEE. <\/span><\/p>\n

          Kessler, C., d\u2019Aquin, M., & Dietze, S. (2013). Linked data for science and education. Journal of Semantic Web, 4<\/i>(1), 1\u20132. <\/span><\/p>\n

          Kontokostas, D., Westphal, P., Auer, S., Hellmann, S., Lehmann, J., Cornelissen, R., & Zaveri, A. (2014). Test-driven evaluation of linked data quality. Proceedings of the 23rd International Conference on World Wide Web <\/i>(WWW\u201914), 7\u201311 April 2014, Seoul, Republic of Korea (pp. 747\u2013758). New York: ACM. <\/span><\/p>\n

          Lausch, A., Schmidt, A., & Tischendorf, L. (2015). Data mining and linked open data: New perspectives for data analysis in environmental research. Ecological Modelling, 295<\/i>, 5\u201317. <\/span><\/p>\n

          Maturana, R. A., Alvarado, M. E., L\u00f3pez-Sola, S., Ib\u00e1\u00f1ez, M. J., & El\u00f3segui, L. R. (2013). Linked data based applications for learning analytics research: Faceted searches, enriched contexts, graph browsing and dynamic graphic visualisation of data. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_03.pdf <\/span><\/p>\n

          Miliki\u0107, N., Krcadinac, U., Jovanovi\u0107, J., Brankov, B., & Keca, S. (2013). Paperista: Visual exploration of semantically annotated research papers. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_04.pdf <\/span><\/p>\n

          Mirriahi, N., Ga\u0161evi\u0107, D., Dawson, S., & Long, P. D. (2014). Scientometrics as an important tool for the growth of the field of learning analytics. Journal of Learning Analytics, 1<\/i>(2), 1\u20134. <\/span><\/p>\n

          Mu\u00f1oz-Merino, P. J., Pardo, A., Kloos, C. D., Mu\u00f1oz-Organero, M., Wolpers, M., Katja, K., & Friedrich, M. (2010). CAM in the semantic web world. In A. Paschke, N. Henze, & T. Pellegrini (Eds.), Proceedings of the 6th International Conference on Semantic Systems <\/i>(I-Semantics\u201910), 1\u20133 September 2010, Graz, Austria. New York: ACM. doi:10.1145\/1839707.1839737Newman, M. E. (2006). Modularity and community structure in networks, Proceedings of the National Academy of Sciences, 103<\/i>(23), 8577\u20138582. <\/span><\/p>\n

          Niemann, K., Wolpers, M., Stoitsis, G., Chinis, G., & Manouselis, N. (2013). Aggregating social and usage datasets for learning analytics: Data-oriented challenges. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 245\u2013249). New York: ACM. <\/span><\/p>\n

          Nunes, B. P., Fetahu, B., & Casanova, M. A. (2013). Cite4Me: Semantic retrieval and analysis of scientific publications. LAK Data Challenge, 974. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_06.pdf <\/span><\/p>\n

          Ochoa, X., Suthers, D., Verbert, K., & Duval, E. (2014). Analysis and reflections on the third learning analytics and knowledge conference (LAK 2013). Journal of Learning Analytics, 1<\/i>(2), 5\u201322. <\/span><\/p>\n

          Piedra, N., Chicaiza, J. A., L\u00f3pez, J., & Tovar, E. (2014). An architecture based on linked data technologies for the integration and reuse of OER in MOOCs context. Open Praxis, 6<\/i>(2), 171\u2013187. <\/span><\/p>\n

          Santos, J. L., Verbert, K., Klerkx, J., Duval, E., Charleer, S., & Ternier, S. (2015). Tracking data in open learning environments. Journal of Universal Computer Science, 21<\/i>(7), 976\u2013996. <\/span><\/p>\n

          Scheffel, M., Niemann, K., Leon Rojas, S., Drachsler, H., & Specht, M. (2014). Spiral me to the core: Getting a visual grasp on text corpora through clusters and keywords. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol- 1137\/lakdatachallenge2014_submission_3.pdf <\/span><\/p>\n

          Schmitz, H. C., Wolpers, M., Kirschenmann, U., & Niemann, K. (2011). Contextualized attention metadata. In C. Roda (Ed.), Human attention in digital environments <\/i>(pp. 186\u2013209). New York: Cambridge University Press. <\/span><\/p>\n

          Sharkey, M., & Ansari, M. (2014). Deconstruct and reconstruct: Using topic modeling on an analytics corpus. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-1137\/lakdatachallenge2014_submission_1.pdf <\/span><\/p>\n

          Siemens, G. (2005). Connectivism: A learning theory for the digital age. International Journal of Instructional Technology and Distance Learning, 2<\/i>(1), 3\u201310. http:\/\/itdl.org\/Journal\/Jan_05\/article01.htm <\/span><\/p>\n

          Softic, S., De Vocht, L., Taraghi, B., Ebner, M., Mannens, E., & De Walle, R. V. (2014). Leveraging learning analytics in a personal learning environment using linked data. Bulletin of the IEEE Technical Committee on Learning Technology, 16<\/i>(4), 10\u201313. <\/span><\/p>\n

          Taibi, D., & Dietze, S. (2013), Fostering analytics on learning analytics research: The LAK dataset. LAK Data Challenge, 974. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_preface.pdf <\/span><\/p>\n

          Zablith, F. (2015). Interconnecting and enriching higher education programs using linked data. Proceedings of the 24th International Conference on World Wide Web <\/i>(WWW\u201915), 18\u201322 May 2015, Florence, Italy (pp. 711\u2013716). New York: ACM. <\/span><\/p>\n

          Zotou, M., Papantoniou, A., Kremer, K., Peristeras, V., & Tambouris, E. (2014). Implementing \u201crethinking education\u201d: Matching skills profiles with open courses through linked open data technologies. Bulletin of the IEEE Technical Committee on Learning Technology, 16<\/i>(4), 18\u201321. <\/span><\/p>\n

          Zouaq, A., Joksimovi\u0107, S., & Ga\u0161evi\u0107, D. (2013). Ontology learning to analyze research trends in learning analytics publications. LAK Data Challenge. http:\/\/ceur-ws.org\/Vol-974\/lakdatachallenge2013_08.pdf<\/span><\/a><\/span><\/span><\/p>\n

          \n
          \n

          1<\/a> connectivist<\/span><\/p>\n<\/div>\n

          \n

          2<\/a> knowledge graph<\/span><\/p>\n<\/div>\n

          \n

          3<\/a> https:\/\/www.w3.org\/TR\/sparql11\u2013query\/<\/span><\/p>\n<\/div>\n

          \n

          4<\/a> http:\/\/lod-cloud.net\/<\/span><\/p>\n<\/div>\n

          \n

          5<\/a> http:\/\/lod-cloud.net\/<\/span><\/p>\n<\/div>\n

          \n

          6<\/a> http:\/\/bit.ly\/yago-naga<\/span><\/p>\n<\/div>\n

          \n

          7<\/a> http:\/\/bit.ly\/wikidata-main<\/span><\/p>\n<\/div>\n

          \n

          8<\/a> http:\/\/www.foaf-project.org\/<\/span><\/p>\n<\/div>\n

          \n

          9<\/a> http:\/\/activitystrea.ms\/<\/span><\/p>\n<\/div>\n

          \n

          10<\/a> http:\/\/www.adlnet.gov\/tla\/experience-api<\/span><\/p>\n<\/div>\n

          \n

          11<\/a> http:\/\/ieeeltsc.org\/wg12LOM\/<\/span><\/p>\n<\/div>\n

          \n

          12<\/a> http:\/\/www.adlnet.org\/<\/span><\/p>\n<\/div>\n

          \n

          13<\/a> http:\/\/www.imsglobal.org\/question\/<\/span><\/p>\n<\/div>\n

          \n

          14<\/a> Avrupa Komisyonu, “ABYNM: Avrupa Beceriler, Yeterlilikler, Nitelikler ve Meslekler”, https:\/\/ec.europa.eu\/esco<\/span><\/p>\n<\/div>\n

          \n

          15<\/a> https:\/\/github.com\/adlnet\/xAPI\u2013Spec<\/span><\/p>\n<\/div>\n

          \n

          16<\/a> http:\/\/www.bnf.fr\/en\/tools\/a.welcome_to_the_bnf.html<\/span><\/p>\n<\/div>\n

          \n

          17<\/a> http:\/\/datahub.io\/dataset\/l3s-dblp<\/span><\/p>\n<\/div>\n

          \n

          18<\/a> http:\/\/lov.okfn.org\/dataset\/lov\/<\/span><\/p>\n<\/div>\n

          \n

          19<\/a> \u00c7evirenin notu: Google\u2019un arama esnas\u0131nda sa\u011f tarafta konu ile ilgili ayr\u0131 bir kutu i\u00e7erisinde getirdi\u011fi bilgi paneli<\/span><\/p>\n<\/div>\n","protected":false},"author":1,"menu_order":7,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":[],"pb_section_license":""},"chapter-type":[48],"contributor":[],"license":[],"class_list":["post-137","chapter","type-chapter","status-publish","hentry","chapter-type-numberless"],"part":109,"_links":{"self":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/137","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/users\/1"}],"version-history":[{"count":0,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/137\/revisions"}],"part":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/parts\/109"}],"metadata":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/137\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/media?parent=137"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapter-type?post=137"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/contributor?post=137"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/license?post=137"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}