{"id":64,"date":"2020-09-03T16:39:06","date_gmt":"2020-09-03T13:39:06","guid":{"rendered":"http:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-11-cok-modlu-ogrenme-analitigi\/"},"modified":"2020-09-03T16:39:06","modified_gmt":"2020-09-03T13:39:06","slug":"bolum-11-cok-modlu-ogrenme-analitigi","status":"publish","type":"chapter","link":"https:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-11-cok-modlu-ogrenme-analitigi\/","title":{"raw":"B\u00f6l\u00fcm 11 \u00c7ok Modlu \u00d6\u011frenme Analiti\u011fi","rendered":"B\u00f6l\u00fcm 11 \u00c7ok Modlu \u00d6\u011frenme Analiti\u011fi"},"content":{"raw":"\n

Xavier Ochoa<\/span><\/span><\/p>\n

Politecnica del Litoral Lisesi, Ekvator <\/span><\/span><\/p>\n

DOI: 10.18608\/hla17.011<\/span><\/span><\/p>\n\n

\u00d6Z<\/h2>\n

Bu b\u00f6l\u00fcm, \u00f6\u011frenme s\u00fcreci hakk\u0131nda daha sa\u011flam ve daha belirgin bir anlay\u0131\u015f elde etmek i\u00e7in tamamlay\u0131c\u0131 \u00f6\u011frenme izleri kaynaklar\u0131n\u0131n yakalanmas\u0131, birle\u015ftirilmesi ve analiz edilmesi yoluyla \u00f6\u011frenme analiti\u011fi (\u00d6A) yakla\u015f\u0131m y\u00f6ntemi sunmaktad\u0131r. \u00c7ok modlu \u00f6\u011frenme analitiklerinde (\u00c7M\u00d6A) kaynaklar veya y\u00f6ntemler1<\/a><\/sup>, \u00e7evrimi\u00e7i sistemler taraf\u0131ndan yakalanan geleneksel kay\u0131t g\u00fcnl\u00fc\u011f\u00fc verilerini i\u00e7erir, fakat ayn\u0131 zamanda yapay nesneleri ve hareketler, bak\u0131\u015f, konu\u015fma veya yazma gibi daha do\u011fal insan hareketlerini (sinyal) \u00f6\u011frenmeyi de i\u00e7erir. \u00c7M\u00d6A'n\u0131n mevcut durumu genellikle uyguland\u0131\u011f\u0131 \u00f6\u011frenme ortamlar\u0131na g\u00f6re tart\u0131\u015f\u0131l\u0131r ve y\u00f6ntemlerine g\u00f6re s\u0131n\u0131fland\u0131r\u0131l\u0131r. Bu b\u00f6l\u00fcm, \u00e7ok modlu tekniklerin uygulay\u0131c\u0131lar\u0131 i\u00e7in ortaya \u00e7\u0131kan sorunlar\u0131n tart\u0131\u015f\u0131lmas\u0131yla sona ermektedir.<\/span><\/span><\/p>\n

Anahtar Kelimeler<\/span>: Ses, video, veri birle\u015ftirme, \u00e7oklu alg\u0131lay\u0131c\u0131<\/span><\/span><\/p>\n

Esasen, \u00f6\u011frenme analiti\u011fi (\u00d6A) alan\u0131n\u0131n odak noktas\u0131, \u00f6\u011frencilerin bir \u00e7e\u015fit dijital ara\u00e7 kullan\u0131rken ger\u00e7ekle\u015ftirdikleri eylemlerin incelenmesiydi. Bu dijital ara\u00e7lar, \u00f6\u011frenme y\u00f6netimi sistemleri (\u00d6YS; Arnold ve Pistilli, 2012), bilgisayar destekli ak\u0131ll\u0131 \u00f6\u011fretim sistemleri (A\u00d6S'ler; Crossley, Roscoe ve McNamara, 2013), kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i dersler (KA\u00c7D'ler, K\u0131z\u0131lcec, Piech ve Schneider, 2013), e\u011fitici video oyunlar\u0131 (Serrano-Laguna ve Fernandez \u2013 Manjon, 2014) veya bir bilgisayar\u0131 \u00f6\u011frenme s\u00fcrecinde aktif bir bile\u015fen olarak kullanan di\u011fer sistemlerdir. \u00d6te yandan, bilgisayarlar\u0131n bulunmad\u0131\u011f\u0131 veya yaln\u0131zca yard\u0131mc\u0131, tan\u0131mlanmam\u0131\u015f bir role sahip oldu\u011fu y\u00fcz y\u00fcze dersler veya \u00e7al\u0131\u015fma gruplar\u0131 gibi di\u011fer \u00f6\u011frenme ba\u011flamlar\u0131nda nispeten daha az \u00d6A ara\u015ft\u0131rmas\u0131 veya uygulamas\u0131 yap\u0131lm\u0131\u015ft\u0131r. Bilgisayar destekli \u00f6\u011frenme ba\u011flamlar\u0131na y\u00f6nelik bu yanl\u0131l\u0131k, her t\u00fcrl\u00fc \u00d6A \u00e7al\u0131\u015fmas\u0131n\u0131n veya sisteminin temel ihtiyac\u0131 ile a\u00e7\u0131klanmaktad\u0131r: \u00f6\u011frenme izlerinin varl\u0131\u011f\u0131 (Siemens, 2013).<\/span><\/p>\n

Bilgisayar tabanl\u0131 \u00f6\u011frenme sistemleri, ba\u015flang\u0131\u00e7ta analitik d\u00fc\u015f\u00fcn\u00fclerek tasarlanmam\u0131\u015f olsa bile, kullan\u0131c\u0131lar\u0131 ile etkile\u015fimlerini otomatik olarak ince tanecikli ayr\u0131nt\u0131 \u015feklinde yakalama e\u011filimindedir. Bu etkile\u015fimleri tan\u0131mlayan veriler, analiz edilecek izlemleri \u00e7\u0131karmak i\u00e7in daha sonra tahmin edilebilecek g\u00fcnl\u00fck dosyalar\u0131 veya kelime i\u015flemci belgeleri gibi bir\u00e7ok formda saklan\u0131r. Kullan\u0131ma haz\u0131r verilerin g\u00f6reli bollu\u011fu ve i\u015flemenin \u00f6n\u00fcndeki teknik engeller, bilgisayar tabanl\u0131 \u00f6\u011frenme sistemlerini \u00d6A i\u00e7in AR-GE yapmada ideal bir yer haline getirir. Buna kar\u015f\u0131l\u0131k, bilgisayarlar\u0131n kullan\u0131lmad\u0131\u011f\u0131 \u00f6\u011frenme ba\u011flamlar\u0131nda, \u00f6\u011frenenlerin eylemleri otomatik olarak yakalanamaz2<\/a><\/sup>. \u00d6\u011frencinin \u00fcretti\u011fi fiziksel belgeler gibi baz\u0131 \u00f6\u011frenme \u00fcr\u00fcnleri3<\/a><\/sup> mevcut olsa bile, i\u015flenmeden \u00f6nce d\u00f6n\u00fc\u015ft\u00fcr\u00fclmeleri gerekir. Analiz edilecek izler olmadan, \u00d6A'da geleneksel olarak kullan\u0131lan bilgi i\u015flemsel modeller ve ara\u00e7lar ge\u00e7erli de\u011fildir.<\/span><\/p>\n

Bilgisayar destekli \u00f6\u011frenme ba\u011flamlar\u0131na y\u00f6nelik bu yanl\u0131l\u0131\u011f\u0131n varl\u0131\u011f\u0131, \u00d6A'da bir sokak lambas\u0131 etkisi (Freedman, 2010) olu\u015fturabilir. Bu etki ismini bir adam\u0131n ev anahtar\u0131n\u0131 parkta kaybetmesine ra\u011fmen onu sokak lambas\u0131n\u0131n alt\u0131nda aramas\u0131n\u0131 ifade eden bir \u015fakadan almaktad\u0131r. Sahneyi izleyen polis neden anahtar\u0131 sokak lambas\u0131n\u0131n alt\u0131nda arad\u0131\u011f\u0131n\u0131 sordu\u011funda adam \"\u00e7\u00fcnk\u00fc buras\u0131 daha ayd\u0131nl\u0131k\" diye cevap verir. Sokak lambas\u0131 efekti, \u00e7\u00f6z\u00fcmleri ger\u00e7ek \u00e7\u00f6z\u00fcmlerin olabilece\u011fi yerde de\u011fil araman\u0131n kolay oldu\u011fu yerde aramak anlam\u0131na gelmektedir. Bu durum s\u00fcrecin b\u00fcy\u00fck bir b\u00f6l\u00fcm\u00fcn\u00fcn ger\u00e7ekle\u015fti\u011fi ger\u00e7ek d\u00fcnya ortamlar\u0131n\u0131 g\u00f6z ard\u0131 edip, \u00f6\u011frenme s\u00fcrecini yaln\u0131zca bilgisayar temelli ba\u011flamlara bakarak anlamaya ve optimize etmeye \u00e7al\u0131\u015fan erken d\u00f6nem \u00d6A ara\u015ft\u0131rmalar\u0131 i\u00e7in d\u00fc\u015f\u00fcn\u00fclebilir. Hatta \u00f6\u011frenenlerin bilgisayar destekli sistemlerde kayd\u0131n\u0131n tutulamad\u0131\u011f\u0131 eylemleri bile genellikle g\u00f6z ard\u0131 edilir. \u00d6rne\u011fin, bir A\u00d6S'de bir problem sunuldu\u011funda kafas\u0131 kar\u0131\u015fan veya \u00e7evrimi\u00e7i bir ders izlerken s\u0131k\u0131l\u0131p esneyen bir \u00f6\u011frenen hakk\u0131ndaki bilgiler geleneksel \u00d6A ara\u015ft\u0131rmalar\u0131nda dikkate al\u0131nmaz. Sokak lambas\u0131 etkisini azaltmak i\u00e7in, ara\u015ft\u0131rmac\u0131lar \u015fimdi ger\u00e7ek d\u00fcnyadaki \u00f6\u011frenme ba\u011flamlar\u0131ndan ince taneli \u00f6\u011frenme izlerinin otomatik olarak toplanmas\u0131na odaklanarak, Bir KA\u00c7D oturumunun analizi kadar y\u00fcz y\u00fcze derslerin analizini de m\u00fcmk\u00fcn hale getiriyorlar. \u00d6A \u00fczerine daha yeni \u00e7al\u0131\u015fmalar, geleneksel g\u00fcnl\u00fck dosyalar\u0131ndan ba\u015fka yeni veri kaynaklar\u0131n\u0131 ara\u015ft\u0131r\u0131yor: \u00f6\u011frenen taraf\u0131ndan \u00fcretilen metinler (Simsek vd., 2015), g\u00f6z izleme bilgileri (Vatrapu, Reimann, Bull ve Johnson, 2013) ve s\u0131n\u0131f yap\u0131land\u0131rmas\u0131 (Almeda, Scupelli, Baker, Weber ve Fisher, 2014) bunlardan birka\u00e7\u0131d\u0131r. Bu farkl\u0131 \u00f6\u011frenme izlerinin kaynaklar\u0131n\u0131n tek bir analizde birle\u015ftirilmesi, \u00e7ok modlu \u00f6\u011frenme analiti\u011finin (\u00c7M\u00d6A) temel amac\u0131d\u0131r.<\/span><\/p>\n

\u00c7M\u00d6A, etkile\u015fimlerin sadece bir bilgisayar veya dijital cihaz arac\u0131l\u0131\u011f\u0131yla sa\u011flanmad\u0131\u011f\u0131 dijital ve ger\u00e7ek d\u00fcnya senaryolar\u0131ndaki \u00f6\u011frenmeyi anlamaya ve optimize etmeye odaklanarak farkl\u0131 \u00f6\u011frenme izleri kaynaklar\u0131n\u0131 \u00d6A ara\u015ft\u0131rmas\u0131na ve uygulamas\u0131na d\u00e2hil etmeye \u00e7al\u0131\u015fan bir alt aland\u0131r (Blikstein, 2013). \u00c7M\u00d6A'da, \u00f6\u011frenme izleri sadece g\u00fcnl\u00fck dosyalar\u0131ndan veya dijital belgelerden de\u011fil, kaydedilmi\u015f video ve seslerden, kalem vuru\u015flar\u0131ndan, konum izleme cihazlar\u0131ndan, biyo-alg\u0131lay\u0131c\u0131lardan ve \u00f6\u011frenme s\u00fcrecini anlamak veya \u00f6l\u00e7mek i\u00e7in yararl\u0131 olabilecek di\u011fer y\u00f6ntemlerden elde edilir. Ayr\u0131ca, \u00c7M\u00d6A'da, farkl\u0131 durum ve formlardan \u00e7\u0131kart\u0131lan izler, eylemlerin ve \u00f6\u011frenenin i\u00e7 durumunun daha kapsaml\u0131 bir g\u00f6r\u00fcn\u00fcm\u00fcn\u00fc sa\u011flamak i\u00e7in birle\u015ftirilir.<\/span><\/p>\n

\u00d6\u011frenmeyi ara\u015ft\u0131r\u0131rken farkl\u0131 y\u00f6ntem ve formlar\u0131 kullanma fikri, \u00d6A ba\u011flam\u0131nda yeni olmakla birlikte, geleneksel deneysel e\u011fitim ara\u015ft\u0131rmalar\u0131nda yayg\u0131nd\u0131r. Do\u011fas\u0131 gere\u011fi \u00e7ok modlu bir alg\u0131lay\u0131c\u0131 olan bir insan g\u00f6zlemcisini, ger\u00e7ek d\u00fcnyadaki \u00f6\u011frenme ba\u011flam\u0131na eklemek, \u00f6\u011frenmeyi do\u011fal ortamda \u00e7al\u0131\u015fman\u0131n ola\u011fan yoludur (Gall, Borg ve Gall, 1996). Video, ses kayd\u0131 ve etiketleme ara\u00e7lar\u0131 gibi teknolojiler bu g\u00f6zlemi daha az m\u00fcdahaleci ve daha \u00f6l\u00e7\u00fclebilir hale getirmi\u015ftir (Cobb vd., 2003; Lund, 2007). Geleneksel e\u011fitsel ara\u015ft\u0131rma yakla\u015f\u0131m\u0131n\u0131n temel sorunu, veri toplama ve analizlerinin, el ile yap\u0131lmalar\u0131 nedeniyle \u00e7ok maliyetli olmalar\u0131 ve \u00f6l\u00e7eklenmemeleridir. Veri toplaman\u0131n hem boyut hem de zaman a\u00e7\u0131s\u0131ndan s\u0131n\u0131rl\u0131 olmas\u0131 gerekir ve veri analizi sonu\u00e7lar\u0131, \u00e7al\u0131\u015f\u0131lan \u00f6\u011frenenler i\u00e7in faydal\u0131 olacak kadar h\u0131zl\u0131 ve kullan\u0131\u015fl\u0131 de\u011fildir. Farkl\u0131 modalite ve formlar kaydedilebilir ve \u00f6\u011frenme izleri bunlardan otomatik olarak \u00e7\u0131kar\u0131labilirse, \u00d6A ara\u00e7lar\u0131, \u00f6\u011frenmeyi oldu\u011fu gibi iyile\u015ftirmek ve s\u00fcrekli bir ger\u00e7ek zamanl\u0131 geribildirim d\u00f6ng\u00fcs\u00fc sa\u011flamak i\u00e7in kullan\u0131labilir.<\/span><\/p>\n

Beklenildi\u011fi gibi, ham \u00e7ok modlu kay\u0131tlardan \u00f6\u011frenme izlerini \u00e7\u0131karmak s\u0131radan bir i\u015f de\u011fildir. Bilgisayar g\u00f6r\u00fc\u015f\u00fc, konu\u015fma i\u015fleme, eskiz tan\u0131ma ve bilgisayar bilimleri alanlar\u0131nda geli\u015ftirilen di\u011fer teknikler, \u00f6\u011frenme bilimi, e\u011fitim ara\u015ft\u0131rmas\u0131 ve davran\u0131\u015f bilimi taraf\u0131ndan sa\u011flanan mevcut \u00f6\u011frenme teorileri taraf\u0131ndan y\u00f6nlendirilmelidir. Karma\u015f\u0131kl\u0131\u011f\u0131 g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, \u00c7M\u00d6A alt alan\u0131 nispeten gen\u00e7 ve ke\u015ffedilmemi\u015ftir. Ancak ilk \u00e7al\u0131\u015fmalar ve ara\u015ft\u0131rmac\u0131lar aras\u0131ndaki erken disiplinler aras\u0131 i\u015f birli\u011fi olumlu sonu\u00e7lar vermi\u015ftir (Scherer, Worsley ve Morency, 2012; Morency, Oviatt, Scherer, Weibel ve Worsley, 2013; Ochoa, Worsley, Chiluiza ve Luz, 2014, Markaki, Lund ve Sanchez, 2015). Bu b\u00f6l\u00fcm, bu alan\u0131 ara\u015ft\u0131rmak isteyen ara\u015ft\u0131rmac\u0131lar ve uygulay\u0131c\u0131lar i\u00e7in bir k\u0131lavuzdur. \u0130lk olarak, \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131nda kullan\u0131lan ana modaliteler sunulacak, analiz edilecek ve \u00f6rneklendirilecektir. \u0130kincisi, \u00c7M\u00d6A'n\u0131n uyguland\u0131\u011f\u0131 ger\u00e7ek d\u00fcnya ortamlar\u0131 ana durum ve modalitelerine g\u00f6re incelenecek ve s\u0131n\u0131fland\u0131r\u0131lacakt\u0131r. Son olarak, \u00c7M\u00d6A ara\u015ft\u0131rmas\u0131 ve uygulamas\u0131 i\u00e7in \u00f6nemli olan \u00e7\u00f6z\u00fclmemi\u015f birka\u00e7 konu tart\u0131\u015f\u0131lacakt\u0131r.<\/span><\/p>\n\n

MODAL\u0130TELER VE MEDYA<\/h2>\n

\u0130leti\u015fim kuram\u0131 tan\u0131m\u0131nda, \u00e7ok modluluk, bireyler aras\u0131nda bilgi ve anlam al\u0131\u015fveri\u015finde bulunmak i\u00e7in \u00e7e\u015fitli ileti\u015fim bi\u00e7imlerinin (metinsel, i\u015fitsel, dilbilimsel, mek\u00e2nsal, g\u00f6rsel, vb.) kullan\u0131lmas\u0131n\u0131 ifade eder (Kress ve Van Leeuwen, 2001). Medya filmleri, kitaplar, web sayfalar\u0131 ve hatta hava, ileti\u015fim modunun kodlanabilece\u011fi fiziksel veya dijital birer alt tabakad\u0131r. Her mod bir veya birka\u00e7 medya arac\u0131l\u0131\u011f\u0131yla ifade edilebilir. \u00d6rne\u011fin, konu\u015fma, havadaki bas\u0131n\u00e7 de\u011fi\u015fimleri (y\u00fcz y\u00fcze diyalogda), kasetteki manyetik y\u00f6n de\u011fi\u015fimleri (kaset kayd\u0131nda) veya dijital say\u0131lar\u0131n de\u011fi\u015fimleri (MP3 dosyas\u0131na) kodlanabilir. Ayr\u0131ca, ayn\u0131 ara\u00e7 birka\u00e7 modu iletmek i\u00e7in kullan\u0131labilir. \u00d6rne\u011fin, bir video kayd\u0131 v\u00fccut dili (duru\u015f), duygular (y\u00fcz ifadesi) ve kullan\u0131lan ara\u00e7lar (eylemler) hakk\u0131nda bilgi i\u00e7erebilir.<\/span><\/p>\n

Kendi do\u011fas\u0131 gere\u011fi, \u00f6\u011frenme genellikle \u00e7ok modludur (Jewitt, 2006). Bir insan, bir kitap okuyarak, bir profes\u00f6r\u00fc dinleyerek, bir i\u015flemi izleyerek, fiziksel veya dijital ara\u00e7lar kullanarak ve g\u00f6rece karma\u015f\u0131k bilgilerin kodlanabilece\u011fi herhangi bir ba\u015fka insan ileti\u015fim moduyla \u00f6\u011frenebilir. \u00d6\u011frenme s\u00fcreci ayn\u0131 zamanda birka\u00e7 geri bildirim d\u00f6ng\u00fcs\u00fc de kullan\u0131r; \u00f6r. \u00f6\u011freten dersin anla\u015f\u0131l\u0131p anla\u015f\u0131lmad\u0131\u011f\u0131n\u0131 sordu\u011funda ba\u015f\u0131n\u0131 sallayan bir \u00f6\u011freneni ya da \u00f6\u011fretenin sesinin bir konuyu a\u00e7\u0131klarken kulland\u0131\u011f\u0131 vurgulama gibi. Bu geri bildirim modlar\u0131 genellikle daha basit fakat s\u00fcre\u00e7 i\u00e7in \u00f6nemli olan bilgileri kodlar. \u00d6\u011frenme analiz edilecek, anla\u015f\u0131lacak ve optimize edilecekse, ilgili modlar\u0131n her birinde meydana gelen etkile\u015fimlerin izleri elde edilmelidir. \u00c7M\u00d6A, bu izlerin kodland\u0131\u011f\u0131 veya kaydedildi\u011fi ortamdan ba\u011f\u0131ms\u0131z olarak, bu izleri farkl\u0131 ileti\u015fim modlar\u0131ndan \u00e7\u0131karmaya odaklan\u0131r.<\/span><\/p>\n

A\u015fa\u011f\u0131daki alt b\u00f6l\u00fcmler, \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131nda kullan\u0131lan en yayg\u0131n y\u00f6ntemlerin yans\u0131t\u0131lmas\u0131 ve izlerinin \u00e7\u0131kar\u0131lmas\u0131 konusundaki mevcut durumu sunmaktad\u0131r. Her modalite i\u00e7in, \u00f6\u011frenme s\u00fcrecini anlama konusundaki \u00f6nemi, en yayg\u0131n yakalama (capture etme) ve kaydetme y\u00f6ntemlerinin listesi ve kullan\u0131ld\u0131\u011f\u0131 yerlerin \u00f6rnekleri ile birlikte k\u0131sa bir tan\u0131m sunulmu\u015ftur. Bu \u00f6\u011frenmeyle ilgili t\u00fcm modlar\u0131n kapsaml\u0131 bir listesi de\u011fil sadece \u00c7M\u00d6A \u00e7al\u0131\u015fmalar\u0131nda kullan\u0131lanlard\u0131r.<\/span><\/p>\n\n

Bak\u0131\u015f<\/h3>\n

\u0130nsanlar do\u011frudan dikkatlerini \u00e7eken \u015feye bakma e\u011filimindedir. Dolay\u0131s\u0131yla bir bireyin g\u00f6r\u00fc\u015f\u00fcn\u00fcn y\u00f6n\u00fc, dikkatinin y\u00f6n\u00fcn\u00fcn bir g\u00f6stergesidir (Frischen, Bayliss ve Tipper, 2007). Dikkat, \u00f6\u011frenme i\u00e7in vazge\u00e7ilmez bir gerekliliktir (Kruschke, 2003). Bir i\u015farete dikkat etmek, bireyin bilgilerini elde etmesine ve ilgili par\u00e7alar\u0131 uzun s\u00fcreli haf\u0131zada saklamas\u0131na yard\u0131mc\u0131 olur. Bak\u0131\u015f, dikkatleri tahmin edebilen tek temsilci ve hatas\u0131z olmasa da e\u011fitim uygulamalar\u0131nda yayg\u0131n olarak kullan\u0131l\u0131r. \u00d6rne\u011fin, bu konuda e\u011fitim alm\u0131\u015f bir \u00f6\u011freten, \u00f6\u011frencilerin bak\u0131\u015flar\u0131n\u0131 g\u00f6zlemleyerek b\u00fct\u00fcn bir s\u0131n\u0131f\u0131n dikkat seviyesini de\u011ferlendirebilir; bir g\u00f6zlemci, bak\u0131\u015f\u0131n konu\u015fmac\u0131dan konu\u015fmac\u0131ya do\u011fru y\u00f6n\u00fcn\u00fc izleyerek bir tart\u0131\u015fmadaki kat\u0131l\u0131mc\u0131n\u0131n dikkat seviyesini belirleyebilir.<\/span><\/p>\n

Bak\u0131\u015flar\u0131n \u00f6nemi uzun zamand\u0131r pazarlamac\u0131lar, davran\u0131\u015f bilimcileri ve insan-bilgisayar etkile\u015fimi ara\u015ft\u0131rmac\u0131lar\u0131 taraf\u0131ndan \u00e7al\u0131\u015f\u0131lm\u0131\u015ft\u0131r. Reklam\u0131n etkinli\u011fini (Krugman, Fox, Fletcher, Fischer ve Rojas, 1994) belirlemek, otizmin erken te\u015fhisinde (Boraston ve Blakemore, 2007) ve bilgisayar aray\u00fczlerinin etkinli\u011fi konusunda (Poole ve Ball, 2006) yard\u0131mc\u0131 olmak i\u00e7in g\u00f6z izleme \u00e7al\u0131\u015fmalar\u0131 yayg\u0131nd\u0131r. Bununla birlikte, bu \u00e7al\u0131\u015fmalarda monit\u00f6re sabitlenmi\u015f g\u00f6z izleyicileri veya \u00f6zel g\u00f6z izleme g\u00f6zl\u00fckleri kullan\u0131larak bak\u0131\u015flar\u0131 kaydetmenin ana y\u00f6ntemleri \u00f6\u011frenme ortamlar\u0131nda yayg\u0131n olarak kullan\u0131lamayacak kadar elveri\u015fsiz ve maliyetlidir. \u00c7M\u00d6A\u2019da bak\u0131\u015f\u0131 yakalaman\u0131n mevcut se\u00e7ene\u011fi, video kay\u0131tlar\u0131d\u0131r (Raca ve Dillenbourg, 2013).<\/span><\/p>\n

\"\"<\/p>\n<\/a> \u015eekil 11.1. Bir s\u0131n\u0131f ortam\u0131nda bak\u0131\u015f kestirimi (Raca, Tormey ve Dillenbourg, 2014).<\/i><\/span><\/span>\n

Bir veya bir dizi kamera, deneklerin ba\u015flar\u0131n\u0131 ve g\u00f6zlerini kaydetmek i\u00e7in konumland\u0131r\u0131lm\u0131\u015ft\u0131r. Sonra, bilgisayar g\u00f6rme teknikleri Lin, Lin, Lin ve Lee (2013) te sunuldu\u011fu gibi, video kayd\u0131ndan bak\u0131\u015f y\u00f6n\u00fc bilgisini \u00e7\u0131karmak i\u00e7in kullan\u0131l\u0131r. Kay\u0131ttaki g\u00f6receli bak\u0131\u015f y\u00f6n\u00fcn\u00fc elde etmek i\u00e7in kontrol edilmesi gereken ana hususlar, y\u00fcz \u00e7\u00f6z\u00fcn\u00fcrl\u00fc\u011f\u00fc ve ortamdaki nesneler veya di\u011fer ki\u015filer taraf\u0131ndan kapanmay\u0131 \u00f6nlemektir (Raca ve Dillenbourg, 2013). Mutlak bak\u0131\u015f y\u00f6n\u00fcn\u00fc hesaplamak i\u00e7in kameralar\u0131n \u00f6\u011frenme ayarlar\u0131ndaki konumu ile ilgili bilgiler de kaydedilmelidir.<\/span><\/p>\n

\u00c7M\u00d6A'da, bir\u00e7ok bak\u0131\u015f izi \u00e7\u0131kar\u0131m \u00f6rne\u011fi vard\u0131r. Raca ve Dillenbourg (2013), b\u00f6l\u00fcme dayal\u0131 bir model kullanarak bir derste oturan \u00f6\u011frencilerin video kay\u0131tlar\u0131nda ba\u015f y\u00f6neliminden bak\u0131\u015f y\u00f6n\u00fcn\u00fc tahmin etmektedir (\u015eekil 11.1). Bu \u015fekilde, \u00f6\u011frenen y\u00fczleri otomatik olarak tan\u0131n\u0131r (dikd\u00f6rtgen) ve bak\u0131\u015flar\u0131 (ok) bir insan y\u00fcz\u00fc modeline dayanarak tahmin edilir. Bu bilgiler daha sonra \u00f6\u011frencilerin bireysel olarak dikkatinin oda\u011f\u0131n\u0131 belirlemek ve bunlar\u0131 kendisi taraf\u0131ndan bildirilen dikkat ile kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in kullan\u0131l\u0131r. Raca ve Dillenbourg, \u00f6\u011frencilerin \u00f6\u011freteni g\u00f6r\u00fc\u015f alanlar\u0131nda tuttuklar\u0131 zaman y\u00fczdesinin dikkat d\u00fczeyinin \u00f6nemli bir belirleyicisi oldu\u011funu bulmu\u015flard\u0131r. Farkl\u0131 bir \u00f6\u011frenme ortam\u0131nda, Echeverria, Avendano, Chiluiza, Vasquez ve Ochoa (2014), g\u00f6z merkezi noktalar\u0131 ile burun ucu noktas\u0131na olan mesafeyi hesaplayarak kafa y\u00f6n\u00fcn\u00fc \u00f6l\u00e7en bak\u0131\u015f y\u00f6n\u00fcn\u00fc de tahmin etmi\u015flerdir. Bu bilgi, \u00f6\u011frencilerin akademik sunumlar s\u0131ras\u0131nda izleyiciyle g\u00f6z temas\u0131n\u0131 sa\u011flay\u0131p sa\u011flamad\u0131klar\u0131n\u0131 belirlemek i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r.<\/span><\/p>\n\n

Duru\u015f, Jestler ve Hareket (Beden Dili)<\/h3>\n

Duru\u015f, jestler ve hareket, her biri farkl\u0131 bilgi t\u00fcrlerini ta\u015f\u0131yabilmesine ra\u011fmen, birlikte beden dili olarak adland\u0131r\u0131lan birbiriyle ili\u015fkili \u00fc\u00e7 moddur (Bull, 2013). Duru\u015f, v\u00fccudun veya v\u00fccudun bir k\u0131sm\u0131n\u0131n belirli bir zamanda ald\u0131\u011f\u0131 pozisyonu ifade eder. Bir \u00f6\u011frenenin duru\u015fu i\u00e7sel durumu hakk\u0131nda bilgi verebilir. \u00d6rne\u011fin, bir \u00f6\u011frenen otururken ba\u015f\u0131n\u0131 masan\u0131n \u00fcst\u00fcne koymu\u015fsa, \u00f6\u011freten, bu durumdan \u00f6\u011frenenin yorgun oldu\u011funu ya da derse kar\u015f\u0131 ilgisinin olmad\u0131\u011f\u0131n\u0131 \u00e7\u0131karabilir. Baz\u0131 \u00f6zel durumlarda, edinilen duru\u015f becerilerin kazan\u0131lmas\u0131 ile ilgilidir. \u00d6rne\u011fin, s\u00f6zl\u00fc sunumlara ili\u015fkin e\u011fitilen \u00f6\u011frencilerin, belirli duru\u015flar\u0131 (eller ve kollar biraz a\u00e7\u0131k) di\u011ferlerinden (eller ceplerinde) daha fazla kullanmalar\u0131 beklenir. \u00d6\u011frenilen jestler i\u00e7sel bir durumu g\u00f6stermez. Jestler, v\u00fccudun farkl\u0131 b\u00f6lgelerinin, \u00f6zellikle ba\u015f, kollar ve ellerin belirli bir anlam\u0131 iletmek i\u00e7in koordine edilmi\u015f hareketleridir. S\u00f6zel olmayan bu ileti\u015fim \u015fekli genellikle bilin\u00e7lidir. \u00d6\u011frenme s\u00fcrecinde k\u0131sa geri bildirim d\u00f6ng\u00fcleri ve alternatif vurgulama kanallar\u0131 sa\u011flaman\u0131n bir yolu olarak kullan\u0131l\u0131r. \u00d6rne\u011fin, tahtadaki belirli bir b\u00f6l\u00fcme i\u015faret eden \u00f6\u011fretmen veya zor bir soru ile kar\u015f\u0131la\u015ft\u0131\u011f\u0131nda omuzlar\u0131n\u0131 kald\u0131ran bir \u00f6\u011frenci. Son olarak, hareket, yeni bir duru\u015f edinmek veya belirli bir jesti icra etme gereklili\u011fi olmadan beden pozisyonundaki herhangi bir de\u011fi\u015fikli\u011fi ifade eder. Bu hareket genellikle \u00f6\u011frenme s\u00fcrecinde bireyin i\u00e7sel durumunu ortaya \u00e7\u0131karan bilin\u00e7siz v\u00fccut hareketlerinin bir sonucudur; \u00f6rne\u011fin, gerginlik veya \u015f\u00fcphe g\u00f6steren de\u011fi\u015fken hareketler.<\/span><\/p>\n

Duru\u015f, jestler ve hareketler, ger\u00e7ek d\u00fcnya ortamlar\u0131nda video yakalamadaki kolayl\u0131k, d\u00fc\u015f\u00fck maliyetli 2-D ve 3-D alg\u0131lay\u0131c\u0131lar\u0131n ve y\u00fcksek performansl\u0131 bilgisayar g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fcn kullan\u0131labilirli\u011fi ile \u00f6zellik \u00e7\u0131kar\u0131m\u0131 i\u00e7in y\u00fcksek performansl\u0131 bilgisayar g\u00f6r\u00fc\u015f\u00fc algoritmalar nedeniyle \u00c7M\u00d6A'da en s\u0131k \u00e7al\u0131\u015f\u0131lan modlar olmu\u015ftur. V\u00fccut dili, farkl\u0131 v\u00fccut par\u00e7alar\u0131na ba\u011fl\u0131 ivme\u00f6l\u00e7er (Mitra ve Acharya, 2007) veya \u00f6zel ara\u00e7lar (\u00f6r. bir Wii Remote; Schlomer, Poppinga, Henze ve Boll, 2008) kullan\u0131larak y\u00fcksek hassasiyetler yakalan\u0131rken, pratikte bunlar\u0131n \u00e7o\u011fu \u00f6\u011frenme etkinli\u011fi i\u00e7inde kullan\u0131m\u0131 mevcut sisteme \u00e7ok yabanc\u0131d\u0131r. Hareketi yakalamak i\u00e7in en yayg\u0131n \u00e7\u00f6z\u00fcm konunun videosunu kaydetmek ile duru\u015f, jestler ve hareketi tahmin etmektir. Herhangi bir kamera t\u00fcr\u00fc, ilgili hareketi yeterli \u00e7\u00f6z\u00fcn\u00fcrl\u00fckte yakalayabildi\u011fi s\u00fcrece kullan\u0131labilir. Gereken \u00e7\u00f6z\u00fcn\u00fcrl\u00fck, video ile y\u00fcr\u00fct\u00fclen \u00f6zellik \u00e7\u0131karma t\u00fcr\u00fcne ba\u011fl\u0131d\u0131r. \u0130nsan hareketinin otomatik olarak \u00e7\u0131kart\u0131lmas\u0131 i\u00e7in kullan\u0131lan en yayg\u0131n cihaz Microsoft Kinect'tir (Zhang, 2012). Kinect, video ve derinlik yakalama kar\u0131\u015f\u0131m\u0131 sayesinde ara\u015ft\u0131rmac\u0131lara, \u00e7ekilen her foto\u011fraf karesi i\u00e7in konunun yeniden yap\u0131land\u0131r\u0131lm\u0131\u015f bir iskeletini sunabilir. Bu da v\u00fccut duru\u015flar\u0131n\u0131 ve hareketlerini yakalamak i\u00e7in idealdir. Kinect alg\u0131lay\u0131c\u0131s\u0131n\u0131n yeni s\u00fcr\u00fcmleri ayn\u0131 zamanda el hareketini de \u00e7\u0131kartabilmektedir (Vasquez, Vargas ve Sucar, 2015).<\/span><\/p>\n

\"\"<\/p>\n<\/a> \u015eekil 11.2. Ger\u00e7ek \u00f6\u011frenen sunucular\u0131n k\u00fcmelenmi\u015f \u00fcst v\u00fccut duru\u015flar\u0131 (Echeverr\u00eda, Avenda\u00f1o, Chiluiza,V\u00e1squez ve Ochoa, 2014).<\/i><\/span><\/span>\n

\u00c7M\u00d6A'da beden dilinin yakalanmas\u0131na ve i\u015flenmesine ili\u015fkin en belirgin \u00f6rnekler, bir s\u0131n\u0131f ortam\u0131nda \u00fcst beden hareket gecikmesi arac\u0131l\u0131\u011f\u0131yla dikkatin tahmin edilmesi (Raca, Tormey ve Dillenbourg, 2014) ve otomatikle\u015ftirilmi\u015f bir sunum \u00f6\u011freticisi olu\u015fturulmas\u0131na y\u00f6nelik gen\u00e7 bir akademisyen sunucunun duru\u015f ve jest analizidir (Echeverria vd., 2014). \u015eekil 11.2, \u00e7al\u0131\u015fmalar\u0131n\u0131 sunan \u00f6\u011frencilerin Kinect verilerinin analizinden elde edilen 23 farkl\u0131 durumu g\u00f6stermektedir. Bu 23 duru\u015f, bir sunum i\u00e7in iyi veya k\u00f6t\u00fc olarak kabul edilebilecek alt\u0131 v\u00fccut hareketi (farkl\u0131 renkler) olarak s\u0131n\u0131fland\u0131r\u0131lm\u0131\u015ft\u0131r. \u015eekil 11.3 ger\u00e7ek sunumlar s\u0131ras\u0131nda bu v\u00fccut hareketlerinin ger\u00e7ek \u00f6rneklerini sunar. Pozun s\u0131n\u0131fland\u0131r\u0131lmas\u0131 (soldaki Kinect noktalar\u0131n\u0131n \u00fcst\u00fcnde), bir insan g\u00f6zlemcinin foto\u011fraftan yorumlayabildi\u011fi ile ayn\u0131d\u0131r (a\u015fa\u011f\u0131da).<\/span><\/p>\n

\"\"<\/p>\n<\/a> \u015eekil 11.3. Prototip duru\u015flara g\u00f6re s\u0131n\u0131fland\u0131r\u0131lm\u0131\u015f ger\u00e7ek duru\u015flar (Echeverria vd., 2014).<\/i><\/span><\/span>\n

Jestleri kullanman\u0131n di\u011fer ilgin\u00e7 \u00f6rnekleri Boncoddo vd. (2013), Alibali, Nathan, Fujimori, Stein ve Raudenbush, (2011) ve Mazur-Palandre, Colletta ve Lund (2014)'d\u0131r. \u0130lk olarak, Boncoddo vd. (2013) matematiksel kan\u0131tlar\u0131n a\u00e7\u0131klanmas\u0131 s\u0131ras\u0131nda ger\u00e7ekle\u015ftirilen ilgili jestlerin say\u0131s\u0131n\u0131 yakalam\u0131\u015f ve \u00f6\u011frencilerin cevaplar\u0131na ula\u015fma \u015fekilleriyle ili\u015fki kurmu\u015ftur. \u0130kincisi, Alibali vd. (2011) \u00f6\u011fretmenlerin matematik derslerinde yapt\u0131klar\u0131 farkl\u0131 jestleri s\u0131n\u0131fland\u0131rm\u0131\u015f ve aralar\u0131ndaki ili\u015fkileri bulmu\u015flard\u0131r. Son olarak, Mazur-Palandre vd. (2014), s\u00fcre\u00e7 ve talimatlar\u0131 a\u00e7\u0131klarken \u00e7ocuklar\u0131n jestleri kullan\u0131m\u0131 \u00fczerine bir \u00e7al\u0131\u015fma sunmu\u015ftur.<\/span><\/p>\n\n

Eylemler<\/h3>\n

Eylem modu, jest ve hareket modlar\u0131na \u00e7ok benzer. Her ikisi de genellikle \u00c7M\u00d6A\u2019daki video kay\u0131tlar\u0131 taraf\u0131ndan yakalanan v\u00fccut hareketleridir. Bununla birlikte, eylemler genellikle \u00f6\u011frenilen bir arac\u0131n manip\u00fclasyonunu i\u00e7eren ama\u00e7l\u0131 hareketlerdir. Bu eylemlerin t\u00fcr\u00fc, dizilim veya do\u011frulu\u011fu, \u00f6\u011frenenin belirli bir beceride elde etti\u011fi ustal\u0131k seviyesinin bir g\u00f6stergesi olarak kullan\u0131labilir. \u00d6rne\u011fin, bir \u00f6\u011frenenin laboratuvardaki \u00e7e\u015fitli ara\u00e7lar\u0131 manip\u00fcle edi\u015findeki d\u00fczen ve g\u00fcvenlik, \u00f6\u011frenenin belirli bir s\u00fcre\u00e7 hakk\u0131ndaki anlay\u0131\u015f\u0131n\u0131 belirlemek i\u00e7in bir sembol olarak kullan\u0131labilir.<\/span><\/p>\n

\u00c7M\u00d6A'da eylem kay\u0131t ve analizinin temel kullan\u0131mlar\u0131 uzmanl\u0131k kestirimindedir. Bir m\u00fchendislik in\u015fas\u0131 faaliyetinde, \u00f6rne\u011fin, el ve bilek hareketinin analizi uzmanl\u0131k seviyesini belirleyebilir (Worsley ve Blikstein, 2014b). Matematiksel problem \u00e7\u00f6zmede, \u00f6\u011frenenin hesap makinesini kulland\u0131\u011f\u0131 zaman y\u00fczdesi \u00f6l\u00e7\u00fclebilir (Ochoa vd., 2013). Ochoa vd. (2013) hesap makinesinin problem \u00e7\u00f6zme oturumlar\u0131ndaki pozisyonunu ve a\u00e7\u0131s\u0131n\u0131 izlemi\u015ftir (\u015eekil 11.4). Bu konum ve a\u00e7\u0131 (do\u011fru) daha sonra videodaki o belirli \u00e7er\u00e7eve s\u0131ras\u0131nda (g\u00f6r\u00fcnt\u00fcn\u00fcn kenarl\u0131\u011f\u0131 ile kesi\u015fme) hesap makinesini hangi \u00f6\u011frenenin kulland\u0131\u011f\u0131n\u0131 tahmin etmek i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r.<\/span><\/p>\n

\"\"<\/p>\n<\/a> \u015eekil 11.4. Uzmanl\u0131k tahmini i\u00e7in hesap makinesi kullan\u0131m\u0131n\u0131n belirlenmesi (Ochoa vd., 2013).<\/i><\/span><\/span>\n

Y\u00fcz \u0130fadeleri<\/h3>\n

Y\u00fcz ifadeleriyle toplanan bilgiler v\u00fccut dili modlar\u0131 ile de b\u00fcy\u00fck oranda ili\u015fkilidir. \u0130nsan y\u00fcz\u00fc \u00e7ok karma\u015f\u0131k zihinsel durumlar\u0131 nispeten basit ifadelerle iletebilir. Bilgisayarla g\u00f6rme alan\u0131nda, videoda kaydedilen y\u00fcz ifadelerinden duygular\u0131 otomatik olarak tan\u0131mlamaya \u00e7al\u0131\u015fan \u00e7ok say\u0131da ba\u015far\u0131l\u0131 ara\u015ft\u0131rma yap\u0131lm\u0131\u015ft\u0131r (Mishra vd., 2015).<\/span><\/p>\n

\u00d6A alan\u0131nda y\u00fcz ifadeleri kullanman\u0131n ana \u00f6rnekleri, Craig, D'Mello, Witherspoon ve Graesser (2008) ve Worsley ve Blikstein (2015b) 'nin eserleridir. Craig vd. (2008), AutoTutor sistemini kullan\u0131rken \u00f6\u011frencilerin duygusal durumlar\u0131n\u0131 otomatik olarak tahmin etmi\u015ftir (Graesser, Chipman, Haynes ve Olney, 2005). Worsley ve Blikstein (2015b), \u00f6\u011frenciler farkl\u0131 in\u015fa al\u0131\u015ft\u0131rmalar\u0131 ile kar\u015f\u0131 kar\u015f\u0131ya kald\u0131klar\u0131nda duygusal de\u011fi\u015fiklikleri ke\u015ffetmek i\u00e7in benzer teknikleri kulland\u0131lar. Her iki \u00e7al\u0131\u015fmada da kar\u0131\u015f\u0131k bir ifadenin \u00f6\u011frenme s\u00fcrecinin ba\u015far\u0131s\u0131n\u0131n iyi bir g\u00f6stergesi oldu\u011funu ke\u015ffetmi\u015ftir.<\/span><\/p>\n\n

Konu\u015fma<\/h3>\n

\u00c7M\u00d6A'da ses kay\u0131tlar\u0131n\u0131n en yayg\u0131n kullan\u0131m\u0131, \u00f6\u011frenenin ne hakk\u0131nda konu\u015ftu\u011funu veya dinledi\u011fini izlemektir. \u0130nsanlar aras\u0131ndaki en temel ve en karma\u015f\u0131k ileti\u015fim \u015fekli olan konu\u015fma, \u00f6\u011frenme s\u00fcrecini anlamada \u00f6zellikle \u00f6nemlidir. Mevcut \u00c7M\u00d6A uygulamas\u0131nda, ses kay\u0131tlar\u0131ndan iki ana sinyal \u00e7\u0131kar\u0131l\u0131r: ne s\u00f6yleniyor ve nas\u0131l s\u00f6yleniyor. \u0130lk yakla\u015f\u0131mda, genellikle konu\u015fmay\u0131 tan\u0131ma ad\u0131 verilen, konu\u015fman\u0131n as\u0131l i\u00e7eri\u011fi \u00e7\u0131kar\u0131l\u0131r. Bu analizin sonucu, konunun neden bahsetti\u011fini belirlemek i\u00e7in do\u011fal dil i\u015fleme (DD\u0130) ara\u00e7lar\u0131 kullan\u0131larak i\u015flenebilecek bir transkripttir. \u0130kinci yakla\u015f\u0131mda, konu\u015fman\u0131n tonlama, vurgu ve ritim gibi prosodik \u00f6zellikleri \u00e7\u0131kar\u0131l\u0131r. Bu \u00f6zellikler i\u00e7 duruma (g\u00fcvenlik, duygusal durum, vb.) veya konu\u015fmac\u0131n\u0131n niyetine (\u015faka, alayc\u0131l\u0131k, vb.) I\u015f\u0131k tutabilir. Konu\u015fma tan\u0131ma, kullan\u0131lan dile b\u00fcy\u00fck \u00f6l\u00e7\u00fcde ba\u011fl\u0131d\u0131r; prosodik \u00f6zellikler dil farkl\u0131l\u0131klar\u0131na kar\u015f\u0131 daha az hassast\u0131r.<\/span><\/p>\n

Ses mikrofonlar arac\u0131l\u0131\u011f\u0131yla yakalan\u0131r. Videodan daha kolay yakalanmakla birlikte, i\u015flenecek kadar y\u00fcksek kalitede ses kayd\u0131 yapmak asl\u0131nda \u00e7ok daha karma\u015f\u0131kt\u0131r. Mikrofonlar\u0131n tipi ve mek\u00e2nsal konfig\u00fcrasyonu, \u00f6\u011frenme ortam\u0131na ve kaydedilen sinyal ile ne t\u00fcr bir analizin yap\u0131laca\u011f\u0131na ba\u011fl\u0131d\u0131r. \u00d6rne\u011fin, otomatik konu\u015fma tan\u0131ma giri\u015fiminde bulunulursa, mikrofon y\u00f6nl\u00fc olmal\u0131 ve dene\u011fin a\u011fz\u0131na yak\u0131n olmal\u0131d\u0131r. \u00d6te yandan, yaln\u0131zca birinin ne zaman konu\u015ftu\u011funun tespiti gerekliyse, odan\u0131n ortas\u0131nda bulunan bir \u00e7evresel mikrofon yeterli olabilir. G\u00fcr\u00fclt\u00fc ve \u00e7oklu sinyallerin varl\u0131\u011f\u0131 sadece otomatik \u00f6zellik \u00e7\u0131kar\u0131m\u0131n\u0131 engellemekle kalmaz, ayn\u0131 zamanda manuel ek a\u00e7\u0131klamalar\u0131 da bozar. Bireysel yak\u0131n kay\u0131t m\u00fcmk\u00fcn olmad\u0131\u011f\u0131nda kay\u0131tlar\u0131 iyile\u015ftirmek i\u00e7in kullan\u0131lan en yayg\u0131n teknik, yaln\u0131zca g\u00fcr\u00fclt\u00fcy\u00fc azaltmakla kalmayacak, ayn\u0131 zamanda sesin mek\u00e2nsal k\u00f6kenini de belirleyebilecek mikrofon dizilerinin kullan\u0131lmas\u0131d\u0131r.<\/span><\/p>\n

\u00d6neminden dolay\u0131, ses bug\u00fcne kadar \u00e7o\u011fu \u00c7M\u00d6A \u00e7al\u0131\u015fmalar\u0131nda da bulunmaktad\u0131r. \u0130\u015fbirlikli \u00f6\u011frenme diyaloglar\u0131n\u0131n benzerlik seviyesini belirlemek i\u00e7in (Luzardo, Guaman, Chiluiza, Castells ve Ochoa, 2014), s\u00f6zl\u00fc sunumlar\u0131n kalitesini de\u011ferlendirmek (Luzardo, Guaman, Chiluiza, Castells ve Ochoa, 2014)ve matematik problem \u00e7\u00f6zme uzmanl\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in farkl\u0131 konu\u015fma analizi t\u00fcrleri kullan\u0131lm\u0131\u015ft\u0131r (Thompson, 2013).<\/span><\/p>\n\n

Yazma ve Taslak<\/h3>\n

Yazma ve eskiz birbiriyle yak\u0131ndan ili\u015fkili iki moddur. Her ikisi de karma\u015f\u0131k d\u00fc\u015f\u00fcnceleri iletmek i\u00e7in bir ara\u00e7, genellikle kalem kullan\u0131r. Bir kalem kullanmak, belki de \u00f6\u011frencilerin yazma ve eskiz yapmada kullanmak i\u00e7in \u00f6\u011frendi\u011fi ilk becerilerden biridir, \u00f6zellikle erken d\u00f6nemlerde, \u00f6\u011frenmede h\u00e2l\u00e2 bask\u0131n bir faaliyettir. Bu moddan en yayg\u0131n bilgi \u00e7\u0131kar\u0131ma, \u00f6\u011frencilerin ne s\u00f6yledi\u011finin d\u00f6k\u00fcm\u00fcn\u00fc, yazma durumunda veya i\u00e7erik hakk\u0131ndaki bilginin \u00e7\u0131kt\u0131s\u0131n\u0131n al\u0131nabilece\u011fi eskizlerin yap\u0131land\u0131r\u0131lm\u0131\u015f bir g\u00f6sterimidir. Bununla birlikte, yazma ve \u00e7izim y\u00f6ntemlerini teknolojik yollarla yakalamak, insan g\u00f6zlemcilerin yazma h\u0131z\u0131, ritmi ve bask\u0131 seviyesi gibi kolayca tespit edemedi\u011fi bilgileri kullanma kap\u0131s\u0131n\u0131 a\u00e7ar. \u00d6\u011frenmeyi anlamadaki de\u011ferleri hala net olmasa da iyi bir uzman tahminci olabileceklerine dair g\u00f6stergeler vard\u0131r (Ochoa vd., 2013).<\/span><\/p>\n

Yazmak ve eskiz yapmak i\u00e7in en yayg\u0131n kullan\u0131lan kay\u0131t arac\u0131 dijital bir kalemdir (Oviatt ve Cohen, 2015). Bu kalemler farkl\u0131 y\u00fczeylerde yap\u0131lan vuru\u015flar\u0131n pozisyonunu, s\u00fcresini ve bas\u0131nc\u0131n\u0131 say\u0131salla\u015ft\u0131rabilir. Dijital formda olduktan sonra, bu bilgi \u00d6A ara\u00e7lar\u0131nda kullan\u0131labilir. Alternatif olarak, e\u011fitimde tabletlerin yayg\u0131n olarak kullan\u0131lmas\u0131 (Clarke ve Svanaes, 2014), \u00f6zellikle eskizlerde, bu modlar\u0131 kolayca yakalamak i\u00e7in bir f\u0131rsat sunmaktad\u0131r.<\/span><\/p>\n

\u00c7M\u00d6A alan\u0131nda, matematiksel veri derlemine dayanan iki \u00e7al\u0131\u015fma (Oviatt, Cohen ve Weibel, 2013), yazma ve eskiz modlar\u0131n\u0131n uzmanl\u0131k tahmine olabilecek katk\u0131s\u0131n\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Ochoa vd. (2013) yaz\u0131 karakteristiklerini (vuru\u015f h\u0131z\u0131 ve s\u00fcresi) \u00e7\u0131karm\u0131\u015f ve kullan\u0131lan basit geometrik fig\u00fcrlerin say\u0131s\u0131n\u0131 belirlemek i\u00e7in eskiz tan\u0131ma i\u015flemlerini ger\u00e7ekle\u015ftirmi\u015ftir. Sonu\u00e7lar, yazma h\u0131z\u0131n\u0131n uzmanl\u0131k d\u00fczeyi ile olduk\u00e7a ili\u015fkili oldu\u011funu belirlemi\u015ftir. Zhou, Hang, Oviatt, Yu ve Chen (2014), gruptaki uzman\u0131 %80 do\u011frulukla tan\u0131mlamak i\u00e7in yazma ve eskiz \u00f6zelliklerine dayal\u0131 s\u0131n\u0131fland\u0131rma sistemlerini kullanm\u0131\u015ft\u0131r.<\/span><\/p>\n\n

BA\u011eLAMLAR<\/h2>\n

\u00c7M\u00d6A ara\u015ft\u0131rmas\u0131n\u0131n temel amac\u0131, \u00d6A ara\u00e7lar\u0131n\u0131n ve metodolojilerinin, kolayca dijital izler sa\u011flamayan \u00f6\u011frenme ba\u011flamlar\u0131na uygulanmas\u0131n\u0131 geni\u015fletmektir. Bu ba\u011flamlar\u0131n bir \u00f6zelli\u011fi, \u00f6\u011frenme s\u00fcrecini anlamak i\u00e7in birden fazla modun yakalanmas\u0131n\u0131n gerekli olmas\u0131d\u0131r. Tablo 11.1, mevcut \u00c7M\u00d6A literat\u00fcr\u00fcnde incelenen ba\u011flam\u0131n, kullan\u0131lan modlar\u0131n, bu ba\u011flamlarda ara\u015ft\u0131r\u0131lan temel \u00f6\u011frenme y\u00f6nlerinin ve bu \u00e7al\u0131\u015fmalar\u0131n yap\u0131ld\u0131\u011f\u0131 \u00e7al\u0131\u015fmalar\u0131n ayr\u0131nt\u0131lar\u0131yla birlikte bir \u00f6zetini sunmaktad\u0131r.<\/span><\/p>\n

<\/a><\/a><\/a><\/a><\/a> Tablo 11.1. \u00c7M\u00d6A Taraf\u0131ndan \u00c7al\u0131\u015f\u0131lan \u00d6\u011frenme Ba\u011flamlar\u0131<\/i><\/span><\/span><\/p>\n\n<\/colgroup>\n\n\n\n\n\n\n\n
\n

Ba\u011flam<\/span><\/p>\n<\/td>\n

\n

Modlar<\/span><\/p>\n<\/td>\n

\n

\u00d6\u011frenme Y\u00f6nleri<\/span><\/p>\n<\/td>\n

\n

\u00c7al\u0131\u015fmalar<\/span><\/p>\n<\/td>\n<\/tr>\n

Dersler<\/span><\/span><\/td>\nHareket, Bak\u0131\u015f, Hareketler, Y\u00fcz \u0130fadesi, Konu\u015fma<\/span><\/span><\/td>\nDikkat, Soru-Cevap Etkile\u015fimleri<\/span><\/span><\/td>\nRaca ve Dillenbourg, 2013; Raca vd., 2014; Dominguez vd., 2015; D'Mello vd., 2015; Alibali vd., 2011<\/span><\/span><\/td>\n<\/tr>\n
S\u00f6zl\u00fc Sunumlar<\/span><\/span><\/td>\nDuru\u015f, Hareket, Jestler, Bak\u0131\u015f, Konu\u015fma, Dijital Belge<\/span><\/span><\/td>\nBeceri Geli\u015ftirme, Geribildirim, Zihinsel Durum<\/span><\/span><\/td>\nLuzardo vd., 2014; Echeverria vd., 2014; Chen vd., 2014; Leong vd., 2015; Schneider vd., 2015;<\/span><\/span>\n\nBoncoddo vd., 2013<\/span><\/span><\/td>\n<\/tr>\n
Problem \u00e7\u00f6zme<\/span><\/span><\/td>\nHareket, Eylemler, Konu\u015fma, Yazma, Eskiz<\/span><\/span><\/td>\nUzmanl\u0131k Tahmini<\/span><\/span><\/td>\nOchoa vd., 2013; Luz, 2013; Thompson, 2013;<\/span><\/span>\n\nZhou vd., 2014<\/span><\/span><\/td>\n<\/tr>\n
Yap\u0131 Al\u0131\u015ft\u0131rmalar\u0131<\/span><\/span><\/td>\nHareketler, Eylemler, Konu\u015fma, Y\u00fcz \u0130fadeleri, Galvanik Cilt Tepkisi<\/span><\/span><\/td>\nYeniye kar\u015f\u0131 Uzman Kal\u0131plar\u0131<\/span><\/span><\/td>\nWorsley ve Blikstein, 2013, 2014b, 2015a<\/span><\/span><\/td>\n<\/tr>\n
Ak\u0131ll\u0131 \u00d6\u011freticilerin Kullan\u0131m\u0131<\/span><\/span><\/td>\nDijital G\u00fcnl\u00fck Dosyalar\u0131, Y\u00fcz \u0130fadeleri, Konu\u015fma<\/span><\/span><\/td>\nDuygular ve \u00d6\u011frenme \u0130li\u015fkisi<\/span><\/span><\/td>\nCraig vd., 2008; D'Mello vd., 2008<\/span><\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Dersler<\/h3>\n

Geleneksel olarak, dersler \u00f6\u011frenme ile ilgili en yayg\u0131n ba\u011flamd\u0131r. Bu ortam\u0131n \u00e7e\u015fitli y\u00f6nleri \u00e7al\u0131\u015fmay\u0131 hak ederken, \u00c7M\u00d6A ara\u015ft\u0131rmac\u0131lar\u0131 bug\u00fcne kadar derste \u00f6\u011frencilerin dikkatini otomatik olarak de\u011ferlendirmeye odaklanm\u0131\u015ft\u0131r. Raca ve Dillenbourg (2013) ve Raca vd. (2014), s\u0131n\u0131fta video kayd\u0131 ile bu kay\u0131ttan \u00f6\u011frenen hareketinin bak\u0131\u015f\u0131n\u0131n otomatik olarak \u00e7\u0131kar\u0131lmas\u0131n\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Bu \u00e7al\u0131\u015fmalar\u0131n sonu\u00e7lar\u0131, her iki modun da \u00f6\u011frenenin dikkatiyle ili\u015fkili oldu\u011funu ancak oturma pozisyonu gibi di\u011fer \u00f6nemli katk\u0131 sa\u011flay\u0131c\u0131lar\u0131n oldu\u011funu g\u00f6stermektedir. Dominguez, Echeverrfa, Chiluiza ve Ochoa (2015), \u00e7oklu model bir kay\u0131t arac\u0131 (\u00c7MKA) kullanarak video, ses ve kalem tabanl\u0131 modlar\u0131 yakalamak i\u00e7in yeni ve da\u011f\u0131t\u0131lm\u0131\u015f bir yol sundu. \u015eekil 11.5, b\u00f6yle bir arac\u0131n tasar\u0131m\u0131n\u0131 sunar. Cihaz\u0131n \u00f6\u011frencilere yak\u0131nl\u0131\u011f\u0131 t\u0131kanma riskini azalt\u0131r ve video ve ses yakalama kalitesini artt\u0131r\u0131r. Son olarak, D\u2019Mello vd. (2015), \u00f6\u011fretim \u00fcyesi ve \u00f6\u011frenciler aras\u0131ndaki soru-cevap etkile\u015fimlerini de\u011ferlendirmek i\u00e7in bir konferans ortam\u0131nda \u00e7e\u015fitli ses kay\u0131tlar\u0131 \u00fcretmi\u015ftir.<\/span><\/p>\n

\"\"<\/p>\n<\/a> \u015eekil 11.5. Ders ayarlar\u0131nda kullan\u0131lacak \u00e7ok modlu bir kay\u0131t arac\u0131n\u0131n (\u00c7MKA) tasar\u0131m\u0131. \u00c7MKA s\u0131n\u0131fta (solda) ve \u00c7MKA \u00f6\u011frenenin bak\u0131\u015f a\u00e7\u0131s\u0131ndan (sa\u011fda)<\/i><\/span><\/span>\n

S\u00f6zl\u00fc Sunumlar<\/h3>\n

Bir akademik konuyu izleyicinin \u00f6n\u00fcnde sunma becerisi, genellikle y\u00fcksek\u00f6\u011frenim \u00f6\u011frencilerinin edinmesi gereken sosyal becerilerden biri olarak kabul edilir (Debnath vd., 2012). D\u00fcnyan\u0131n d\u00f6rt bir yan\u0131ndaki bir\u00e7ok ba\u011f\u0131ms\u0131z grup, yeni \u00f6\u011frencilerin k\u00f6t\u00fc uygulamalar\u0131 d\u00fczeltmelerine ve s\u00f6zl\u00fc sunumlarda ustal\u0131k kazanmalar\u0131na yard\u0131mc\u0131 olacak \u00c7M\u00d6A sistemleri kurmaya ba\u015flad\u0131. Echeverrfa vd. (2014) ve Luzardo vd. (2014) ayn\u0131 sistemin jest, duru\u015f, hareket, bak\u0131\u015f, konu\u015fma ve dijital sunum dosyalar\u0131n\u0131n analizini kullanan ve bir insan de\u011ferlendiricinin \u00f6\u011frenciye verece\u011fi notu tahmin edebilen farkl\u0131 y\u00f6nlerini sunmaktad\u0131r. Ayn\u0131 verileri analiz eden Chen, Leong, Feng ve Lee (2014), ayn\u0131 y\u00f6ntemi tahmin etmekte kullan\u0131lan bile\u015fik de\u011fi\u015fkenlerdeki farkl\u0131 y\u00f6ntemlerle de birle\u015ftirebilmi\u015ftir. B\u00f6rner, van Rosmalen ve Specht (2015) Kinect'i pozlar\u0131 tan\u0131mak ve ger\u00e7ek zamanl\u0131 olarak geri bildirim sa\u011flamak i\u00e7in kullanan sanal bir sunum beceri e\u011fitici4<\/a><\/sup> olu\u015fturdu.<\/span><\/p>\n\n

Problem \u00c7\u00f6zme<\/h3>\n

\u00d6zellikle STEM derslerinde \u00f6\u011frenme s\u0131kl\u0131kla, bireysel ve grupla problem \u00e7\u00f6zme oturumlar\u0131nda g\u00f6r\u00fcl\u00fcr (Silver, 2013). Matematik ve geometri problemlerini \u00e7\u00f6zen \u00fc\u00e7 farkl\u0131 lise \u00f6\u011frenci grubunun \u00e7oklu model kay\u0131tlar\u0131ndan olu\u015fan bir dizi matematiksel veri derleminin (Oviatt vd., 2013) varl\u0131\u011f\u0131, bu ortamda \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131n\u0131 h\u0131zland\u0131rd\u0131. Veri k\u00fcmesinde yer alan medya, her \u00f6\u011frenenin \u00f6nden video kayd\u0131n\u0131, \u00e7al\u0131\u015fma masas\u0131n\u0131n video kayd\u0131n\u0131, her \u00f6\u011frenenin ses kayd\u0131n\u0131 ve odan\u0131n genel sesini i\u00e7erir. Ayr\u0131ca, \u00f6\u011frencilerin dijital kalemler ile donat\u0131lm\u0131\u015flard\u0131r. \u00d6\u011frencilerin uzmanl\u0131k d\u00fczeyi ile ilgili ger\u00e7ek referans de\u011fer sa\u011flanm\u0131\u015ft\u0131r. Luz (2013), Thompson (2013), Ochoa vd. (2013) ve Zhou vd. (2014) bu veri k\u00fcmesini farkl\u0131 modlar kullanarak analiz etmi\u015f, t\u00fcm modalitelerin uzmanl\u0131k seviyesinin y\u00fcksek bir do\u011fruluk d\u00fczeyiyle (> %70) belirlenmesine katk\u0131da bulunmu\u015ftur.<\/span><\/p>\n\n

Yap\u0131 Al\u0131\u015ft\u0131rmalar\u0131<\/h3>\n

M\u00fchendislik tasar\u0131m\u0131 ve in\u015fas\u0131 i\u00e7in gerekli bilgi ve beceriler, k\u00fc\u00e7\u00fck yap\u0131 zorluklar\u0131 ile test edilebilir (Householder ve Hailey, 2012). Worsley ve Blikstein'\u0131n (2013, 2014b, 2015a) son derece \u00f6nemli eserleri, \u00e7ok modlu analizlerle, yap\u0131lar\u0131n tasar\u0131m\u0131 ve elle montaj\u0131nda uzmanlar ve yeni ba\u015flayanlar taraf\u0131ndan ger\u00e7ekle\u015ftirilen eylem \u00f6r\u00fcnt\u00fclerini ara\u015ft\u0131r\u0131yor. Analiz i\u00e7in kullan\u0131lan ana modlar; hareketler, eylemler, konu\u015fma, y\u00fcz ifadesi ve galvanik cilt tepkisi idi. Bu modlardan \u00e7\u0131kar\u0131lan izlerin kombinasyonu, yap\u0131m s\u00fcrecinde m\u00fchendislik tasar\u0131m\u0131ndaki ustal\u0131k seviyesini tan\u0131mlamaya yard\u0131mc\u0131 olan farkl\u0131l\u0131klar\u0131 ortaya koyuyor.<\/span><\/p>\n\n

Ak\u0131ll\u0131 \u00d6\u011freticilerin Kullan\u0131m\u0131<\/h3>\n

A\u00d6S'ler genellikle geleneksel \u00d6A taraf\u0131ndan kay\u0131t g\u00fcnl\u00fc\u011f\u00fc dosyalar\u0131 kullan\u0131larak incelenir. Bununla birlikte, etkile\u015fim verilerini tamamlayan yeni modlar eklemek i\u00e7in \u00f6\u011frenenin videosu ve sesi \u00e7ekilmi\u015ftir. Videodan \u00e7\u0131kar\u0131lan ana modlar y\u00fcz ifadesi (Craig vd., 2008) ve konu\u015fma (D'Mello vd., 2008) olup, \u00f6\u011frenenin i\u00e7sel duygusal durumunu temsil eder. Her ikisi de A\u00d6S'yi kullan\u0131rken can s\u0131k\u0131nt\u0131s\u0131, kar\u0131\u015f\u0131kl\u0131k ve hayal k\u0131r\u0131kl\u0131\u011f\u0131 gibi duygusal durumlar\u0131 ba\u015far\u0131yla tespit edebilir.<\/span><\/p>\n\n

\u00d6ZEL KONULAR<\/h2>\n

\u00d6A modellerinde ve uygulamalar\u0131nda \u00e7ok modelli izlerin kullan\u0131lmas\u0131, ayn\u0131 moddan elde edilen farkl\u0131 izlerin kullan\u0131lmas\u0131na benzer. Bununla birlikte, \u00c7M\u00d6A ara\u015ft\u0131rma ve uygulamalar\u0131, belirli modaliteler yakaland\u0131\u011f\u0131nda, i\u015flendi\u011finde ve analiz edildi\u011finde baz\u0131 spesifik sorunlar\u0131 ortaya \u00e7\u0131karmaktad\u0131r. Bu meseleler, bir\u00e7ok modaliteden izlerin teknik olarak \u00e7\u0131kar\u0131lmas\u0131na paralel olarak ancak \u00c7M\u00d6A \u00e7\u00f6z\u00fcmlerinin ger\u00e7ek d\u00fcnyada etkili bir \u015fekilde konu\u015fland\u0131r\u0131lmas\u0131 i\u00e7in \u00f6nemli olan a\u00e7\u0131k ara\u015ft\u0131rma alanlar\u0131 olmaya devam etmektedir.<\/span><\/p>\n\n

Kay\u0131t<\/h3>\n

Dijital bir ara\u00e7ta etkile\u015fim bilgilerini yakalamak, kodun ilgili k\u0131s\u0131mlar\u0131na g\u00fcnl\u00fck ifadeleri eklemek kadar kolay ve ucuzdur. Bu ifadeler, \u00f6\u011frenenden herhangi bir kat\u0131l\u0131m gerektirmeden, \u015feffaf ve genel olarak g\u00fcvenilir ve hatas\u0131z bir \u015fekilde otomatik olarak ger\u00e7ekle\u015ftirilir. \u00d6te yandan, ger\u00e7ek d\u00fcnyadaki medyay\u0131 yakalamak, kay\u0131t ara\u00e7lar\u0131n\u0131n (kameralar, mikrofonlar, dijital kalemler, vb.) edinilmesini, kurulmas\u0131n\u0131 ve kullan\u0131lmas\u0131n\u0131, sistemi a\u00e7\u0131p kapatmas\u0131n\u0131 ve izlenmesini ve t\u0131kan\u0131kl\u0131k, parazit veya g\u00fcr\u00fclt\u00fc yoluyla kayd\u0131n bozulmas\u0131n\u0131 \u00f6nlemeyi gerektirir. Dijital kay\u0131t kadar zahmetsiz ve verimli \u00e7al\u0131\u015fan kay\u0131t sistemlerinin geli\u015ftirilmesi, \u00c7M\u00d6A'n\u0131n geli\u015fmesinin \u00f6n\u00fcndeki en b\u00fcy\u00fck engellerden biridir. Bu bir m\u00fchendislik problemi olsa da ara\u015ft\u0131rmac\u0131lar \u00e7\u00f6z\u00fcmlerinin uygulanabilirli\u011fi ve \u00f6l\u00e7eklenebilirli\u011finin fark\u0131nda olmal\u0131d\u0131rlar. Ana tekliflerden biri, kay\u0131tlar\u0131n her zaman a\u00e7\u0131kta kalan ucuz alg\u0131lay\u0131c\u0131lar\u0131 kullanarak merkezden da\u011f\u0131t\u0131lmas\u0131d\u0131r. Bir veya daha fazla kay\u0131t sorun \u00e7\u0131kar\u0131rsa, genel bilgiler kalan \u00e7al\u0131\u015fma alg\u0131lay\u0131c\u0131lar\u0131ndan yeniden olu\u015fturulabilir.<\/span><\/p>\n\n

Gizlilik<\/h3>\n

Etkile\u015fim bilgilerini dijital ara\u00e7larla yakalamak, \u00f6\u011frenciler ve \u00f6\u011fretenler aras\u0131ndaki \u015fimdiden mahremiyet endi\u015felerini ortaya \u00e7\u0131karmaktad\u0131r (Pardo ve Siemens, 2014). \u201c19845<\/a><\/sup>\u201d g\u00f6zetim seviyelerini taklit eden kay\u0131t sistemlerinin kurulmas\u0131 ve kullan\u0131lmas\u0131n\u0131n, g\u00fc\u00e7l\u00fc bir diren\u00e7 ortaya \u00e7\u0131karaca\u011f\u0131 muhakkakt\u0131r. A\u00e7\u0131k r\u0131za formlar\u0131 erken ara\u015ft\u0131rma a\u015famalar\u0131 i\u00e7in i\u015fe yarayabilirdi ancak ger\u00e7ek d\u00fcnyada \u00c7M\u00d6A sistemlerinin benimsenmesi farkl\u0131 ve daha yarat\u0131c\u0131 bir yakla\u015f\u0131m gerektiriyordu. Bu alandaki en umut verici \u00e7\u00f6z\u00fcmlerden biri, verinin m\u00fclkiyetini \u00f6\u011frenene aktarmakt\u0131r. En mahrem ki\u015fisel bilgileriniz al\u0131nsa bile, neyin ne zaman payla\u015f\u0131laca\u011f\u0131 \u00f6\u011frencinin kontrol\u00fcnde kal\u0131rsa gizlilik kayg\u0131lar\u0131 etkisiz hale getirilir. Bu yakla\u015f\u0131m, \u00e7e\u015fitli niceliksel kendi kendine yap\u0131lan uygulamalara benzer (Swan, 2013). <\/span><\/p>\n\n

Entegrasyon<\/h4>\n

B\u00fcy\u00fck miktarda ham \u00f6\u011frenme izlerinin mevcudiyeti ile ilgili bir soru, \u00f6\u011frenme s\u00fcrecini anlamak, optimize etmek ve yararl\u0131 bilgiler \u00fcretmek i\u00e7in bunlar\u0131 nas\u0131l birle\u015ftirece\u011fimizdir. Farkl\u0131 i\u015flemler kullan\u0131larak farkl\u0131 modlardan \u00e7\u0131kar\u0131lan izler \u00e7ok farkl\u0131 \u00f6zelliklere sahip olmak zorundad\u0131r. \u00d6rne\u011fin, farkl\u0131 modlardan \u00e7\u0131kar\u0131lan izlerin zaman taneciklili\u011fi geni\u015f \u00f6l\u00e7\u00fcde de\u011fi\u015febilir. Konu\u015fman\u0131n prosodik y\u00f6nlerinden \u00e7\u0131kar\u0131lan izler saniyenin onda birinde de\u011fi\u015febilirken, duru\u015flar daha yava\u015f de\u011fi\u015fiyor. \u00c7\u0131kar\u0131lan izlerin kesinli\u011fi de farkl\u0131 olabilir. Y\u00fcksek kaliteli kay\u0131tlarla yap\u0131lan konu\u015fma tan\u0131ma %90 do\u011frulu\u011fa ula\u015f\u0131rken, web kameras\u0131 kaynaklar\u0131ndan gelen duygusal durum tespiti 70'lerin alt\u0131nda kalabilir. Bu fark ba\u015far\u0131l\u0131 analizleri engellemez ancak sahte sonu\u00e7lar\u0131 \u00f6nlemek i\u00e7in duyarl\u0131 tasar\u0131m gereklidir. \u00c7M\u00d6A, Worsley ve Blikstein (2014a) 'daki bu ara\u015ft\u0131rma hatt\u0131na \u00f6nc\u00fcl\u00fck etmek, Newell (1994) ve Anderson (2002) taraf\u0131ndan insan bili\u015fine bir a\u00e7\u0131klama olarak \u00f6nerilen \u201cbili\u015f bantlar\u0131\u201d \u00e7er\u00e7evesine dayanan \u00e7e\u015fitli birle\u015ftirme stratejileri \u00f6nermektedir. Entegrasyon \u00e7er\u00e7evelerinin geli\u015ftirilmesi sadece \u00c7M\u00d6A\u2019ya de\u011fil t\u00fcm \u00d6A toplulu\u011funa fayda sa\u011flayacakt\u0131r.<\/span><\/p>\n\n

\u00d6\u011frenmeye Etkisi<\/h3>\n

\u00c7ok kullan\u0131c\u0131l\u0131 \u00f6\u011frenme analitiklerine dayanan son kullan\u0131c\u0131 ara\u00e7lar\u0131 ve m\u00fcdahaleleri, tek modlu analizlere dayananlara benzer olsa da \u00e7ok modlu olanlar i\u00e7in gerekli kullan\u0131\u015fl\u0131l\u0131k, veri toplama ek karma\u015f\u0131kl\u0131\u011f\u0131n\u0131 hakl\u0131 \u00e7\u0131karmak i\u00e7in daha y\u00fcksek olmal\u0131d\u0131r. \u00d6rne\u011fin, \u00d6YS taraf\u0131ndan otomatik olarak toplanan verilere dayal\u0131 bir g\u00f6sterge tablosu uygulamas\u0131n\u0131n kabul edilmesi, t\u00fcm s\u0131n\u0131flar\u0131n video kameralarla donat\u0131lmas\u0131n\u0131 gerektiren benzer bir g\u00f6sterge panelinden daha kolay olacakt\u0131r. Artan karma\u015f\u0131kl\u0131\u011fa, \u00f6\u011frenme s\u00fcreci \u00fczerinde daha b\u00fcy\u00fck bir olumlu etki e\u015flik etmelidir. \u00d6\u011frenmeyi analiz etmek i\u00e7in \u00e7oklu ger\u00e7ek d\u00fcnya sinyallerini kullanma ihtiyac\u0131, s\u00fcre\u00e7 hakk\u0131nda daha yararl\u0131 bilgiler ve \u00f6\u011frenenler \u00fczerinde daha \u00f6l\u00e7\u00fclebilir etkiler sa\u011flama vaadini de sunmal\u0131d\u0131r.<\/span><\/p>\n\n

SONU\u00c7<\/h2>\n

\u00d6A, \u00f6\u011frenme s\u00fcrecini anlamak ve optimize etmek i\u00e7in kullan\u0131lan yakla\u015f\u0131mlarda devrim yaratt\u0131. Ancak yaln\u0131zca bilgisayar tabanl\u0131 \u00f6\u011frenmeyi i\u00e7eren \u00e7al\u0131\u015fmalara ve ara\u00e7lara y\u00f6nelik mevcut yanl\u0131l\u0131\u011f\u0131, genel olarak \u00f6\u011frenmeye uygulanabilirli\u011fini tehlikeye atmaktad\u0131r. \u00c7M\u00d6A, bilgisayar destekli olmayan \u00f6\u011frenme ba\u011flamlar\u0131n\u0131 \u00d6A'n\u0131n ana ara\u015ft\u0131rma ve uygulamas\u0131na entegre etmeye \u00e7al\u0131\u015fan bir alt aland\u0131r.<\/span><\/p>\n

Bu b\u00f6l\u00fcm \u00c7M\u00d6A\u2019daki en son teknolojiyi sunmaktad\u0131r. Daha geleneksel t\u0131klama ak\u0131\u015f bilgisi ve metin i\u00e7eri\u011fi modlar\u0131n\u0131n yan\u0131 s\u0131ra duru\u015f, konu\u015fma ve eskiz gibi \u00e7e\u015fitli modlar, ara\u015ft\u0131rma sorular\u0131n\u0131 cevaplamak ve \u00f6\u011frenme ba\u011flamlar\u0131nda geri bildirim sistemleri olu\u015fturmak i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r. Bilgisayar bilimi tekniklerinin ve e\u011fitim ve davran\u0131\u015f bilimcilerinin sa\u011flad\u0131\u011f\u0131 i\u00e7g\u00f6r\u00fclerden olu\u015fan bir birliktelik, s\u0131n\u0131flar, \u00e7al\u0131\u015fma gruplar\u0131 ve s\u00f6zl\u00fc sunumlar gibi \u00e7ok \u00e7e\u015fitli \u00f6\u011frenme ba\u011flamlar\u0131n\u0131n otomatik olarak de\u011ferlendirilmesini sa\u011flar.<\/span><\/p>\n

Bu b\u00f6l\u00fcmde sunulan ara\u015ft\u0131rma listesinden \u00e7\u0131kar\u0131labilece\u011fi gibi, \u00c7M\u00d6A hala k\u00fc\u00e7\u00fck ama \u00e7ok aktif ve a\u00e7\u0131k bir ara\u015ft\u0131rmac\u0131 toplulu\u011funa sahip olan yeni bir aland\u0131r. \u00c7ok modlu veri k\u00fcmelerinin serbest\u00e7e payla\u015f\u0131ld\u0131\u011f\u0131 ve ortakla\u015fa tart\u0131\u015f\u0131lan yeni tasar\u0131m fikirleriyle birlikte analiz edildi\u011fi d\u00fczenli zorluklar\u0131n ve at\u00f6lye \u00e7al\u0131\u015fmalar\u0131n\u0131n varl\u0131\u011f\u0131, yeni bilgilerin h\u0131zla \u00fcretildi\u011fi bir ara\u015ft\u0131rma ortam\u0131 yarat\u0131r.<\/span><\/p>\n

Baz\u0131 konular hala \u00c7M\u00d6A'n\u0131n ana uygulama haline gelmesini engellerken, aktif ara\u015ft\u0131rma projeleri bu konulara y\u00f6nelik \u00e7\u00f6z\u00fcmleri ara\u015ft\u0131rmaktad\u0131r, \u00e7oklu model \u00f6\u011frenme izlerini daha ucuz, daha az m\u00fcdahaleci ve daha otomatik hale getirmektedir. \u00c7M\u00d6A toplulu\u011fundan do\u011fan, mahremiyetle ilgili kayg\u0131lar\u0131 gidermek i\u00e7in ortaya \u00e7\u0131kan yeni \u00e7\u00f6z\u00fcmler, \u00f6rne\u011fin da\u011f\u0131t\u0131lm\u0131\u015f kay\u0131t sa\u011flamak ve verilerin \u00f6\u011frenenle birlikte dinlenmesi gibi, bir g\u00fcn genel \u00d6A uygulamalar\u0131 i\u00e7in normlar\u0131 olu\u015fturabilir.<\/span><\/p>\n

Son olarak, yazar \u00d6A ara\u015ft\u0131rmac\u0131lar\u0131n\u0131 ve uygulay\u0131c\u0131lar\u0131n\u0131 kendi \u00e7al\u0131\u015fmalar\u0131nda ve ara\u00e7lar\u0131nda birden fazla y\u00f6ntemi kullanmay\u0131 ke\u015ffetmeye davet ediyor. \u00c7M\u00d6A toplulu\u011fu, bilgilerini, verilerini, kodunu ve \u00e7er\u00e7evelerini a\u00e7\u0131k\u00e7a payla\u015facakt\u0131r. Sadece bu farkl\u0131 y\u00f6ntemlerin benimsenmesi, \u00d6A'n\u0131n \u00f6\u011frenmenin ger\u00e7ekle\u015fti\u011fi t\u00fcm ba\u011flamlarda bir etkiye sahip olmas\u0131na izin verecektir.<\/span><\/p>\n\n

KAYNAK\u00c7A<\/h2>\nAlibali, M. W., Nathan, M. J., Fujimori, Y., Stein, N., & Raudenbush, S. (2011). Gestures in the mathematics classroom: What\u2019s the point? In N. Stein & S. Raudenbush (Eds.), Developmental cognitive science goes to school <\/i>(pp. 219\u2013234). New York: Routledge, Taylor & Francis. <\/span><\/span>\n\nAlmeda, M. V., Scupelli, P., Baker, R. S., Weber, M., & Fisher, A. (2014). Clustering of design decisions in classroom visual displays. Proceedings of the 4th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201914), 24\u201328 March 2014, Indianapolis, IN, USA (pp. 44\u201348). New York: ACM. <\/span><\/span>\n\nAnderson, J. R. (2002). Spanning seven orders of magnitude: A challenge for cognitive modeling. Cognitive Science, 26<\/i>(1), 85\u2013112. <\/span><\/span>\n\nArnold, K. E., & Pistilli, M. D. (2012). Course Signals at Purdue: Using learning analytics to increase student success. Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 267\u2013270). New York: ACM. <\/span><\/span>\n\nBlikstein, P. (2013). Multimodal learning analytics. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 102\u2013106). New York: ACM. <\/span><\/span>\n\nBoraston, Z., & Blakemore, S.-J. (2007). The application of eye-tracking technology in the study of autism. The Journal of Physiology, 581<\/i>(3), 893\u2013898. <\/span><\/span>\n\nBull, P. E. (2013). Posture & Gesture<\/i>. Elsevier. <\/span><\/span>\n\nBoncoddo, R., Williams, C., Pier, E., Walkington, C., Alibali, M., Nathan, M., Dogan, M. & Waala, J. (2013). Gesture as a window to justification and proof. In M. V. Martinez & A. C. Superfine (Eds.), Proceedings of the 35th annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education <\/i>(PME-NA 35) 14\u201317 November 2013, Chicago, IL, USA (pp. 229\u2013236). http:\/\/www.pmena.org\/proceedings\/ <\/span><\/span>\n\nChen, L., Leong, C. W., Feng, G., & Lee, C. M. (2014). Using multimodal cues to analyze MLA\u201914 oral presentation quality corpus: Presentation delivery and slides quality. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 45\u201352). New York: ACM. <\/span><\/span>\n\nClarke, B., & Svanaes, S. (2014, April 9). An updated literature review on the use of tablets in education. Family Kids and Youth. https:\/\/smartfuse.s3.amazonaws.com\/mysandstorm.org\/uploads\/2014\/05\/T4S-Use-of- Tablets-in-Education.pdf <\/span><\/span>\n\nCobb, P., Confrey, J., Lehrer, R., Schauble, L., & others. (2003). Design experiments in educational research. Educational Researcher, 32<\/i>(1), 9\u201313.<\/span><\/span>\n\nCraig, S. D., D\u2019Mello, S., Witherspoon, A., & Graesser, A. (2008). Emote aloud during learning with AutoTutor: Applying the facial action coding system to cognitive\u2013affective states during learning. Cognition and Emotion, 22<\/i>(5), 777\u2013788. <\/span><\/span>\n\nCrossley, S. A., Roscoe, R. D., & McNamara, D. S. (2013). Using automatic scoring models to detect changes in student writing in an intelligent tutoring system. Proceedings of the 26th Annual Florida Artificial Intelligence Research Society Conference <\/i>(FLAIRS-13), 20\u201322 May 2013, St. Pete Beach, FL, USA (pp. 208\u2013213). Menlo Park, CA: The AAAI Press. <\/span><\/span>\n\nDebnath, M., Pandey, M., Chaplot, N., Gottimukkula, M. R., Tiwari, P. K., & Gupta, S. N. (2012). Role of soft skills in engineering education: Students\u2019 perceptions and feedback. In C. S. Nair, A. Patil, & P. Mertova (Eds.), Enhancing learning and teaching through student feedback in engineering <\/i>(pp. 61\u201382). ScienceDirect. http: \/\/ www.sciencedirect.com\/science\/book\/9781843346456 <\/span><\/span>\n\nD\u2019Mello, S. K., Jackson, G. T., Craig, S. D., Morgan, B., Chipman, P., White, H., Person, N., Kort, B., el Kaliouby, R., Picard, R., & Graesser, A. C. (2008). AutoTutor detects and responds to learners\u2019 affective and cognitive states. Workshop on Emotional and Cognitive Issues in ITS, held in conjunction with the 9th International Conference on Intelligent Tutoring Systems (ITS 2008), 23\u201327 June 2008, Montreal, PQ, Canada. https:\/\/ www.researchgate.net\/publication\/228673992_AutoTutor_detects_and_responds_to_learners_affective_and_cognitive_states <\/span><\/span>\n\nD\u2019Mello, S., Olney, A., Blanchard, N., Samei, B., Sun, X., Ward, B., & Kelly, S. (2015). Multimodal capture of teacher\u2013student interactions for automated dialogic analysis in live classrooms. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 557\u2013566). New York: ACM. <\/span><\/span>\n\nDominguez, F., Echeverr\u00eda, V., Chiluiza, K., & Ochoa, X. (2015). Multimodal selfies: Designing a multimodal recording device for students in traditional classrooms. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 567\u2013574). New York: ACM. <\/span><\/span>\n\nEcheverr\u00eda, V., Avenda\u00f1o, A., Chiluiza, K., V\u00e1squez, A., & Ochoa, X. (2014). Presentation skills estimation based on video and Kinect data analysis. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 53\u201360). New York: ACM. <\/span><\/span>\n\nFreedman, D. H. (2010, December 10). Why scientific studies are so often wrong: The streetlight effect. Discover Magazine, 26<\/i>. http:\/\/discovermagazine.com\/2010\/jul-aug\/29-why-scientific-studies-often-wrong-streetlight-effect <\/span><\/span>\n\nFrischen, A., Bayliss, A. P., & Tipper, S. P. (2007). Gaze cueing of attention: Visual attention, social cognition, and individual differences. Psychological Bulletin, 133<\/i>(4), 694\u2013724. <\/span><\/span>\n\nGall, M. D., Borg, W. R., & Gall, J. P. (1996). Educational research: An introduction<\/i>. Longman Publishing. <\/span><\/span>\n\nGraesser, A. C., Chipman, P., Haynes, B. C., & Olney, A. (2005). AutoTutor: An intelligent tutoring system with mixed-initiative dialogue. IEEE Transactions on Education, 48<\/i>(4), 612\u2013618. <\/span><\/span>\n\nHouseholder, D. L., & Hailey, C. E. (2012). Incorporating engineering design challenges into STEM courses<\/i>. National Center for Engineering and Technology Education. http:\/\/ncete.org\/flash\/pdfs\/NCETECaucusReport.pdf <\/span><\/span>\n\nJewitt, C. (2006). Technology, literacy and learning: A multimodal approach. Psychology Press. <\/span><\/span>\n\nKizilcec, R. F., Piech, C., & Schneider, E. (2013). Deconstructing disengagement: Analyzing learner subpopulations in massive open online courses. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 170\u2013179). New York: ACM. <\/span><\/span>\n\nKress, G., & Van Leeuwen, T. (2001). Multimodal discourse: The modes and media of contemporary communication. Edward Arnold. <\/span><\/span>\n\nKrugman, D. M., Fox, R. J., Fletcher, J. E., Fischer, P. M., & Rojas, T. H. (1994). Do adolescents attend to warnings in cigarette advertising? An eye-tracking approach. Journal of Advertising Research, 34<\/i>, 39\u201351. <\/span><\/span>\n\nKruschke, J. K. (2003). Attention in learning. Current Directions in Psychological Science, 12<\/i>(5), 171\u2013175.<\/span><\/span>\n\nLeong, C. W., Chen, L., Feng, G., Lee, C. M., & Mulholland, M. (2015). Utilizing depth sensors for analyzing multimodal presentations: Hardware, software and toolkits. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 547\u2013556). New York: ACM. <\/span><\/span>\n\nLin, Y.-T., Lin, R.-Y., Lin, Y.-C., & Lee, G. C. (2013). Real-time eye-gaze estimation using a low-resolution webcam. Multimedia Tools and Applications, 65<\/i>(3), 543\u2013568. <\/span><\/span>\n\nLubold, N., & Pon-Barry, H. (2014). Acoustic-prosodic entrainment and rapport in collaborative learning dialogues. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 5\u201312). New York: ACM. <\/span><\/span>\n\nLund, K. (2007). The importance of gaze and gesture in interactive multimodal explanation. Language Resources and Evaluation, 41<\/i>(3\u20134), 289\u2013303. <\/span><\/span>\n\nLuzardo, G., Guam\u00e1n, B., Chiluiza, K., Castells, J., & Ochoa, X. (2014). Estimation of presentations skills based on slides and audio features. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 37\u201344). New York: ACM. <\/span><\/span>\n\nLuz, S. (2013). Automatic identification of experts and performance prediction in the multimodal math data corpus through analysis of speech interaction. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 575\u2013582). New York: ACM. <\/span><\/span>\n\nMarkaki, V., Lund, K., & Sanchez, E. (2015). Design digital epistemic games: A longitudinal multimodal analysis. Paper presented at the conference Revisiting Participation: Language and Bodies in Interaction<\/i>, 24\u201327 June 2015, Basel, Switzerland. <\/span><\/span>\n\nMazur-Palandre, A., Colletta, J. M., & Lund, K. (2014). Context sensitive \u201chow\u201d explanation in children\u2019s multimodal behavior, Journal of Multimodal Communication Studies, 2<\/i>, 1\u201317. <\/span><\/span>\n\nMishra, B., Fernandes, S. L., Abhishek, K., Alva, A., Shetty, C., Ajila, C. V., \u2026 Shetty, P. (2015). Facial expression recognition using feature based techniques and model based techniques: A survey. Proceedings of the 2nd International Conference on Electronics and Communication Systems <\/i>(ICECS 2015), 26\u201327 February 2015, Coimbatore, India (pp. 589\u2013594). IEEE. <\/span><\/span>\n\nMitra, S., & Acharya, T. (2007). Gesture recognition: A survey. IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, 37(3), 311\u2013324. <\/span><\/span>\n\nMorency, L.-P., Oviatt, S., Scherer, S., Weibel, N., & Worsley, M. (2013). ICMI 2013 grand challenge workshop on multimodal learning analytics. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 373\u2013378). New York: ACM. <\/span><\/span>\n\nNewell, A. (1994). Unified theories of cognition<\/i>. Harvard University Press. <\/span><\/span>\n\nOchoa, X., Chiluiza, K., M\u00e9ndez, G., Luzardo, G., Guam\u00e1n, B., & Castells, J. (2013). Expertise estimation based on simple multimodal features. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 583\u2013590). New York: ACM. <\/span><\/span>\n\nOchoa, X., Worsley, M., Chiluiza, K., & Luz, S. (2014). MLA\u201914: Third multimodal learning analytics workshop and grand challenges. Proceedings of the 16th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 531\u2013532). New York: ACM. <\/span><\/span>\n\nOviatt, S., Cohen, A., & Weibel, N. (2013). Multimodal learning analytics: Description of math data corpus for ICMI grand challenge workshop. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 583\u2013590). New York: ACM. <\/span><\/span>\n\nOviatt, S., & Cohen, P. R. (2015). The paradigm shift to multimodality in contemporary computer interfaces<\/i>. San Rafael, CA: Morgan & Claypool Publishers. <\/span><\/span>\n\nPardo, A., & Siemens, G. (2014). Ethical and privacy principles for learning analytics. British Journal of Educational Technology, 45<\/i>(3), 438\u2013450. <\/span><\/span>\n\nPoole, A., & Ball, L. J. (2006). Eye tracking in HCI and usability research. Encyclopedia of Human Computer Interaction, 1<\/i>, 211\u2013219.<\/span><\/span>\n\nRaca, M., & Dillenbourg, P. (2013). System for assessing classroom attention. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 265\u2013269). New York: ACM. <\/span><\/span>\n\nRaca, M., Tormey, R., & Dillenbourg, P. (2014). Sleepers\u2019 lag: Study on motion and attention. Proceedings of the 4th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201914), 24\u201328 March 2014, Indianapolis, IN, USA (pp. 36\u201343). New York: ACM. <\/span><\/span>\n\nScherer, S., Worsley, M., & Morency, L.-P. (2012). 1st international workshop on multimodal learning analytics. Proceedings of the 14th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201912), 22\u201326 October 2012, Santa Monica, CA, USA (pp. 609\u2013610). New York: ACM. <\/span><\/span>\n\nSchl\u00f6mer, T., Poppinga, B., Henze, N., & Boll, S. (2008). Gesture recognition with a Wii controller. Proceedings of the 2nd International Conference on Tangible and Embedded Interaction <\/i>(TEI\u201908), 18\u201321 February 2008, Bonn, Germany (pp. 11\u201314). New York: ACM. <\/span><\/span>\n\nSchneider, J., B\u00f6rner, D., van Rosmalen, P., & Specht, M. (2015). Presentation trainer, your public speaking multimodal coach. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 539\u2013546). New York: ACM. <\/span><\/span>\n\nSerrano-Laguna, A., & Fern\u00e1ndez-Manj\u00f3n, B. (2014). Applying learning analytics to simplify serious games deployment in the classroom. Proceedings of the 2014 IEEE Global Engineering Education Conference <\/i>(EDUCON 2014), 3\u20135 April 2014, Istanbul, Turkey (pp. 872\u2013877). IEEE. <\/span><\/span>\n\nSiemens, G. (2013). Learning analytics: The emergence of a discipline. American Behavioral Scientist, 9<\/i>, 51\u201360. doi:0002764213498851. <\/span><\/span>\n\nSilver, E.A. (2013). Teaching and learning mathematical problem solving: Multiple research perspectives. Routledge. <\/span><\/span>\n\nSimsek, D., S\u00e1ndor, \u00c1., Buckingham Shum, S., Ferguson, R., De Liddo, A., & Whitelock, D. (2015). Correlations between automated rhetorical analysis and tutors\u2019 grades on student essays. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 355\u2013359). New York: ACM. <\/span><\/span>\n\nSwan, M. (2013). The quantified self: Fundamental disruption in big data science and biological discovery. Big Data, 1<\/i>(2), 85\u201399. <\/span><\/span>\n\nThompson, K. (2013). Using micro-patterns of speech to predict the correctness of answers to mathematics problems: An exercise in multimodal learning analytics. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 591\u2013598). New York: ACM. <\/span><\/span>\n\nVasquez, H. A., Vargas, H. S., & Sucar, L. E. (2015). Using gestures to interact with a service robot using Kinect 2. Advances in Computer Vision and Pattern Recognition, 85<\/i>. Springer. <\/span><\/span>\n\nVatrapu, R., Reimann, P., Bull, S., & Johnson, M. (2013). An eye-tracking study of notational, informational, and emotional aspects of learning analytics representations. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 125\u2013134). New York: ACM. <\/span><\/span>\n\nWorsley, M., & Blikstein, P. (2013). Towards the development of multimodal action based assessment. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 94\u2013101). New York: ACM. <\/span><\/span>\n\nWorsley, M., & Blikstein, P. (2014a). Deciphering the practices and affordances of different reasoning strategies through multimodal learning analytics. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 21\u201327). New York: ACM. <\/span><\/span>\n\nWorsley, M., & Blikstein, P. (2014b). Using multimodal learning analytics to study learning mechanisms. In J. Stamper, Z. Pardos, M. Mavrikis, & B. M. McLaren (Eds.), Proceedings of the 7th International Conference on Educational Data Mining <\/i>(EDM2014), 4\u20137 July, London, UK (pp. 431\u2013432). International Educational Data Mining Society.<\/span><\/span>\n\nWorsley, M., & Blikstein, P. (2015a). Leveraging multimodal learning analytics to differentiate student learning strategies. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 360\u2013367). New York: ACM. <\/span><\/span>\n\nWorsley, M., & Blikstein, P. (2015b). Using learning analytics to study cognitive disequilibrium in a complex learning environment. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 426\u2013427). New York: ACM. <\/span><\/span>\n\nZhang, Z. (2012). Microsoft Kinect sensor and its effect. IEEE MultiMedia, 19<\/i>(2), 4\u201310. <\/span><\/span>\n\nZhou, J., Hang, K., Oviatt, S., Yu, K., & Chen, F. (2014). Combining empirical and machine learning techniques to predict math expertise using pen signal features. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 29\u201336). New York: ACM.<\/span><\/span>\n\n
\n\n
\n

1<\/a> orj. modalities<\/span><\/span><\/p>\n\n<\/div>\n

\n

2<\/a> orj.capture<\/span><\/span><\/p>\n\n<\/div>\n

\n

3<\/a> orj.artefact<\/span><\/span><\/p>\n\n<\/div>\n

\n

4<\/a> orj. trainer<\/span><\/span><\/p>\n\n<\/div>\n

\n

5<\/a> \u00c7evirenin notu: Yazar George Orwel\u2019in 1984 isimli kitab\u0131na at\u0131fta bulunuyor.<\/span><\/span><\/p>\n\n<\/div>\n","rendered":"

Xavier Ochoa<\/span><\/span><\/p>\n

Politecnica del Litoral Lisesi, Ekvator <\/span><\/span><\/p>\n

DOI: 10.18608\/hla17.011<\/span><\/span><\/p>\n

\u00d6Z<\/h2>\n

Bu b\u00f6l\u00fcm, \u00f6\u011frenme s\u00fcreci hakk\u0131nda daha sa\u011flam ve daha belirgin bir anlay\u0131\u015f elde etmek i\u00e7in tamamlay\u0131c\u0131 \u00f6\u011frenme izleri kaynaklar\u0131n\u0131n yakalanmas\u0131, birle\u015ftirilmesi ve analiz edilmesi yoluyla \u00f6\u011frenme analiti\u011fi (\u00d6A) yakla\u015f\u0131m y\u00f6ntemi sunmaktad\u0131r. \u00c7ok modlu \u00f6\u011frenme analitiklerinde (\u00c7M\u00d6A) kaynaklar veya y\u00f6ntemler1<\/a><\/sup>, \u00e7evrimi\u00e7i sistemler taraf\u0131ndan yakalanan geleneksel kay\u0131t g\u00fcnl\u00fc\u011f\u00fc verilerini i\u00e7erir, fakat ayn\u0131 zamanda yapay nesneleri ve hareketler, bak\u0131\u015f, konu\u015fma veya yazma gibi daha do\u011fal insan hareketlerini (sinyal) \u00f6\u011frenmeyi de i\u00e7erir. \u00c7M\u00d6A’n\u0131n mevcut durumu genellikle uyguland\u0131\u011f\u0131 \u00f6\u011frenme ortamlar\u0131na g\u00f6re tart\u0131\u015f\u0131l\u0131r ve y\u00f6ntemlerine g\u00f6re s\u0131n\u0131fland\u0131r\u0131l\u0131r. Bu b\u00f6l\u00fcm, \u00e7ok modlu tekniklerin uygulay\u0131c\u0131lar\u0131 i\u00e7in ortaya \u00e7\u0131kan sorunlar\u0131n tart\u0131\u015f\u0131lmas\u0131yla sona ermektedir.<\/span><\/span><\/p>\n

Anahtar Kelimeler<\/span>: Ses, video, veri birle\u015ftirme, \u00e7oklu alg\u0131lay\u0131c\u0131<\/span><\/span><\/p>\n

Esasen, \u00f6\u011frenme analiti\u011fi (\u00d6A) alan\u0131n\u0131n odak noktas\u0131, \u00f6\u011frencilerin bir \u00e7e\u015fit dijital ara\u00e7 kullan\u0131rken ger\u00e7ekle\u015ftirdikleri eylemlerin incelenmesiydi. Bu dijital ara\u00e7lar, \u00f6\u011frenme y\u00f6netimi sistemleri (\u00d6YS; Arnold ve Pistilli, 2012), bilgisayar destekli ak\u0131ll\u0131 \u00f6\u011fretim sistemleri (A\u00d6S’ler; Crossley, Roscoe ve McNamara, 2013), kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i dersler (KA\u00c7D’ler, K\u0131z\u0131lcec, Piech ve Schneider, 2013), e\u011fitici video oyunlar\u0131 (Serrano-Laguna ve Fernandez \u2013 Manjon, 2014) veya bir bilgisayar\u0131 \u00f6\u011frenme s\u00fcrecinde aktif bir bile\u015fen olarak kullanan di\u011fer sistemlerdir. \u00d6te yandan, bilgisayarlar\u0131n bulunmad\u0131\u011f\u0131 veya yaln\u0131zca yard\u0131mc\u0131, tan\u0131mlanmam\u0131\u015f bir role sahip oldu\u011fu y\u00fcz y\u00fcze dersler veya \u00e7al\u0131\u015fma gruplar\u0131 gibi di\u011fer \u00f6\u011frenme ba\u011flamlar\u0131nda nispeten daha az \u00d6A ara\u015ft\u0131rmas\u0131 veya uygulamas\u0131 yap\u0131lm\u0131\u015ft\u0131r. Bilgisayar destekli \u00f6\u011frenme ba\u011flamlar\u0131na y\u00f6nelik bu yanl\u0131l\u0131k, her t\u00fcrl\u00fc \u00d6A \u00e7al\u0131\u015fmas\u0131n\u0131n veya sisteminin temel ihtiyac\u0131 ile a\u00e7\u0131klanmaktad\u0131r: \u00f6\u011frenme izlerinin varl\u0131\u011f\u0131 (Siemens, 2013).<\/span><\/p>\n

Bilgisayar tabanl\u0131 \u00f6\u011frenme sistemleri, ba\u015flang\u0131\u00e7ta analitik d\u00fc\u015f\u00fcn\u00fclerek tasarlanmam\u0131\u015f olsa bile, kullan\u0131c\u0131lar\u0131 ile etkile\u015fimlerini otomatik olarak ince tanecikli ayr\u0131nt\u0131 \u015feklinde yakalama e\u011filimindedir. Bu etkile\u015fimleri tan\u0131mlayan veriler, analiz edilecek izlemleri \u00e7\u0131karmak i\u00e7in daha sonra tahmin edilebilecek g\u00fcnl\u00fck dosyalar\u0131 veya kelime i\u015flemci belgeleri gibi bir\u00e7ok formda saklan\u0131r. Kullan\u0131ma haz\u0131r verilerin g\u00f6reli bollu\u011fu ve i\u015flemenin \u00f6n\u00fcndeki teknik engeller, bilgisayar tabanl\u0131 \u00f6\u011frenme sistemlerini \u00d6A i\u00e7in AR-GE yapmada ideal bir yer haline getirir. Buna kar\u015f\u0131l\u0131k, bilgisayarlar\u0131n kullan\u0131lmad\u0131\u011f\u0131 \u00f6\u011frenme ba\u011flamlar\u0131nda, \u00f6\u011frenenlerin eylemleri otomatik olarak yakalanamaz2<\/a><\/sup>. \u00d6\u011frencinin \u00fcretti\u011fi fiziksel belgeler gibi baz\u0131 \u00f6\u011frenme \u00fcr\u00fcnleri3<\/a><\/sup> mevcut olsa bile, i\u015flenmeden \u00f6nce d\u00f6n\u00fc\u015ft\u00fcr\u00fclmeleri gerekir. Analiz edilecek izler olmadan, \u00d6A’da geleneksel olarak kullan\u0131lan bilgi i\u015flemsel modeller ve ara\u00e7lar ge\u00e7erli de\u011fildir.<\/span><\/p>\n

Bilgisayar destekli \u00f6\u011frenme ba\u011flamlar\u0131na y\u00f6nelik bu yanl\u0131l\u0131\u011f\u0131n varl\u0131\u011f\u0131, \u00d6A’da bir sokak lambas\u0131 etkisi (Freedman, 2010) olu\u015fturabilir. Bu etki ismini bir adam\u0131n ev anahtar\u0131n\u0131 parkta kaybetmesine ra\u011fmen onu sokak lambas\u0131n\u0131n alt\u0131nda aramas\u0131n\u0131 ifade eden bir \u015fakadan almaktad\u0131r. Sahneyi izleyen polis neden anahtar\u0131 sokak lambas\u0131n\u0131n alt\u0131nda arad\u0131\u011f\u0131n\u0131 sordu\u011funda adam “\u00e7\u00fcnk\u00fc buras\u0131 daha ayd\u0131nl\u0131k” diye cevap verir. Sokak lambas\u0131 efekti, \u00e7\u00f6z\u00fcmleri ger\u00e7ek \u00e7\u00f6z\u00fcmlerin olabilece\u011fi yerde de\u011fil araman\u0131n kolay oldu\u011fu yerde aramak anlam\u0131na gelmektedir. Bu durum s\u00fcrecin b\u00fcy\u00fck bir b\u00f6l\u00fcm\u00fcn\u00fcn ger\u00e7ekle\u015fti\u011fi ger\u00e7ek d\u00fcnya ortamlar\u0131n\u0131 g\u00f6z ard\u0131 edip, \u00f6\u011frenme s\u00fcrecini yaln\u0131zca bilgisayar temelli ba\u011flamlara bakarak anlamaya ve optimize etmeye \u00e7al\u0131\u015fan erken d\u00f6nem \u00d6A ara\u015ft\u0131rmalar\u0131 i\u00e7in d\u00fc\u015f\u00fcn\u00fclebilir. Hatta \u00f6\u011frenenlerin bilgisayar destekli sistemlerde kayd\u0131n\u0131n tutulamad\u0131\u011f\u0131 eylemleri bile genellikle g\u00f6z ard\u0131 edilir. \u00d6rne\u011fin, bir A\u00d6S’de bir problem sunuldu\u011funda kafas\u0131 kar\u0131\u015fan veya \u00e7evrimi\u00e7i bir ders izlerken s\u0131k\u0131l\u0131p esneyen bir \u00f6\u011frenen hakk\u0131ndaki bilgiler geleneksel \u00d6A ara\u015ft\u0131rmalar\u0131nda dikkate al\u0131nmaz. Sokak lambas\u0131 etkisini azaltmak i\u00e7in, ara\u015ft\u0131rmac\u0131lar \u015fimdi ger\u00e7ek d\u00fcnyadaki \u00f6\u011frenme ba\u011flamlar\u0131ndan ince taneli \u00f6\u011frenme izlerinin otomatik olarak toplanmas\u0131na odaklanarak, Bir KA\u00c7D oturumunun analizi kadar y\u00fcz y\u00fcze derslerin analizini de m\u00fcmk\u00fcn hale getiriyorlar. \u00d6A \u00fczerine daha yeni \u00e7al\u0131\u015fmalar, geleneksel g\u00fcnl\u00fck dosyalar\u0131ndan ba\u015fka yeni veri kaynaklar\u0131n\u0131 ara\u015ft\u0131r\u0131yor: \u00f6\u011frenen taraf\u0131ndan \u00fcretilen metinler (Simsek vd., 2015), g\u00f6z izleme bilgileri (Vatrapu, Reimann, Bull ve Johnson, 2013) ve s\u0131n\u0131f yap\u0131land\u0131rmas\u0131 (Almeda, Scupelli, Baker, Weber ve Fisher, 2014) bunlardan birka\u00e7\u0131d\u0131r. Bu farkl\u0131 \u00f6\u011frenme izlerinin kaynaklar\u0131n\u0131n tek bir analizde birle\u015ftirilmesi, \u00e7ok modlu \u00f6\u011frenme analiti\u011finin (\u00c7M\u00d6A) temel amac\u0131d\u0131r.<\/span><\/p>\n

\u00c7M\u00d6A, etkile\u015fimlerin sadece bir bilgisayar veya dijital cihaz arac\u0131l\u0131\u011f\u0131yla sa\u011flanmad\u0131\u011f\u0131 dijital ve ger\u00e7ek d\u00fcnya senaryolar\u0131ndaki \u00f6\u011frenmeyi anlamaya ve optimize etmeye odaklanarak farkl\u0131 \u00f6\u011frenme izleri kaynaklar\u0131n\u0131 \u00d6A ara\u015ft\u0131rmas\u0131na ve uygulamas\u0131na d\u00e2hil etmeye \u00e7al\u0131\u015fan bir alt aland\u0131r (Blikstein, 2013). \u00c7M\u00d6A’da, \u00f6\u011frenme izleri sadece g\u00fcnl\u00fck dosyalar\u0131ndan veya dijital belgelerden de\u011fil, kaydedilmi\u015f video ve seslerden, kalem vuru\u015flar\u0131ndan, konum izleme cihazlar\u0131ndan, biyo-alg\u0131lay\u0131c\u0131lardan ve \u00f6\u011frenme s\u00fcrecini anlamak veya \u00f6l\u00e7mek i\u00e7in yararl\u0131 olabilecek di\u011fer y\u00f6ntemlerden elde edilir. Ayr\u0131ca, \u00c7M\u00d6A’da, farkl\u0131 durum ve formlardan \u00e7\u0131kart\u0131lan izler, eylemlerin ve \u00f6\u011frenenin i\u00e7 durumunun daha kapsaml\u0131 bir g\u00f6r\u00fcn\u00fcm\u00fcn\u00fc sa\u011flamak i\u00e7in birle\u015ftirilir.<\/span><\/p>\n

\u00d6\u011frenmeyi ara\u015ft\u0131r\u0131rken farkl\u0131 y\u00f6ntem ve formlar\u0131 kullanma fikri, \u00d6A ba\u011flam\u0131nda yeni olmakla birlikte, geleneksel deneysel e\u011fitim ara\u015ft\u0131rmalar\u0131nda yayg\u0131nd\u0131r. Do\u011fas\u0131 gere\u011fi \u00e7ok modlu bir alg\u0131lay\u0131c\u0131 olan bir insan g\u00f6zlemcisini, ger\u00e7ek d\u00fcnyadaki \u00f6\u011frenme ba\u011flam\u0131na eklemek, \u00f6\u011frenmeyi do\u011fal ortamda \u00e7al\u0131\u015fman\u0131n ola\u011fan yoludur (Gall, Borg ve Gall, 1996). Video, ses kayd\u0131 ve etiketleme ara\u00e7lar\u0131 gibi teknolojiler bu g\u00f6zlemi daha az m\u00fcdahaleci ve daha \u00f6l\u00e7\u00fclebilir hale getirmi\u015ftir (Cobb vd., 2003; Lund, 2007). Geleneksel e\u011fitsel ara\u015ft\u0131rma yakla\u015f\u0131m\u0131n\u0131n temel sorunu, veri toplama ve analizlerinin, el ile yap\u0131lmalar\u0131 nedeniyle \u00e7ok maliyetli olmalar\u0131 ve \u00f6l\u00e7eklenmemeleridir. Veri toplaman\u0131n hem boyut hem de zaman a\u00e7\u0131s\u0131ndan s\u0131n\u0131rl\u0131 olmas\u0131 gerekir ve veri analizi sonu\u00e7lar\u0131, \u00e7al\u0131\u015f\u0131lan \u00f6\u011frenenler i\u00e7in faydal\u0131 olacak kadar h\u0131zl\u0131 ve kullan\u0131\u015fl\u0131 de\u011fildir. Farkl\u0131 modalite ve formlar kaydedilebilir ve \u00f6\u011frenme izleri bunlardan otomatik olarak \u00e7\u0131kar\u0131labilirse, \u00d6A ara\u00e7lar\u0131, \u00f6\u011frenmeyi oldu\u011fu gibi iyile\u015ftirmek ve s\u00fcrekli bir ger\u00e7ek zamanl\u0131 geribildirim d\u00f6ng\u00fcs\u00fc sa\u011flamak i\u00e7in kullan\u0131labilir.<\/span><\/p>\n

Beklenildi\u011fi gibi, ham \u00e7ok modlu kay\u0131tlardan \u00f6\u011frenme izlerini \u00e7\u0131karmak s\u0131radan bir i\u015f de\u011fildir. Bilgisayar g\u00f6r\u00fc\u015f\u00fc, konu\u015fma i\u015fleme, eskiz tan\u0131ma ve bilgisayar bilimleri alanlar\u0131nda geli\u015ftirilen di\u011fer teknikler, \u00f6\u011frenme bilimi, e\u011fitim ara\u015ft\u0131rmas\u0131 ve davran\u0131\u015f bilimi taraf\u0131ndan sa\u011flanan mevcut \u00f6\u011frenme teorileri taraf\u0131ndan y\u00f6nlendirilmelidir. Karma\u015f\u0131kl\u0131\u011f\u0131 g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, \u00c7M\u00d6A alt alan\u0131 nispeten gen\u00e7 ve ke\u015ffedilmemi\u015ftir. Ancak ilk \u00e7al\u0131\u015fmalar ve ara\u015ft\u0131rmac\u0131lar aras\u0131ndaki erken disiplinler aras\u0131 i\u015f birli\u011fi olumlu sonu\u00e7lar vermi\u015ftir (Scherer, Worsley ve Morency, 2012; Morency, Oviatt, Scherer, Weibel ve Worsley, 2013; Ochoa, Worsley, Chiluiza ve Luz, 2014, Markaki, Lund ve Sanchez, 2015). Bu b\u00f6l\u00fcm, bu alan\u0131 ara\u015ft\u0131rmak isteyen ara\u015ft\u0131rmac\u0131lar ve uygulay\u0131c\u0131lar i\u00e7in bir k\u0131lavuzdur. \u0130lk olarak, \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131nda kullan\u0131lan ana modaliteler sunulacak, analiz edilecek ve \u00f6rneklendirilecektir. \u0130kincisi, \u00c7M\u00d6A’n\u0131n uyguland\u0131\u011f\u0131 ger\u00e7ek d\u00fcnya ortamlar\u0131 ana durum ve modalitelerine g\u00f6re incelenecek ve s\u0131n\u0131fland\u0131r\u0131lacakt\u0131r. Son olarak, \u00c7M\u00d6A ara\u015ft\u0131rmas\u0131 ve uygulamas\u0131 i\u00e7in \u00f6nemli olan \u00e7\u00f6z\u00fclmemi\u015f birka\u00e7 konu tart\u0131\u015f\u0131lacakt\u0131r.<\/span><\/p>\n

MODAL\u0130TELER VE MEDYA<\/h2>\n

\u0130leti\u015fim kuram\u0131 tan\u0131m\u0131nda, \u00e7ok modluluk, bireyler aras\u0131nda bilgi ve anlam al\u0131\u015fveri\u015finde bulunmak i\u00e7in \u00e7e\u015fitli ileti\u015fim bi\u00e7imlerinin (metinsel, i\u015fitsel, dilbilimsel, mek\u00e2nsal, g\u00f6rsel, vb.) kullan\u0131lmas\u0131n\u0131 ifade eder (Kress ve Van Leeuwen, 2001). Medya filmleri, kitaplar, web sayfalar\u0131 ve hatta hava, ileti\u015fim modunun kodlanabilece\u011fi fiziksel veya dijital birer alt tabakad\u0131r. Her mod bir veya birka\u00e7 medya arac\u0131l\u0131\u011f\u0131yla ifade edilebilir. \u00d6rne\u011fin, konu\u015fma, havadaki bas\u0131n\u00e7 de\u011fi\u015fimleri (y\u00fcz y\u00fcze diyalogda), kasetteki manyetik y\u00f6n de\u011fi\u015fimleri (kaset kayd\u0131nda) veya dijital say\u0131lar\u0131n de\u011fi\u015fimleri (MP3 dosyas\u0131na) kodlanabilir. Ayr\u0131ca, ayn\u0131 ara\u00e7 birka\u00e7 modu iletmek i\u00e7in kullan\u0131labilir. \u00d6rne\u011fin, bir video kayd\u0131 v\u00fccut dili (duru\u015f), duygular (y\u00fcz ifadesi) ve kullan\u0131lan ara\u00e7lar (eylemler) hakk\u0131nda bilgi i\u00e7erebilir.<\/span><\/p>\n

Kendi do\u011fas\u0131 gere\u011fi, \u00f6\u011frenme genellikle \u00e7ok modludur (Jewitt, 2006). Bir insan, bir kitap okuyarak, bir profes\u00f6r\u00fc dinleyerek, bir i\u015flemi izleyerek, fiziksel veya dijital ara\u00e7lar kullanarak ve g\u00f6rece karma\u015f\u0131k bilgilerin kodlanabilece\u011fi herhangi bir ba\u015fka insan ileti\u015fim moduyla \u00f6\u011frenebilir. \u00d6\u011frenme s\u00fcreci ayn\u0131 zamanda birka\u00e7 geri bildirim d\u00f6ng\u00fcs\u00fc de kullan\u0131r; \u00f6r. \u00f6\u011freten dersin anla\u015f\u0131l\u0131p anla\u015f\u0131lmad\u0131\u011f\u0131n\u0131 sordu\u011funda ba\u015f\u0131n\u0131 sallayan bir \u00f6\u011freneni ya da \u00f6\u011fretenin sesinin bir konuyu a\u00e7\u0131klarken kulland\u0131\u011f\u0131 vurgulama gibi. Bu geri bildirim modlar\u0131 genellikle daha basit fakat s\u00fcre\u00e7 i\u00e7in \u00f6nemli olan bilgileri kodlar. \u00d6\u011frenme analiz edilecek, anla\u015f\u0131lacak ve optimize edilecekse, ilgili modlar\u0131n her birinde meydana gelen etkile\u015fimlerin izleri elde edilmelidir. \u00c7M\u00d6A, bu izlerin kodland\u0131\u011f\u0131 veya kaydedildi\u011fi ortamdan ba\u011f\u0131ms\u0131z olarak, bu izleri farkl\u0131 ileti\u015fim modlar\u0131ndan \u00e7\u0131karmaya odaklan\u0131r.<\/span><\/p>\n

A\u015fa\u011f\u0131daki alt b\u00f6l\u00fcmler, \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131nda kullan\u0131lan en yayg\u0131n y\u00f6ntemlerin yans\u0131t\u0131lmas\u0131 ve izlerinin \u00e7\u0131kar\u0131lmas\u0131 konusundaki mevcut durumu sunmaktad\u0131r. Her modalite i\u00e7in, \u00f6\u011frenme s\u00fcrecini anlama konusundaki \u00f6nemi, en yayg\u0131n yakalama (capture etme) ve kaydetme y\u00f6ntemlerinin listesi ve kullan\u0131ld\u0131\u011f\u0131 yerlerin \u00f6rnekleri ile birlikte k\u0131sa bir tan\u0131m sunulmu\u015ftur. Bu \u00f6\u011frenmeyle ilgili t\u00fcm modlar\u0131n kapsaml\u0131 bir listesi de\u011fil sadece \u00c7M\u00d6A \u00e7al\u0131\u015fmalar\u0131nda kullan\u0131lanlard\u0131r.<\/span><\/p>\n

Bak\u0131\u015f<\/h3>\n

\u0130nsanlar do\u011frudan dikkatlerini \u00e7eken \u015feye bakma e\u011filimindedir. Dolay\u0131s\u0131yla bir bireyin g\u00f6r\u00fc\u015f\u00fcn\u00fcn y\u00f6n\u00fc, dikkatinin y\u00f6n\u00fcn\u00fcn bir g\u00f6stergesidir (Frischen, Bayliss ve Tipper, 2007). Dikkat, \u00f6\u011frenme i\u00e7in vazge\u00e7ilmez bir gerekliliktir (Kruschke, 2003). Bir i\u015farete dikkat etmek, bireyin bilgilerini elde etmesine ve ilgili par\u00e7alar\u0131 uzun s\u00fcreli haf\u0131zada saklamas\u0131na yard\u0131mc\u0131 olur. Bak\u0131\u015f, dikkatleri tahmin edebilen tek temsilci ve hatas\u0131z olmasa da e\u011fitim uygulamalar\u0131nda yayg\u0131n olarak kullan\u0131l\u0131r. \u00d6rne\u011fin, bu konuda e\u011fitim alm\u0131\u015f bir \u00f6\u011freten, \u00f6\u011frencilerin bak\u0131\u015flar\u0131n\u0131 g\u00f6zlemleyerek b\u00fct\u00fcn bir s\u0131n\u0131f\u0131n dikkat seviyesini de\u011ferlendirebilir; bir g\u00f6zlemci, bak\u0131\u015f\u0131n konu\u015fmac\u0131dan konu\u015fmac\u0131ya do\u011fru y\u00f6n\u00fcn\u00fc izleyerek bir tart\u0131\u015fmadaki kat\u0131l\u0131mc\u0131n\u0131n dikkat seviyesini belirleyebilir.<\/span><\/p>\n

Bak\u0131\u015flar\u0131n \u00f6nemi uzun zamand\u0131r pazarlamac\u0131lar, davran\u0131\u015f bilimcileri ve insan-bilgisayar etkile\u015fimi ara\u015ft\u0131rmac\u0131lar\u0131 taraf\u0131ndan \u00e7al\u0131\u015f\u0131lm\u0131\u015ft\u0131r. Reklam\u0131n etkinli\u011fini (Krugman, Fox, Fletcher, Fischer ve Rojas, 1994) belirlemek, otizmin erken te\u015fhisinde (Boraston ve Blakemore, 2007) ve bilgisayar aray\u00fczlerinin etkinli\u011fi konusunda (Poole ve Ball, 2006) yard\u0131mc\u0131 olmak i\u00e7in g\u00f6z izleme \u00e7al\u0131\u015fmalar\u0131 yayg\u0131nd\u0131r. Bununla birlikte, bu \u00e7al\u0131\u015fmalarda monit\u00f6re sabitlenmi\u015f g\u00f6z izleyicileri veya \u00f6zel g\u00f6z izleme g\u00f6zl\u00fckleri kullan\u0131larak bak\u0131\u015flar\u0131 kaydetmenin ana y\u00f6ntemleri \u00f6\u011frenme ortamlar\u0131nda yayg\u0131n olarak kullan\u0131lamayacak kadar elveri\u015fsiz ve maliyetlidir. \u00c7M\u00d6A\u2019da bak\u0131\u015f\u0131 yakalaman\u0131n mevcut se\u00e7ene\u011fi, video kay\u0131tlar\u0131d\u0131r (Raca ve Dillenbourg, 2013).<\/span><\/p>\n

\"\"<\/p>\n

<\/a> \u015eekil 11.1. Bir s\u0131n\u0131f ortam\u0131nda bak\u0131\u015f kestirimi (Raca, Tormey ve Dillenbourg, 2014).<\/i><\/span><\/span><\/p>\n

Bir veya bir dizi kamera, deneklerin ba\u015flar\u0131n\u0131 ve g\u00f6zlerini kaydetmek i\u00e7in konumland\u0131r\u0131lm\u0131\u015ft\u0131r. Sonra, bilgisayar g\u00f6rme teknikleri Lin, Lin, Lin ve Lee (2013) te sunuldu\u011fu gibi, video kayd\u0131ndan bak\u0131\u015f y\u00f6n\u00fc bilgisini \u00e7\u0131karmak i\u00e7in kullan\u0131l\u0131r. Kay\u0131ttaki g\u00f6receli bak\u0131\u015f y\u00f6n\u00fcn\u00fc elde etmek i\u00e7in kontrol edilmesi gereken ana hususlar, y\u00fcz \u00e7\u00f6z\u00fcn\u00fcrl\u00fc\u011f\u00fc ve ortamdaki nesneler veya di\u011fer ki\u015filer taraf\u0131ndan kapanmay\u0131 \u00f6nlemektir (Raca ve Dillenbourg, 2013). Mutlak bak\u0131\u015f y\u00f6n\u00fcn\u00fc hesaplamak i\u00e7in kameralar\u0131n \u00f6\u011frenme ayarlar\u0131ndaki konumu ile ilgili bilgiler de kaydedilmelidir.<\/span><\/p>\n

\u00c7M\u00d6A’da, bir\u00e7ok bak\u0131\u015f izi \u00e7\u0131kar\u0131m \u00f6rne\u011fi vard\u0131r. Raca ve Dillenbourg (2013), b\u00f6l\u00fcme dayal\u0131 bir model kullanarak bir derste oturan \u00f6\u011frencilerin video kay\u0131tlar\u0131nda ba\u015f y\u00f6neliminden bak\u0131\u015f y\u00f6n\u00fcn\u00fc tahmin etmektedir (\u015eekil 11.1). Bu \u015fekilde, \u00f6\u011frenen y\u00fczleri otomatik olarak tan\u0131n\u0131r (dikd\u00f6rtgen) ve bak\u0131\u015flar\u0131 (ok) bir insan y\u00fcz\u00fc modeline dayanarak tahmin edilir. Bu bilgiler daha sonra \u00f6\u011frencilerin bireysel olarak dikkatinin oda\u011f\u0131n\u0131 belirlemek ve bunlar\u0131 kendisi taraf\u0131ndan bildirilen dikkat ile kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in kullan\u0131l\u0131r. Raca ve Dillenbourg, \u00f6\u011frencilerin \u00f6\u011freteni g\u00f6r\u00fc\u015f alanlar\u0131nda tuttuklar\u0131 zaman y\u00fczdesinin dikkat d\u00fczeyinin \u00f6nemli bir belirleyicisi oldu\u011funu bulmu\u015flard\u0131r. Farkl\u0131 bir \u00f6\u011frenme ortam\u0131nda, Echeverria, Avendano, Chiluiza, Vasquez ve Ochoa (2014), g\u00f6z merkezi noktalar\u0131 ile burun ucu noktas\u0131na olan mesafeyi hesaplayarak kafa y\u00f6n\u00fcn\u00fc \u00f6l\u00e7en bak\u0131\u015f y\u00f6n\u00fcn\u00fc de tahmin etmi\u015flerdir. Bu bilgi, \u00f6\u011frencilerin akademik sunumlar s\u0131ras\u0131nda izleyiciyle g\u00f6z temas\u0131n\u0131 sa\u011flay\u0131p sa\u011flamad\u0131klar\u0131n\u0131 belirlemek i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r.<\/span><\/p>\n

Duru\u015f, Jestler ve Hareket (Beden Dili)<\/h3>\n

Duru\u015f, jestler ve hareket, her biri farkl\u0131 bilgi t\u00fcrlerini ta\u015f\u0131yabilmesine ra\u011fmen, birlikte beden dili olarak adland\u0131r\u0131lan birbiriyle ili\u015fkili \u00fc\u00e7 moddur (Bull, 2013). Duru\u015f, v\u00fccudun veya v\u00fccudun bir k\u0131sm\u0131n\u0131n belirli bir zamanda ald\u0131\u011f\u0131 pozisyonu ifade eder. Bir \u00f6\u011frenenin duru\u015fu i\u00e7sel durumu hakk\u0131nda bilgi verebilir. \u00d6rne\u011fin, bir \u00f6\u011frenen otururken ba\u015f\u0131n\u0131 masan\u0131n \u00fcst\u00fcne koymu\u015fsa, \u00f6\u011freten, bu durumdan \u00f6\u011frenenin yorgun oldu\u011funu ya da derse kar\u015f\u0131 ilgisinin olmad\u0131\u011f\u0131n\u0131 \u00e7\u0131karabilir. Baz\u0131 \u00f6zel durumlarda, edinilen duru\u015f becerilerin kazan\u0131lmas\u0131 ile ilgilidir. \u00d6rne\u011fin, s\u00f6zl\u00fc sunumlara ili\u015fkin e\u011fitilen \u00f6\u011frencilerin, belirli duru\u015flar\u0131 (eller ve kollar biraz a\u00e7\u0131k) di\u011ferlerinden (eller ceplerinde) daha fazla kullanmalar\u0131 beklenir. \u00d6\u011frenilen jestler i\u00e7sel bir durumu g\u00f6stermez. Jestler, v\u00fccudun farkl\u0131 b\u00f6lgelerinin, \u00f6zellikle ba\u015f, kollar ve ellerin belirli bir anlam\u0131 iletmek i\u00e7in koordine edilmi\u015f hareketleridir. S\u00f6zel olmayan bu ileti\u015fim \u015fekli genellikle bilin\u00e7lidir. \u00d6\u011frenme s\u00fcrecinde k\u0131sa geri bildirim d\u00f6ng\u00fcleri ve alternatif vurgulama kanallar\u0131 sa\u011flaman\u0131n bir yolu olarak kullan\u0131l\u0131r. \u00d6rne\u011fin, tahtadaki belirli bir b\u00f6l\u00fcme i\u015faret eden \u00f6\u011fretmen veya zor bir soru ile kar\u015f\u0131la\u015ft\u0131\u011f\u0131nda omuzlar\u0131n\u0131 kald\u0131ran bir \u00f6\u011frenci. Son olarak, hareket, yeni bir duru\u015f edinmek veya belirli bir jesti icra etme gereklili\u011fi olmadan beden pozisyonundaki herhangi bir de\u011fi\u015fikli\u011fi ifade eder. Bu hareket genellikle \u00f6\u011frenme s\u00fcrecinde bireyin i\u00e7sel durumunu ortaya \u00e7\u0131karan bilin\u00e7siz v\u00fccut hareketlerinin bir sonucudur; \u00f6rne\u011fin, gerginlik veya \u015f\u00fcphe g\u00f6steren de\u011fi\u015fken hareketler.<\/span><\/p>\n

Duru\u015f, jestler ve hareketler, ger\u00e7ek d\u00fcnya ortamlar\u0131nda video yakalamadaki kolayl\u0131k, d\u00fc\u015f\u00fck maliyetli 2-D ve 3-D alg\u0131lay\u0131c\u0131lar\u0131n ve y\u00fcksek performansl\u0131 bilgisayar g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fcn kullan\u0131labilirli\u011fi ile \u00f6zellik \u00e7\u0131kar\u0131m\u0131 i\u00e7in y\u00fcksek performansl\u0131 bilgisayar g\u00f6r\u00fc\u015f\u00fc algoritmalar nedeniyle \u00c7M\u00d6A’da en s\u0131k \u00e7al\u0131\u015f\u0131lan modlar olmu\u015ftur. V\u00fccut dili, farkl\u0131 v\u00fccut par\u00e7alar\u0131na ba\u011fl\u0131 ivme\u00f6l\u00e7er (Mitra ve Acharya, 2007) veya \u00f6zel ara\u00e7lar (\u00f6r. bir Wii Remote; Schlomer, Poppinga, Henze ve Boll, 2008) kullan\u0131larak y\u00fcksek hassasiyetler yakalan\u0131rken, pratikte bunlar\u0131n \u00e7o\u011fu \u00f6\u011frenme etkinli\u011fi i\u00e7inde kullan\u0131m\u0131 mevcut sisteme \u00e7ok yabanc\u0131d\u0131r. Hareketi yakalamak i\u00e7in en yayg\u0131n \u00e7\u00f6z\u00fcm konunun videosunu kaydetmek ile duru\u015f, jestler ve hareketi tahmin etmektir. Herhangi bir kamera t\u00fcr\u00fc, ilgili hareketi yeterli \u00e7\u00f6z\u00fcn\u00fcrl\u00fckte yakalayabildi\u011fi s\u00fcrece kullan\u0131labilir. Gereken \u00e7\u00f6z\u00fcn\u00fcrl\u00fck, video ile y\u00fcr\u00fct\u00fclen \u00f6zellik \u00e7\u0131karma t\u00fcr\u00fcne ba\u011fl\u0131d\u0131r. \u0130nsan hareketinin otomatik olarak \u00e7\u0131kart\u0131lmas\u0131 i\u00e7in kullan\u0131lan en yayg\u0131n cihaz Microsoft Kinect’tir (Zhang, 2012). Kinect, video ve derinlik yakalama kar\u0131\u015f\u0131m\u0131 sayesinde ara\u015ft\u0131rmac\u0131lara, \u00e7ekilen her foto\u011fraf karesi i\u00e7in konunun yeniden yap\u0131land\u0131r\u0131lm\u0131\u015f bir iskeletini sunabilir. Bu da v\u00fccut duru\u015flar\u0131n\u0131 ve hareketlerini yakalamak i\u00e7in idealdir. Kinect alg\u0131lay\u0131c\u0131s\u0131n\u0131n yeni s\u00fcr\u00fcmleri ayn\u0131 zamanda el hareketini de \u00e7\u0131kartabilmektedir (Vasquez, Vargas ve Sucar, 2015).<\/span><\/p>\n

\"\"<\/p>\n

<\/a> \u015eekil 11.2. Ger\u00e7ek \u00f6\u011frenen sunucular\u0131n k\u00fcmelenmi\u015f \u00fcst v\u00fccut duru\u015flar\u0131 (Echeverr\u00eda, Avenda\u00f1o, Chiluiza,V\u00e1squez ve Ochoa, 2014).<\/i><\/span><\/span><\/p>\n

\u00c7M\u00d6A’da beden dilinin yakalanmas\u0131na ve i\u015flenmesine ili\u015fkin en belirgin \u00f6rnekler, bir s\u0131n\u0131f ortam\u0131nda \u00fcst beden hareket gecikmesi arac\u0131l\u0131\u011f\u0131yla dikkatin tahmin edilmesi (Raca, Tormey ve Dillenbourg, 2014) ve otomatikle\u015ftirilmi\u015f bir sunum \u00f6\u011freticisi olu\u015fturulmas\u0131na y\u00f6nelik gen\u00e7 bir akademisyen sunucunun duru\u015f ve jest analizidir (Echeverria vd., 2014). \u015eekil 11.2, \u00e7al\u0131\u015fmalar\u0131n\u0131 sunan \u00f6\u011frencilerin Kinect verilerinin analizinden elde edilen 23 farkl\u0131 durumu g\u00f6stermektedir. Bu 23 duru\u015f, bir sunum i\u00e7in iyi veya k\u00f6t\u00fc olarak kabul edilebilecek alt\u0131 v\u00fccut hareketi (farkl\u0131 renkler) olarak s\u0131n\u0131fland\u0131r\u0131lm\u0131\u015ft\u0131r. \u015eekil 11.3 ger\u00e7ek sunumlar s\u0131ras\u0131nda bu v\u00fccut hareketlerinin ger\u00e7ek \u00f6rneklerini sunar. Pozun s\u0131n\u0131fland\u0131r\u0131lmas\u0131 (soldaki Kinect noktalar\u0131n\u0131n \u00fcst\u00fcnde), bir insan g\u00f6zlemcinin foto\u011fraftan yorumlayabildi\u011fi ile ayn\u0131d\u0131r (a\u015fa\u011f\u0131da).<\/span><\/p>\n

\"\"<\/p>\n

<\/a> \u015eekil 11.3. Prototip duru\u015flara g\u00f6re s\u0131n\u0131fland\u0131r\u0131lm\u0131\u015f ger\u00e7ek duru\u015flar (Echeverria vd., 2014).<\/i><\/span><\/span><\/p>\n

Jestleri kullanman\u0131n di\u011fer ilgin\u00e7 \u00f6rnekleri Boncoddo vd. (2013), Alibali, Nathan, Fujimori, Stein ve Raudenbush, (2011) ve Mazur-Palandre, Colletta ve Lund (2014)’d\u0131r. \u0130lk olarak, Boncoddo vd. (2013) matematiksel kan\u0131tlar\u0131n a\u00e7\u0131klanmas\u0131 s\u0131ras\u0131nda ger\u00e7ekle\u015ftirilen ilgili jestlerin say\u0131s\u0131n\u0131 yakalam\u0131\u015f ve \u00f6\u011frencilerin cevaplar\u0131na ula\u015fma \u015fekilleriyle ili\u015fki kurmu\u015ftur. \u0130kincisi, Alibali vd. (2011) \u00f6\u011fretmenlerin matematik derslerinde yapt\u0131klar\u0131 farkl\u0131 jestleri s\u0131n\u0131fland\u0131rm\u0131\u015f ve aralar\u0131ndaki ili\u015fkileri bulmu\u015flard\u0131r. Son olarak, Mazur-Palandre vd. (2014), s\u00fcre\u00e7 ve talimatlar\u0131 a\u00e7\u0131klarken \u00e7ocuklar\u0131n jestleri kullan\u0131m\u0131 \u00fczerine bir \u00e7al\u0131\u015fma sunmu\u015ftur.<\/span><\/p>\n

Eylemler<\/h3>\n

Eylem modu, jest ve hareket modlar\u0131na \u00e7ok benzer. Her ikisi de genellikle \u00c7M\u00d6A\u2019daki video kay\u0131tlar\u0131 taraf\u0131ndan yakalanan v\u00fccut hareketleridir. Bununla birlikte, eylemler genellikle \u00f6\u011frenilen bir arac\u0131n manip\u00fclasyonunu i\u00e7eren ama\u00e7l\u0131 hareketlerdir. Bu eylemlerin t\u00fcr\u00fc, dizilim veya do\u011frulu\u011fu, \u00f6\u011frenenin belirli bir beceride elde etti\u011fi ustal\u0131k seviyesinin bir g\u00f6stergesi olarak kullan\u0131labilir. \u00d6rne\u011fin, bir \u00f6\u011frenenin laboratuvardaki \u00e7e\u015fitli ara\u00e7lar\u0131 manip\u00fcle edi\u015findeki d\u00fczen ve g\u00fcvenlik, \u00f6\u011frenenin belirli bir s\u00fcre\u00e7 hakk\u0131ndaki anlay\u0131\u015f\u0131n\u0131 belirlemek i\u00e7in bir sembol olarak kullan\u0131labilir.<\/span><\/p>\n

\u00c7M\u00d6A’da eylem kay\u0131t ve analizinin temel kullan\u0131mlar\u0131 uzmanl\u0131k kestirimindedir. Bir m\u00fchendislik in\u015fas\u0131 faaliyetinde, \u00f6rne\u011fin, el ve bilek hareketinin analizi uzmanl\u0131k seviyesini belirleyebilir (Worsley ve Blikstein, 2014b). Matematiksel problem \u00e7\u00f6zmede, \u00f6\u011frenenin hesap makinesini kulland\u0131\u011f\u0131 zaman y\u00fczdesi \u00f6l\u00e7\u00fclebilir (Ochoa vd., 2013). Ochoa vd. (2013) hesap makinesinin problem \u00e7\u00f6zme oturumlar\u0131ndaki pozisyonunu ve a\u00e7\u0131s\u0131n\u0131 izlemi\u015ftir (\u015eekil 11.4). Bu konum ve a\u00e7\u0131 (do\u011fru) daha sonra videodaki o belirli \u00e7er\u00e7eve s\u0131ras\u0131nda (g\u00f6r\u00fcnt\u00fcn\u00fcn kenarl\u0131\u011f\u0131 ile kesi\u015fme) hesap makinesini hangi \u00f6\u011frenenin kulland\u0131\u011f\u0131n\u0131 tahmin etmek i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r.<\/span><\/p>\n

\"\"<\/p>\n

<\/a> \u015eekil 11.4. Uzmanl\u0131k tahmini i\u00e7in hesap makinesi kullan\u0131m\u0131n\u0131n belirlenmesi (Ochoa vd., 2013).<\/i><\/span><\/span><\/p>\n

Y\u00fcz \u0130fadeleri<\/h3>\n

Y\u00fcz ifadeleriyle toplanan bilgiler v\u00fccut dili modlar\u0131 ile de b\u00fcy\u00fck oranda ili\u015fkilidir. \u0130nsan y\u00fcz\u00fc \u00e7ok karma\u015f\u0131k zihinsel durumlar\u0131 nispeten basit ifadelerle iletebilir. Bilgisayarla g\u00f6rme alan\u0131nda, videoda kaydedilen y\u00fcz ifadelerinden duygular\u0131 otomatik olarak tan\u0131mlamaya \u00e7al\u0131\u015fan \u00e7ok say\u0131da ba\u015far\u0131l\u0131 ara\u015ft\u0131rma yap\u0131lm\u0131\u015ft\u0131r (Mishra vd., 2015).<\/span><\/p>\n

\u00d6A alan\u0131nda y\u00fcz ifadeleri kullanman\u0131n ana \u00f6rnekleri, Craig, D’Mello, Witherspoon ve Graesser (2008) ve Worsley ve Blikstein (2015b) ‘nin eserleridir. Craig vd. (2008), AutoTutor sistemini kullan\u0131rken \u00f6\u011frencilerin duygusal durumlar\u0131n\u0131 otomatik olarak tahmin etmi\u015ftir (Graesser, Chipman, Haynes ve Olney, 2005). Worsley ve Blikstein (2015b), \u00f6\u011frenciler farkl\u0131 in\u015fa al\u0131\u015ft\u0131rmalar\u0131 ile kar\u015f\u0131 kar\u015f\u0131ya kald\u0131klar\u0131nda duygusal de\u011fi\u015fiklikleri ke\u015ffetmek i\u00e7in benzer teknikleri kulland\u0131lar. Her iki \u00e7al\u0131\u015fmada da kar\u0131\u015f\u0131k bir ifadenin \u00f6\u011frenme s\u00fcrecinin ba\u015far\u0131s\u0131n\u0131n iyi bir g\u00f6stergesi oldu\u011funu ke\u015ffetmi\u015ftir.<\/span><\/p>\n

Konu\u015fma<\/h3>\n

\u00c7M\u00d6A’da ses kay\u0131tlar\u0131n\u0131n en yayg\u0131n kullan\u0131m\u0131, \u00f6\u011frenenin ne hakk\u0131nda konu\u015ftu\u011funu veya dinledi\u011fini izlemektir. \u0130nsanlar aras\u0131ndaki en temel ve en karma\u015f\u0131k ileti\u015fim \u015fekli olan konu\u015fma, \u00f6\u011frenme s\u00fcrecini anlamada \u00f6zellikle \u00f6nemlidir. Mevcut \u00c7M\u00d6A uygulamas\u0131nda, ses kay\u0131tlar\u0131ndan iki ana sinyal \u00e7\u0131kar\u0131l\u0131r: ne s\u00f6yleniyor ve nas\u0131l s\u00f6yleniyor. \u0130lk yakla\u015f\u0131mda, genellikle konu\u015fmay\u0131 tan\u0131ma ad\u0131 verilen, konu\u015fman\u0131n as\u0131l i\u00e7eri\u011fi \u00e7\u0131kar\u0131l\u0131r. Bu analizin sonucu, konunun neden bahsetti\u011fini belirlemek i\u00e7in do\u011fal dil i\u015fleme (DD\u0130) ara\u00e7lar\u0131 kullan\u0131larak i\u015flenebilecek bir transkripttir. \u0130kinci yakla\u015f\u0131mda, konu\u015fman\u0131n tonlama, vurgu ve ritim gibi prosodik \u00f6zellikleri \u00e7\u0131kar\u0131l\u0131r. Bu \u00f6zellikler i\u00e7 duruma (g\u00fcvenlik, duygusal durum, vb.) veya konu\u015fmac\u0131n\u0131n niyetine (\u015faka, alayc\u0131l\u0131k, vb.) I\u015f\u0131k tutabilir. Konu\u015fma tan\u0131ma, kullan\u0131lan dile b\u00fcy\u00fck \u00f6l\u00e7\u00fcde ba\u011fl\u0131d\u0131r; prosodik \u00f6zellikler dil farkl\u0131l\u0131klar\u0131na kar\u015f\u0131 daha az hassast\u0131r.<\/span><\/p>\n

Ses mikrofonlar arac\u0131l\u0131\u011f\u0131yla yakalan\u0131r. Videodan daha kolay yakalanmakla birlikte, i\u015flenecek kadar y\u00fcksek kalitede ses kayd\u0131 yapmak asl\u0131nda \u00e7ok daha karma\u015f\u0131kt\u0131r. Mikrofonlar\u0131n tipi ve mek\u00e2nsal konfig\u00fcrasyonu, \u00f6\u011frenme ortam\u0131na ve kaydedilen sinyal ile ne t\u00fcr bir analizin yap\u0131laca\u011f\u0131na ba\u011fl\u0131d\u0131r. \u00d6rne\u011fin, otomatik konu\u015fma tan\u0131ma giri\u015fiminde bulunulursa, mikrofon y\u00f6nl\u00fc olmal\u0131 ve dene\u011fin a\u011fz\u0131na yak\u0131n olmal\u0131d\u0131r. \u00d6te yandan, yaln\u0131zca birinin ne zaman konu\u015ftu\u011funun tespiti gerekliyse, odan\u0131n ortas\u0131nda bulunan bir \u00e7evresel mikrofon yeterli olabilir. G\u00fcr\u00fclt\u00fc ve \u00e7oklu sinyallerin varl\u0131\u011f\u0131 sadece otomatik \u00f6zellik \u00e7\u0131kar\u0131m\u0131n\u0131 engellemekle kalmaz, ayn\u0131 zamanda manuel ek a\u00e7\u0131klamalar\u0131 da bozar. Bireysel yak\u0131n kay\u0131t m\u00fcmk\u00fcn olmad\u0131\u011f\u0131nda kay\u0131tlar\u0131 iyile\u015ftirmek i\u00e7in kullan\u0131lan en yayg\u0131n teknik, yaln\u0131zca g\u00fcr\u00fclt\u00fcy\u00fc azaltmakla kalmayacak, ayn\u0131 zamanda sesin mek\u00e2nsal k\u00f6kenini de belirleyebilecek mikrofon dizilerinin kullan\u0131lmas\u0131d\u0131r.<\/span><\/p>\n

\u00d6neminden dolay\u0131, ses bug\u00fcne kadar \u00e7o\u011fu \u00c7M\u00d6A \u00e7al\u0131\u015fmalar\u0131nda da bulunmaktad\u0131r. \u0130\u015fbirlikli \u00f6\u011frenme diyaloglar\u0131n\u0131n benzerlik seviyesini belirlemek i\u00e7in (Luzardo, Guaman, Chiluiza, Castells ve Ochoa, 2014), s\u00f6zl\u00fc sunumlar\u0131n kalitesini de\u011ferlendirmek (Luzardo, Guaman, Chiluiza, Castells ve Ochoa, 2014)ve matematik problem \u00e7\u00f6zme uzmanl\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in farkl\u0131 konu\u015fma analizi t\u00fcrleri kullan\u0131lm\u0131\u015ft\u0131r (Thompson, 2013).<\/span><\/p>\n

Yazma ve Taslak<\/h3>\n

Yazma ve eskiz birbiriyle yak\u0131ndan ili\u015fkili iki moddur. Her ikisi de karma\u015f\u0131k d\u00fc\u015f\u00fcnceleri iletmek i\u00e7in bir ara\u00e7, genellikle kalem kullan\u0131r. Bir kalem kullanmak, belki de \u00f6\u011frencilerin yazma ve eskiz yapmada kullanmak i\u00e7in \u00f6\u011frendi\u011fi ilk becerilerden biridir, \u00f6zellikle erken d\u00f6nemlerde, \u00f6\u011frenmede h\u00e2l\u00e2 bask\u0131n bir faaliyettir. Bu moddan en yayg\u0131n bilgi \u00e7\u0131kar\u0131ma, \u00f6\u011frencilerin ne s\u00f6yledi\u011finin d\u00f6k\u00fcm\u00fcn\u00fc, yazma durumunda veya i\u00e7erik hakk\u0131ndaki bilginin \u00e7\u0131kt\u0131s\u0131n\u0131n al\u0131nabilece\u011fi eskizlerin yap\u0131land\u0131r\u0131lm\u0131\u015f bir g\u00f6sterimidir. Bununla birlikte, yazma ve \u00e7izim y\u00f6ntemlerini teknolojik yollarla yakalamak, insan g\u00f6zlemcilerin yazma h\u0131z\u0131, ritmi ve bask\u0131 seviyesi gibi kolayca tespit edemedi\u011fi bilgileri kullanma kap\u0131s\u0131n\u0131 a\u00e7ar. \u00d6\u011frenmeyi anlamadaki de\u011ferleri hala net olmasa da iyi bir uzman tahminci olabileceklerine dair g\u00f6stergeler vard\u0131r (Ochoa vd., 2013).<\/span><\/p>\n

Yazmak ve eskiz yapmak i\u00e7in en yayg\u0131n kullan\u0131lan kay\u0131t arac\u0131 dijital bir kalemdir (Oviatt ve Cohen, 2015). Bu kalemler farkl\u0131 y\u00fczeylerde yap\u0131lan vuru\u015flar\u0131n pozisyonunu, s\u00fcresini ve bas\u0131nc\u0131n\u0131 say\u0131salla\u015ft\u0131rabilir. Dijital formda olduktan sonra, bu bilgi \u00d6A ara\u00e7lar\u0131nda kullan\u0131labilir. Alternatif olarak, e\u011fitimde tabletlerin yayg\u0131n olarak kullan\u0131lmas\u0131 (Clarke ve Svanaes, 2014), \u00f6zellikle eskizlerde, bu modlar\u0131 kolayca yakalamak i\u00e7in bir f\u0131rsat sunmaktad\u0131r.<\/span><\/p>\n

\u00c7M\u00d6A alan\u0131nda, matematiksel veri derlemine dayanan iki \u00e7al\u0131\u015fma (Oviatt, Cohen ve Weibel, 2013), yazma ve eskiz modlar\u0131n\u0131n uzmanl\u0131k tahmine olabilecek katk\u0131s\u0131n\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Ochoa vd. (2013) yaz\u0131 karakteristiklerini (vuru\u015f h\u0131z\u0131 ve s\u00fcresi) \u00e7\u0131karm\u0131\u015f ve kullan\u0131lan basit geometrik fig\u00fcrlerin say\u0131s\u0131n\u0131 belirlemek i\u00e7in eskiz tan\u0131ma i\u015flemlerini ger\u00e7ekle\u015ftirmi\u015ftir. Sonu\u00e7lar, yazma h\u0131z\u0131n\u0131n uzmanl\u0131k d\u00fczeyi ile olduk\u00e7a ili\u015fkili oldu\u011funu belirlemi\u015ftir. Zhou, Hang, Oviatt, Yu ve Chen (2014), gruptaki uzman\u0131 %80 do\u011frulukla tan\u0131mlamak i\u00e7in yazma ve eskiz \u00f6zelliklerine dayal\u0131 s\u0131n\u0131fland\u0131rma sistemlerini kullanm\u0131\u015ft\u0131r.<\/span><\/p>\n

BA\u011eLAMLAR<\/h2>\n

\u00c7M\u00d6A ara\u015ft\u0131rmas\u0131n\u0131n temel amac\u0131, \u00d6A ara\u00e7lar\u0131n\u0131n ve metodolojilerinin, kolayca dijital izler sa\u011flamayan \u00f6\u011frenme ba\u011flamlar\u0131na uygulanmas\u0131n\u0131 geni\u015fletmektir. Bu ba\u011flamlar\u0131n bir \u00f6zelli\u011fi, \u00f6\u011frenme s\u00fcrecini anlamak i\u00e7in birden fazla modun yakalanmas\u0131n\u0131n gerekli olmas\u0131d\u0131r. Tablo 11.1, mevcut \u00c7M\u00d6A literat\u00fcr\u00fcnde incelenen ba\u011flam\u0131n, kullan\u0131lan modlar\u0131n, bu ba\u011flamlarda ara\u015ft\u0131r\u0131lan temel \u00f6\u011frenme y\u00f6nlerinin ve bu \u00e7al\u0131\u015fmalar\u0131n yap\u0131ld\u0131\u011f\u0131 \u00e7al\u0131\u015fmalar\u0131n ayr\u0131nt\u0131lar\u0131yla birlikte bir \u00f6zetini sunmaktad\u0131r.<\/span><\/p>\n

<\/a><\/a><\/a><\/a><\/a> Tablo 11.1. \u00c7M\u00d6A Taraf\u0131ndan \u00c7al\u0131\u015f\u0131lan \u00d6\u011frenme Ba\u011flamlar\u0131<\/i><\/span><\/span><\/p>\n\n\n\n\n\n<\/colgroup>\n\n\n\n\n\n\n\n
\n

Ba\u011flam<\/span><\/p>\n<\/td>\n

\n

Modlar<\/span><\/p>\n<\/td>\n

\n

\u00d6\u011frenme Y\u00f6nleri<\/span><\/p>\n<\/td>\n

\n

\u00c7al\u0131\u015fmalar<\/span><\/p>\n<\/td>\n<\/tr>\n

Dersler<\/span><\/span><\/td>\nHareket, Bak\u0131\u015f, Hareketler, Y\u00fcz \u0130fadesi, Konu\u015fma<\/span><\/span><\/td>\nDikkat, Soru-Cevap Etkile\u015fimleri<\/span><\/span><\/td>\nRaca ve Dillenbourg, 2013; Raca vd., 2014; Dominguez vd., 2015; D’Mello vd., 2015; Alibali vd., 2011<\/span><\/span><\/td>\n<\/tr>\n
S\u00f6zl\u00fc Sunumlar<\/span><\/span><\/td>\nDuru\u015f, Hareket, Jestler, Bak\u0131\u015f, Konu\u015fma, Dijital Belge<\/span><\/span><\/td>\nBeceri Geli\u015ftirme, Geribildirim, Zihinsel Durum<\/span><\/span><\/td>\nLuzardo vd., 2014; Echeverria vd., 2014; Chen vd., 2014; Leong vd., 2015; Schneider vd., 2015;<\/span><\/span><\/p>\n

Boncoddo vd., 2013<\/span><\/span><\/td>\n<\/tr>\n

Problem \u00e7\u00f6zme<\/span><\/span><\/td>\nHareket, Eylemler, Konu\u015fma, Yazma, Eskiz<\/span><\/span><\/td>\nUzmanl\u0131k Tahmini<\/span><\/span><\/td>\nOchoa vd., 2013; Luz, 2013; Thompson, 2013;<\/span><\/span><\/p>\n

Zhou vd., 2014<\/span><\/span><\/td>\n<\/tr>\n

Yap\u0131 Al\u0131\u015ft\u0131rmalar\u0131<\/span><\/span><\/td>\nHareketler, Eylemler, Konu\u015fma, Y\u00fcz \u0130fadeleri, Galvanik Cilt Tepkisi<\/span><\/span><\/td>\nYeniye kar\u015f\u0131 Uzman Kal\u0131plar\u0131<\/span><\/span><\/td>\nWorsley ve Blikstein, 2013, 2014b, 2015a<\/span><\/span><\/td>\n<\/tr>\n
Ak\u0131ll\u0131 \u00d6\u011freticilerin Kullan\u0131m\u0131<\/span><\/span><\/td>\nDijital G\u00fcnl\u00fck Dosyalar\u0131, Y\u00fcz \u0130fadeleri, Konu\u015fma<\/span><\/span><\/td>\nDuygular ve \u00d6\u011frenme \u0130li\u015fkisi<\/span><\/span><\/td>\nCraig vd., 2008; D’Mello vd., 2008<\/span><\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Dersler<\/h3>\n

Geleneksel olarak, dersler \u00f6\u011frenme ile ilgili en yayg\u0131n ba\u011flamd\u0131r. Bu ortam\u0131n \u00e7e\u015fitli y\u00f6nleri \u00e7al\u0131\u015fmay\u0131 hak ederken, \u00c7M\u00d6A ara\u015ft\u0131rmac\u0131lar\u0131 bug\u00fcne kadar derste \u00f6\u011frencilerin dikkatini otomatik olarak de\u011ferlendirmeye odaklanm\u0131\u015ft\u0131r. Raca ve Dillenbourg (2013) ve Raca vd. (2014), s\u0131n\u0131fta video kayd\u0131 ile bu kay\u0131ttan \u00f6\u011frenen hareketinin bak\u0131\u015f\u0131n\u0131n otomatik olarak \u00e7\u0131kar\u0131lmas\u0131n\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Bu \u00e7al\u0131\u015fmalar\u0131n sonu\u00e7lar\u0131, her iki modun da \u00f6\u011frenenin dikkatiyle ili\u015fkili oldu\u011funu ancak oturma pozisyonu gibi di\u011fer \u00f6nemli katk\u0131 sa\u011flay\u0131c\u0131lar\u0131n oldu\u011funu g\u00f6stermektedir. Dominguez, Echeverrfa, Chiluiza ve Ochoa (2015), \u00e7oklu model bir kay\u0131t arac\u0131 (\u00c7MKA) kullanarak video, ses ve kalem tabanl\u0131 modlar\u0131 yakalamak i\u00e7in yeni ve da\u011f\u0131t\u0131lm\u0131\u015f bir yol sundu. \u015eekil 11.5, b\u00f6yle bir arac\u0131n tasar\u0131m\u0131n\u0131 sunar. Cihaz\u0131n \u00f6\u011frencilere yak\u0131nl\u0131\u011f\u0131 t\u0131kanma riskini azalt\u0131r ve video ve ses yakalama kalitesini artt\u0131r\u0131r. Son olarak, D\u2019Mello vd. (2015), \u00f6\u011fretim \u00fcyesi ve \u00f6\u011frenciler aras\u0131ndaki soru-cevap etkile\u015fimlerini de\u011ferlendirmek i\u00e7in bir konferans ortam\u0131nda \u00e7e\u015fitli ses kay\u0131tlar\u0131 \u00fcretmi\u015ftir.<\/span><\/p>\n

\"\"<\/p>\n

<\/a> \u015eekil 11.5. Ders ayarlar\u0131nda kullan\u0131lacak \u00e7ok modlu bir kay\u0131t arac\u0131n\u0131n (\u00c7MKA) tasar\u0131m\u0131. \u00c7MKA s\u0131n\u0131fta (solda) ve \u00c7MKA \u00f6\u011frenenin bak\u0131\u015f a\u00e7\u0131s\u0131ndan (sa\u011fda)<\/i><\/span><\/span><\/p>\n

S\u00f6zl\u00fc Sunumlar<\/h3>\n

Bir akademik konuyu izleyicinin \u00f6n\u00fcnde sunma becerisi, genellikle y\u00fcksek\u00f6\u011frenim \u00f6\u011frencilerinin edinmesi gereken sosyal becerilerden biri olarak kabul edilir (Debnath vd., 2012). D\u00fcnyan\u0131n d\u00f6rt bir yan\u0131ndaki bir\u00e7ok ba\u011f\u0131ms\u0131z grup, yeni \u00f6\u011frencilerin k\u00f6t\u00fc uygulamalar\u0131 d\u00fczeltmelerine ve s\u00f6zl\u00fc sunumlarda ustal\u0131k kazanmalar\u0131na yard\u0131mc\u0131 olacak \u00c7M\u00d6A sistemleri kurmaya ba\u015flad\u0131. Echeverrfa vd. (2014) ve Luzardo vd. (2014) ayn\u0131 sistemin jest, duru\u015f, hareket, bak\u0131\u015f, konu\u015fma ve dijital sunum dosyalar\u0131n\u0131n analizini kullanan ve bir insan de\u011ferlendiricinin \u00f6\u011frenciye verece\u011fi notu tahmin edebilen farkl\u0131 y\u00f6nlerini sunmaktad\u0131r. Ayn\u0131 verileri analiz eden Chen, Leong, Feng ve Lee (2014), ayn\u0131 y\u00f6ntemi tahmin etmekte kullan\u0131lan bile\u015fik de\u011fi\u015fkenlerdeki farkl\u0131 y\u00f6ntemlerle de birle\u015ftirebilmi\u015ftir. B\u00f6rner, van Rosmalen ve Specht (2015) Kinect’i pozlar\u0131 tan\u0131mak ve ger\u00e7ek zamanl\u0131 olarak geri bildirim sa\u011flamak i\u00e7in kullanan sanal bir sunum beceri e\u011fitici4<\/a><\/sup> olu\u015fturdu.<\/span><\/p>\n

Problem \u00c7\u00f6zme<\/h3>\n

\u00d6zellikle STEM derslerinde \u00f6\u011frenme s\u0131kl\u0131kla, bireysel ve grupla problem \u00e7\u00f6zme oturumlar\u0131nda g\u00f6r\u00fcl\u00fcr (Silver, 2013). Matematik ve geometri problemlerini \u00e7\u00f6zen \u00fc\u00e7 farkl\u0131 lise \u00f6\u011frenci grubunun \u00e7oklu model kay\u0131tlar\u0131ndan olu\u015fan bir dizi matematiksel veri derleminin (Oviatt vd., 2013) varl\u0131\u011f\u0131, bu ortamda \u00c7M\u00d6A ara\u015ft\u0131rmalar\u0131n\u0131 h\u0131zland\u0131rd\u0131. Veri k\u00fcmesinde yer alan medya, her \u00f6\u011frenenin \u00f6nden video kayd\u0131n\u0131, \u00e7al\u0131\u015fma masas\u0131n\u0131n video kayd\u0131n\u0131, her \u00f6\u011frenenin ses kayd\u0131n\u0131 ve odan\u0131n genel sesini i\u00e7erir. Ayr\u0131ca, \u00f6\u011frencilerin dijital kalemler ile donat\u0131lm\u0131\u015flard\u0131r. \u00d6\u011frencilerin uzmanl\u0131k d\u00fczeyi ile ilgili ger\u00e7ek referans de\u011fer sa\u011flanm\u0131\u015ft\u0131r. Luz (2013), Thompson (2013), Ochoa vd. (2013) ve Zhou vd. (2014) bu veri k\u00fcmesini farkl\u0131 modlar kullanarak analiz etmi\u015f, t\u00fcm modalitelerin uzmanl\u0131k seviyesinin y\u00fcksek bir do\u011fruluk d\u00fczeyiyle (> %70) belirlenmesine katk\u0131da bulunmu\u015ftur.<\/span><\/p>\n

Yap\u0131 Al\u0131\u015ft\u0131rmalar\u0131<\/h3>\n

M\u00fchendislik tasar\u0131m\u0131 ve in\u015fas\u0131 i\u00e7in gerekli bilgi ve beceriler, k\u00fc\u00e7\u00fck yap\u0131 zorluklar\u0131 ile test edilebilir (Householder ve Hailey, 2012). Worsley ve Blikstein’\u0131n (2013, 2014b, 2015a) son derece \u00f6nemli eserleri, \u00e7ok modlu analizlerle, yap\u0131lar\u0131n tasar\u0131m\u0131 ve elle montaj\u0131nda uzmanlar ve yeni ba\u015flayanlar taraf\u0131ndan ger\u00e7ekle\u015ftirilen eylem \u00f6r\u00fcnt\u00fclerini ara\u015ft\u0131r\u0131yor. Analiz i\u00e7in kullan\u0131lan ana modlar; hareketler, eylemler, konu\u015fma, y\u00fcz ifadesi ve galvanik cilt tepkisi idi. Bu modlardan \u00e7\u0131kar\u0131lan izlerin kombinasyonu, yap\u0131m s\u00fcrecinde m\u00fchendislik tasar\u0131m\u0131ndaki ustal\u0131k seviyesini tan\u0131mlamaya yard\u0131mc\u0131 olan farkl\u0131l\u0131klar\u0131 ortaya koyuyor.<\/span><\/p>\n

Ak\u0131ll\u0131 \u00d6\u011freticilerin Kullan\u0131m\u0131<\/h3>\n

A\u00d6S’ler genellikle geleneksel \u00d6A taraf\u0131ndan kay\u0131t g\u00fcnl\u00fc\u011f\u00fc dosyalar\u0131 kullan\u0131larak incelenir. Bununla birlikte, etkile\u015fim verilerini tamamlayan yeni modlar eklemek i\u00e7in \u00f6\u011frenenin videosu ve sesi \u00e7ekilmi\u015ftir. Videodan \u00e7\u0131kar\u0131lan ana modlar y\u00fcz ifadesi (Craig vd., 2008) ve konu\u015fma (D’Mello vd., 2008) olup, \u00f6\u011frenenin i\u00e7sel duygusal durumunu temsil eder. Her ikisi de A\u00d6S’yi kullan\u0131rken can s\u0131k\u0131nt\u0131s\u0131, kar\u0131\u015f\u0131kl\u0131k ve hayal k\u0131r\u0131kl\u0131\u011f\u0131 gibi duygusal durumlar\u0131 ba\u015far\u0131yla tespit edebilir.<\/span><\/p>\n

\u00d6ZEL KONULAR<\/h2>\n

\u00d6A modellerinde ve uygulamalar\u0131nda \u00e7ok modelli izlerin kullan\u0131lmas\u0131, ayn\u0131 moddan elde edilen farkl\u0131 izlerin kullan\u0131lmas\u0131na benzer. Bununla birlikte, \u00c7M\u00d6A ara\u015ft\u0131rma ve uygulamalar\u0131, belirli modaliteler yakaland\u0131\u011f\u0131nda, i\u015flendi\u011finde ve analiz edildi\u011finde baz\u0131 spesifik sorunlar\u0131 ortaya \u00e7\u0131karmaktad\u0131r. Bu meseleler, bir\u00e7ok modaliteden izlerin teknik olarak \u00e7\u0131kar\u0131lmas\u0131na paralel olarak ancak \u00c7M\u00d6A \u00e7\u00f6z\u00fcmlerinin ger\u00e7ek d\u00fcnyada etkili bir \u015fekilde konu\u015fland\u0131r\u0131lmas\u0131 i\u00e7in \u00f6nemli olan a\u00e7\u0131k ara\u015ft\u0131rma alanlar\u0131 olmaya devam etmektedir.<\/span><\/p>\n

Kay\u0131t<\/h3>\n

Dijital bir ara\u00e7ta etkile\u015fim bilgilerini yakalamak, kodun ilgili k\u0131s\u0131mlar\u0131na g\u00fcnl\u00fck ifadeleri eklemek kadar kolay ve ucuzdur. Bu ifadeler, \u00f6\u011frenenden herhangi bir kat\u0131l\u0131m gerektirmeden, \u015feffaf ve genel olarak g\u00fcvenilir ve hatas\u0131z bir \u015fekilde otomatik olarak ger\u00e7ekle\u015ftirilir. \u00d6te yandan, ger\u00e7ek d\u00fcnyadaki medyay\u0131 yakalamak, kay\u0131t ara\u00e7lar\u0131n\u0131n (kameralar, mikrofonlar, dijital kalemler, vb.) edinilmesini, kurulmas\u0131n\u0131 ve kullan\u0131lmas\u0131n\u0131, sistemi a\u00e7\u0131p kapatmas\u0131n\u0131 ve izlenmesini ve t\u0131kan\u0131kl\u0131k, parazit veya g\u00fcr\u00fclt\u00fc yoluyla kayd\u0131n bozulmas\u0131n\u0131 \u00f6nlemeyi gerektirir. Dijital kay\u0131t kadar zahmetsiz ve verimli \u00e7al\u0131\u015fan kay\u0131t sistemlerinin geli\u015ftirilmesi, \u00c7M\u00d6A’n\u0131n geli\u015fmesinin \u00f6n\u00fcndeki en b\u00fcy\u00fck engellerden biridir. Bu bir m\u00fchendislik problemi olsa da ara\u015ft\u0131rmac\u0131lar \u00e7\u00f6z\u00fcmlerinin uygulanabilirli\u011fi ve \u00f6l\u00e7eklenebilirli\u011finin fark\u0131nda olmal\u0131d\u0131rlar. Ana tekliflerden biri, kay\u0131tlar\u0131n her zaman a\u00e7\u0131kta kalan ucuz alg\u0131lay\u0131c\u0131lar\u0131 kullanarak merkezden da\u011f\u0131t\u0131lmas\u0131d\u0131r. Bir veya daha fazla kay\u0131t sorun \u00e7\u0131kar\u0131rsa, genel bilgiler kalan \u00e7al\u0131\u015fma alg\u0131lay\u0131c\u0131lar\u0131ndan yeniden olu\u015fturulabilir.<\/span><\/p>\n

Gizlilik<\/h3>\n

Etkile\u015fim bilgilerini dijital ara\u00e7larla yakalamak, \u00f6\u011frenciler ve \u00f6\u011fretenler aras\u0131ndaki \u015fimdiden mahremiyet endi\u015felerini ortaya \u00e7\u0131karmaktad\u0131r (Pardo ve Siemens, 2014). \u201c19845<\/a><\/sup>\u201d g\u00f6zetim seviyelerini taklit eden kay\u0131t sistemlerinin kurulmas\u0131 ve kullan\u0131lmas\u0131n\u0131n, g\u00fc\u00e7l\u00fc bir diren\u00e7 ortaya \u00e7\u0131karaca\u011f\u0131 muhakkakt\u0131r. A\u00e7\u0131k r\u0131za formlar\u0131 erken ara\u015ft\u0131rma a\u015famalar\u0131 i\u00e7in i\u015fe yarayabilirdi ancak ger\u00e7ek d\u00fcnyada \u00c7M\u00d6A sistemlerinin benimsenmesi farkl\u0131 ve daha yarat\u0131c\u0131 bir yakla\u015f\u0131m gerektiriyordu. Bu alandaki en umut verici \u00e7\u00f6z\u00fcmlerden biri, verinin m\u00fclkiyetini \u00f6\u011frenene aktarmakt\u0131r. En mahrem ki\u015fisel bilgileriniz al\u0131nsa bile, neyin ne zaman payla\u015f\u0131laca\u011f\u0131 \u00f6\u011frencinin kontrol\u00fcnde kal\u0131rsa gizlilik kayg\u0131lar\u0131 etkisiz hale getirilir. Bu yakla\u015f\u0131m, \u00e7e\u015fitli niceliksel kendi kendine yap\u0131lan uygulamalara benzer (Swan, 2013). <\/span><\/p>\n

Entegrasyon<\/h4>\n

B\u00fcy\u00fck miktarda ham \u00f6\u011frenme izlerinin mevcudiyeti ile ilgili bir soru, \u00f6\u011frenme s\u00fcrecini anlamak, optimize etmek ve yararl\u0131 bilgiler \u00fcretmek i\u00e7in bunlar\u0131 nas\u0131l birle\u015ftirece\u011fimizdir. Farkl\u0131 i\u015flemler kullan\u0131larak farkl\u0131 modlardan \u00e7\u0131kar\u0131lan izler \u00e7ok farkl\u0131 \u00f6zelliklere sahip olmak zorundad\u0131r. \u00d6rne\u011fin, farkl\u0131 modlardan \u00e7\u0131kar\u0131lan izlerin zaman taneciklili\u011fi geni\u015f \u00f6l\u00e7\u00fcde de\u011fi\u015febilir. Konu\u015fman\u0131n prosodik y\u00f6nlerinden \u00e7\u0131kar\u0131lan izler saniyenin onda birinde de\u011fi\u015febilirken, duru\u015flar daha yava\u015f de\u011fi\u015fiyor. \u00c7\u0131kar\u0131lan izlerin kesinli\u011fi de farkl\u0131 olabilir. Y\u00fcksek kaliteli kay\u0131tlarla yap\u0131lan konu\u015fma tan\u0131ma %90 do\u011frulu\u011fa ula\u015f\u0131rken, web kameras\u0131 kaynaklar\u0131ndan gelen duygusal durum tespiti 70’lerin alt\u0131nda kalabilir. Bu fark ba\u015far\u0131l\u0131 analizleri engellemez ancak sahte sonu\u00e7lar\u0131 \u00f6nlemek i\u00e7in duyarl\u0131 tasar\u0131m gereklidir. \u00c7M\u00d6A, Worsley ve Blikstein (2014a) ‘daki bu ara\u015ft\u0131rma hatt\u0131na \u00f6nc\u00fcl\u00fck etmek, Newell (1994) ve Anderson (2002) taraf\u0131ndan insan bili\u015fine bir a\u00e7\u0131klama olarak \u00f6nerilen \u201cbili\u015f bantlar\u0131\u201d \u00e7er\u00e7evesine dayanan \u00e7e\u015fitli birle\u015ftirme stratejileri \u00f6nermektedir. Entegrasyon \u00e7er\u00e7evelerinin geli\u015ftirilmesi sadece \u00c7M\u00d6A\u2019ya de\u011fil t\u00fcm \u00d6A toplulu\u011funa fayda sa\u011flayacakt\u0131r.<\/span><\/p>\n

\u00d6\u011frenmeye Etkisi<\/h3>\n

\u00c7ok kullan\u0131c\u0131l\u0131 \u00f6\u011frenme analitiklerine dayanan son kullan\u0131c\u0131 ara\u00e7lar\u0131 ve m\u00fcdahaleleri, tek modlu analizlere dayananlara benzer olsa da \u00e7ok modlu olanlar i\u00e7in gerekli kullan\u0131\u015fl\u0131l\u0131k, veri toplama ek karma\u015f\u0131kl\u0131\u011f\u0131n\u0131 hakl\u0131 \u00e7\u0131karmak i\u00e7in daha y\u00fcksek olmal\u0131d\u0131r. \u00d6rne\u011fin, \u00d6YS taraf\u0131ndan otomatik olarak toplanan verilere dayal\u0131 bir g\u00f6sterge tablosu uygulamas\u0131n\u0131n kabul edilmesi, t\u00fcm s\u0131n\u0131flar\u0131n video kameralarla donat\u0131lmas\u0131n\u0131 gerektiren benzer bir g\u00f6sterge panelinden daha kolay olacakt\u0131r. Artan karma\u015f\u0131kl\u0131\u011fa, \u00f6\u011frenme s\u00fcreci \u00fczerinde daha b\u00fcy\u00fck bir olumlu etki e\u015flik etmelidir. \u00d6\u011frenmeyi analiz etmek i\u00e7in \u00e7oklu ger\u00e7ek d\u00fcnya sinyallerini kullanma ihtiyac\u0131, s\u00fcre\u00e7 hakk\u0131nda daha yararl\u0131 bilgiler ve \u00f6\u011frenenler \u00fczerinde daha \u00f6l\u00e7\u00fclebilir etkiler sa\u011flama vaadini de sunmal\u0131d\u0131r.<\/span><\/p>\n

SONU\u00c7<\/h2>\n

\u00d6A, \u00f6\u011frenme s\u00fcrecini anlamak ve optimize etmek i\u00e7in kullan\u0131lan yakla\u015f\u0131mlarda devrim yaratt\u0131. Ancak yaln\u0131zca bilgisayar tabanl\u0131 \u00f6\u011frenmeyi i\u00e7eren \u00e7al\u0131\u015fmalara ve ara\u00e7lara y\u00f6nelik mevcut yanl\u0131l\u0131\u011f\u0131, genel olarak \u00f6\u011frenmeye uygulanabilirli\u011fini tehlikeye atmaktad\u0131r. \u00c7M\u00d6A, bilgisayar destekli olmayan \u00f6\u011frenme ba\u011flamlar\u0131n\u0131 \u00d6A’n\u0131n ana ara\u015ft\u0131rma ve uygulamas\u0131na entegre etmeye \u00e7al\u0131\u015fan bir alt aland\u0131r.<\/span><\/p>\n

Bu b\u00f6l\u00fcm \u00c7M\u00d6A\u2019daki en son teknolojiyi sunmaktad\u0131r. Daha geleneksel t\u0131klama ak\u0131\u015f bilgisi ve metin i\u00e7eri\u011fi modlar\u0131n\u0131n yan\u0131 s\u0131ra duru\u015f, konu\u015fma ve eskiz gibi \u00e7e\u015fitli modlar, ara\u015ft\u0131rma sorular\u0131n\u0131 cevaplamak ve \u00f6\u011frenme ba\u011flamlar\u0131nda geri bildirim sistemleri olu\u015fturmak i\u00e7in kullan\u0131lm\u0131\u015ft\u0131r. Bilgisayar bilimi tekniklerinin ve e\u011fitim ve davran\u0131\u015f bilimcilerinin sa\u011flad\u0131\u011f\u0131 i\u00e7g\u00f6r\u00fclerden olu\u015fan bir birliktelik, s\u0131n\u0131flar, \u00e7al\u0131\u015fma gruplar\u0131 ve s\u00f6zl\u00fc sunumlar gibi \u00e7ok \u00e7e\u015fitli \u00f6\u011frenme ba\u011flamlar\u0131n\u0131n otomatik olarak de\u011ferlendirilmesini sa\u011flar.<\/span><\/p>\n

Bu b\u00f6l\u00fcmde sunulan ara\u015ft\u0131rma listesinden \u00e7\u0131kar\u0131labilece\u011fi gibi, \u00c7M\u00d6A hala k\u00fc\u00e7\u00fck ama \u00e7ok aktif ve a\u00e7\u0131k bir ara\u015ft\u0131rmac\u0131 toplulu\u011funa sahip olan yeni bir aland\u0131r. \u00c7ok modlu veri k\u00fcmelerinin serbest\u00e7e payla\u015f\u0131ld\u0131\u011f\u0131 ve ortakla\u015fa tart\u0131\u015f\u0131lan yeni tasar\u0131m fikirleriyle birlikte analiz edildi\u011fi d\u00fczenli zorluklar\u0131n ve at\u00f6lye \u00e7al\u0131\u015fmalar\u0131n\u0131n varl\u0131\u011f\u0131, yeni bilgilerin h\u0131zla \u00fcretildi\u011fi bir ara\u015ft\u0131rma ortam\u0131 yarat\u0131r.<\/span><\/p>\n

Baz\u0131 konular hala \u00c7M\u00d6A’n\u0131n ana uygulama haline gelmesini engellerken, aktif ara\u015ft\u0131rma projeleri bu konulara y\u00f6nelik \u00e7\u00f6z\u00fcmleri ara\u015ft\u0131rmaktad\u0131r, \u00e7oklu model \u00f6\u011frenme izlerini daha ucuz, daha az m\u00fcdahaleci ve daha otomatik hale getirmektedir. \u00c7M\u00d6A toplulu\u011fundan do\u011fan, mahremiyetle ilgili kayg\u0131lar\u0131 gidermek i\u00e7in ortaya \u00e7\u0131kan yeni \u00e7\u00f6z\u00fcmler, \u00f6rne\u011fin da\u011f\u0131t\u0131lm\u0131\u015f kay\u0131t sa\u011flamak ve verilerin \u00f6\u011frenenle birlikte dinlenmesi gibi, bir g\u00fcn genel \u00d6A uygulamalar\u0131 i\u00e7in normlar\u0131 olu\u015fturabilir.<\/span><\/p>\n

Son olarak, yazar \u00d6A ara\u015ft\u0131rmac\u0131lar\u0131n\u0131 ve uygulay\u0131c\u0131lar\u0131n\u0131 kendi \u00e7al\u0131\u015fmalar\u0131nda ve ara\u00e7lar\u0131nda birden fazla y\u00f6ntemi kullanmay\u0131 ke\u015ffetmeye davet ediyor. \u00c7M\u00d6A toplulu\u011fu, bilgilerini, verilerini, kodunu ve \u00e7er\u00e7evelerini a\u00e7\u0131k\u00e7a payla\u015facakt\u0131r. Sadece bu farkl\u0131 y\u00f6ntemlerin benimsenmesi, \u00d6A’n\u0131n \u00f6\u011frenmenin ger\u00e7ekle\u015fti\u011fi t\u00fcm ba\u011flamlarda bir etkiye sahip olmas\u0131na izin verecektir.<\/span><\/p>\n

KAYNAK\u00c7A<\/h2>\n

Alibali, M. W., Nathan, M. J., Fujimori, Y., Stein, N., & Raudenbush, S. (2011). Gestures in the mathematics classroom: What\u2019s the point? In N. Stein & S. Raudenbush (Eds.), Developmental cognitive science goes to school <\/i>(pp. 219\u2013234). New York: Routledge, Taylor & Francis. <\/span><\/span><\/p>\n

Almeda, M. V., Scupelli, P., Baker, R. S., Weber, M., & Fisher, A. (2014). Clustering of design decisions in classroom visual displays. Proceedings of the 4th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201914), 24\u201328 March 2014, Indianapolis, IN, USA (pp. 44\u201348). New York: ACM. <\/span><\/span><\/p>\n

Anderson, J. R. (2002). Spanning seven orders of magnitude: A challenge for cognitive modeling. Cognitive Science, 26<\/i>(1), 85\u2013112. <\/span><\/span><\/p>\n

Arnold, K. E., & Pistilli, M. D. (2012). Course Signals at Purdue: Using learning analytics to increase student success. Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 267\u2013270). New York: ACM. <\/span><\/span><\/p>\n

Blikstein, P. (2013). Multimodal learning analytics. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 102\u2013106). New York: ACM. <\/span><\/span><\/p>\n

Boraston, Z., & Blakemore, S.-J. (2007). The application of eye-tracking technology in the study of autism. The Journal of Physiology, 581<\/i>(3), 893\u2013898. <\/span><\/span><\/p>\n

Bull, P. E. (2013). Posture & Gesture<\/i>. Elsevier. <\/span><\/span><\/p>\n

Boncoddo, R., Williams, C., Pier, E., Walkington, C., Alibali, M., Nathan, M., Dogan, M. & Waala, J. (2013). Gesture as a window to justification and proof. In M. V. Martinez & A. C. Superfine (Eds.), Proceedings of the 35th annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education <\/i>(PME-NA 35) 14\u201317 November 2013, Chicago, IL, USA (pp. 229\u2013236). http:\/\/www.pmena.org\/proceedings\/ <\/span><\/span><\/p>\n

Chen, L., Leong, C. W., Feng, G., & Lee, C. M. (2014). Using multimodal cues to analyze MLA\u201914 oral presentation quality corpus: Presentation delivery and slides quality. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 45\u201352). New York: ACM. <\/span><\/span><\/p>\n

Clarke, B., & Svanaes, S. (2014, April 9). An updated literature review on the use of tablets in education. Family Kids and Youth. https:\/\/smartfuse.s3.amazonaws.com\/mysandstorm.org\/uploads\/2014\/05\/T4S-Use-of- Tablets-in-Education.pdf <\/span><\/span><\/p>\n

Cobb, P., Confrey, J., Lehrer, R., Schauble, L., & others. (2003). Design experiments in educational research. Educational Researcher, 32<\/i>(1), 9\u201313.<\/span><\/span><\/p>\n

Craig, S. D., D\u2019Mello, S., Witherspoon, A., & Graesser, A. (2008). Emote aloud during learning with AutoTutor: Applying the facial action coding system to cognitive\u2013affective states during learning. Cognition and Emotion, 22<\/i>(5), 777\u2013788. <\/span><\/span><\/p>\n

Crossley, S. A., Roscoe, R. D., & McNamara, D. S. (2013). Using automatic scoring models to detect changes in student writing in an intelligent tutoring system. Proceedings of the 26th Annual Florida Artificial Intelligence Research Society Conference <\/i>(FLAIRS-13), 20\u201322 May 2013, St. Pete Beach, FL, USA (pp. 208\u2013213). Menlo Park, CA: The AAAI Press. <\/span><\/span><\/p>\n

Debnath, M., Pandey, M., Chaplot, N., Gottimukkula, M. R., Tiwari, P. K., & Gupta, S. N. (2012). Role of soft skills in engineering education: Students\u2019 perceptions and feedback. In C. S. Nair, A. Patil, & P. Mertova (Eds.), Enhancing learning and teaching through student feedback in engineering <\/i>(pp. 61\u201382). ScienceDirect. http: \/\/ www.sciencedirect.com\/science\/book\/9781843346456 <\/span><\/span><\/p>\n

D\u2019Mello, S. K., Jackson, G. T., Craig, S. D., Morgan, B., Chipman, P., White, H., Person, N., Kort, B., el Kaliouby, R., Picard, R., & Graesser, A. C. (2008). AutoTutor detects and responds to learners\u2019 affective and cognitive states. Workshop on Emotional and Cognitive Issues in ITS, held in conjunction with the 9th International Conference on Intelligent Tutoring Systems (ITS 2008), 23\u201327 June 2008, Montreal, PQ, Canada. https:\/\/ www.researchgate.net\/publication\/228673992_AutoTutor_detects_and_responds_to_learners_affective_and_cognitive_states <\/span><\/span><\/p>\n

D\u2019Mello, S., Olney, A., Blanchard, N., Samei, B., Sun, X., Ward, B., & Kelly, S. (2015). Multimodal capture of teacher\u2013student interactions for automated dialogic analysis in live classrooms. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 557\u2013566). New York: ACM. <\/span><\/span><\/p>\n

Dominguez, F., Echeverr\u00eda, V., Chiluiza, K., & Ochoa, X. (2015). Multimodal selfies: Designing a multimodal recording device for students in traditional classrooms. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 567\u2013574). New York: ACM. <\/span><\/span><\/p>\n

Echeverr\u00eda, V., Avenda\u00f1o, A., Chiluiza, K., V\u00e1squez, A., & Ochoa, X. (2014). Presentation skills estimation based on video and Kinect data analysis. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 53\u201360). New York: ACM. <\/span><\/span><\/p>\n

Freedman, D. H. (2010, December 10). Why scientific studies are so often wrong: The streetlight effect. Discover Magazine, 26<\/i>. http:\/\/discovermagazine.com\/2010\/jul-aug\/29-why-scientific-studies-often-wrong-streetlight-effect <\/span><\/span><\/p>\n

Frischen, A., Bayliss, A. P., & Tipper, S. P. (2007). Gaze cueing of attention: Visual attention, social cognition, and individual differences. Psychological Bulletin, 133<\/i>(4), 694\u2013724. <\/span><\/span><\/p>\n

Gall, M. D., Borg, W. R., & Gall, J. P. (1996). Educational research: An introduction<\/i>. Longman Publishing. <\/span><\/span><\/p>\n

Graesser, A. C., Chipman, P., Haynes, B. C., & Olney, A. (2005). AutoTutor: An intelligent tutoring system with mixed-initiative dialogue. IEEE Transactions on Education, 48<\/i>(4), 612\u2013618. <\/span><\/span><\/p>\n

Householder, D. L., & Hailey, C. E. (2012). Incorporating engineering design challenges into STEM courses<\/i>. National Center for Engineering and Technology Education. http:\/\/ncete.org\/flash\/pdfs\/NCETECaucusReport.pdf <\/span><\/span><\/p>\n

Jewitt, C. (2006). Technology, literacy and learning: A multimodal approach. Psychology Press. <\/span><\/span><\/p>\n

Kizilcec, R. F., Piech, C., & Schneider, E. (2013). Deconstructing disengagement: Analyzing learner subpopulations in massive open online courses. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 170\u2013179). New York: ACM. <\/span><\/span><\/p>\n

Kress, G., & Van Leeuwen, T. (2001). Multimodal discourse: The modes and media of contemporary communication. Edward Arnold. <\/span><\/span><\/p>\n

Krugman, D. M., Fox, R. J., Fletcher, J. E., Fischer, P. M., & Rojas, T. H. (1994). Do adolescents attend to warnings in cigarette advertising? An eye-tracking approach. Journal of Advertising Research, 34<\/i>, 39\u201351. <\/span><\/span><\/p>\n

Kruschke, J. K. (2003). Attention in learning. Current Directions in Psychological Science, 12<\/i>(5), 171\u2013175.<\/span><\/span><\/p>\n

Leong, C. W., Chen, L., Feng, G., Lee, C. M., & Mulholland, M. (2015). Utilizing depth sensors for analyzing multimodal presentations: Hardware, software and toolkits. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 547\u2013556). New York: ACM. <\/span><\/span><\/p>\n

Lin, Y.-T., Lin, R.-Y., Lin, Y.-C., & Lee, G. C. (2013). Real-time eye-gaze estimation using a low-resolution webcam. Multimedia Tools and Applications, 65<\/i>(3), 543\u2013568. <\/span><\/span><\/p>\n

Lubold, N., & Pon-Barry, H. (2014). Acoustic-prosodic entrainment and rapport in collaborative learning dialogues. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 5\u201312). New York: ACM. <\/span><\/span><\/p>\n

Lund, K. (2007). The importance of gaze and gesture in interactive multimodal explanation. Language Resources and Evaluation, 41<\/i>(3\u20134), 289\u2013303. <\/span><\/span><\/p>\n

Luzardo, G., Guam\u00e1n, B., Chiluiza, K., Castells, J., & Ochoa, X. (2014). Estimation of presentations skills based on slides and audio features. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 37\u201344). New York: ACM. <\/span><\/span><\/p>\n

Luz, S. (2013). Automatic identification of experts and performance prediction in the multimodal math data corpus through analysis of speech interaction. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 575\u2013582). New York: ACM. <\/span><\/span><\/p>\n

Markaki, V., Lund, K., & Sanchez, E. (2015). Design digital epistemic games: A longitudinal multimodal analysis. Paper presented at the conference Revisiting Participation: Language and Bodies in Interaction<\/i>, 24\u201327 June 2015, Basel, Switzerland. <\/span><\/span><\/p>\n

Mazur-Palandre, A., Colletta, J. M., & Lund, K. (2014). Context sensitive \u201chow\u201d explanation in children\u2019s multimodal behavior, Journal of Multimodal Communication Studies, 2<\/i>, 1\u201317. <\/span><\/span><\/p>\n

Mishra, B., Fernandes, S. L., Abhishek, K., Alva, A., Shetty, C., Ajila, C. V., \u2026 Shetty, P. (2015). Facial expression recognition using feature based techniques and model based techniques: A survey. Proceedings of the 2nd International Conference on Electronics and Communication Systems <\/i>(ICECS 2015), 26\u201327 February 2015, Coimbatore, India (pp. 589\u2013594). IEEE. <\/span><\/span><\/p>\n

Mitra, S., & Acharya, T. (2007). Gesture recognition: A survey. IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, 37(3), 311\u2013324. <\/span><\/span><\/p>\n

Morency, L.-P., Oviatt, S., Scherer, S., Weibel, N., & Worsley, M. (2013). ICMI 2013 grand challenge workshop on multimodal learning analytics. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 373\u2013378). New York: ACM. <\/span><\/span><\/p>\n

Newell, A. (1994). Unified theories of cognition<\/i>. Harvard University Press. <\/span><\/span><\/p>\n

Ochoa, X., Chiluiza, K., M\u00e9ndez, G., Luzardo, G., Guam\u00e1n, B., & Castells, J. (2013). Expertise estimation based on simple multimodal features. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 583\u2013590). New York: ACM. <\/span><\/span><\/p>\n

Ochoa, X., Worsley, M., Chiluiza, K., & Luz, S. (2014). MLA\u201914: Third multimodal learning analytics workshop and grand challenges. Proceedings of the 16th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 531\u2013532). New York: ACM. <\/span><\/span><\/p>\n

Oviatt, S., Cohen, A., & Weibel, N. (2013). Multimodal learning analytics: Description of math data corpus for ICMI grand challenge workshop. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 583\u2013590). New York: ACM. <\/span><\/span><\/p>\n

Oviatt, S., & Cohen, P. R. (2015). The paradigm shift to multimodality in contemporary computer interfaces<\/i>. San Rafael, CA: Morgan & Claypool Publishers. <\/span><\/span><\/p>\n

Pardo, A., & Siemens, G. (2014). Ethical and privacy principles for learning analytics. British Journal of Educational Technology, 45<\/i>(3), 438\u2013450. <\/span><\/span><\/p>\n

Poole, A., & Ball, L. J. (2006). Eye tracking in HCI and usability research. Encyclopedia of Human Computer Interaction, 1<\/i>, 211\u2013219.<\/span><\/span><\/p>\n

Raca, M., & Dillenbourg, P. (2013). System for assessing classroom attention. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 265\u2013269). New York: ACM. <\/span><\/span><\/p>\n

Raca, M., Tormey, R., & Dillenbourg, P. (2014). Sleepers\u2019 lag: Study on motion and attention. Proceedings of the 4th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201914), 24\u201328 March 2014, Indianapolis, IN, USA (pp. 36\u201343). New York: ACM. <\/span><\/span><\/p>\n

Scherer, S., Worsley, M., & Morency, L.-P. (2012). 1st international workshop on multimodal learning analytics. Proceedings of the 14th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201912), 22\u201326 October 2012, Santa Monica, CA, USA (pp. 609\u2013610). New York: ACM. <\/span><\/span><\/p>\n

Schl\u00f6mer, T., Poppinga, B., Henze, N., & Boll, S. (2008). Gesture recognition with a Wii controller. Proceedings of the 2nd International Conference on Tangible and Embedded Interaction <\/i>(TEI\u201908), 18\u201321 February 2008, Bonn, Germany (pp. 11\u201314). New York: ACM. <\/span><\/span><\/p>\n

Schneider, J., B\u00f6rner, D., van Rosmalen, P., & Specht, M. (2015). Presentation trainer, your public speaking multimodal coach. Proceedings of the 17th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201915), 9\u201313 November 2015, Seattle, WA, USA (pp. 539\u2013546). New York: ACM. <\/span><\/span><\/p>\n

Serrano-Laguna, A., & Fern\u00e1ndez-Manj\u00f3n, B. (2014). Applying learning analytics to simplify serious games deployment in the classroom. Proceedings of the 2014 IEEE Global Engineering Education Conference <\/i>(EDUCON 2014), 3\u20135 April 2014, Istanbul, Turkey (pp. 872\u2013877). IEEE. <\/span><\/span><\/p>\n

Siemens, G. (2013). Learning analytics: The emergence of a discipline. American Behavioral Scientist, 9<\/i>, 51\u201360. doi:0002764213498851. <\/span><\/span><\/p>\n

Silver, E.A. (2013). Teaching and learning mathematical problem solving: Multiple research perspectives. Routledge. <\/span><\/span><\/p>\n

Simsek, D., S\u00e1ndor, \u00c1., Buckingham Shum, S., Ferguson, R., De Liddo, A., & Whitelock, D. (2015). Correlations between automated rhetorical analysis and tutors\u2019 grades on student essays. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 355\u2013359). New York: ACM. <\/span><\/span><\/p>\n

Swan, M. (2013). The quantified self: Fundamental disruption in big data science and biological discovery. Big Data, 1<\/i>(2), 85\u201399. <\/span><\/span><\/p>\n

Thompson, K. (2013). Using micro-patterns of speech to predict the correctness of answers to mathematics problems: An exercise in multimodal learning analytics. Proceedings of the 15th ACM International Conference on Multimodal Interaction <\/i>(ICMI\u201913), 9\u201313 December 2013, Sydney, Australia (pp. 591\u2013598). New York: ACM. <\/span><\/span><\/p>\n

Vasquez, H. A., Vargas, H. S., & Sucar, L. E. (2015). Using gestures to interact with a service robot using Kinect 2. Advances in Computer Vision and Pattern Recognition, 85<\/i>. Springer. <\/span><\/span><\/p>\n

Vatrapu, R., Reimann, P., Bull, S., & Johnson, M. (2013). An eye-tracking study of notational, informational, and emotional aspects of learning analytics representations. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 125\u2013134). New York: ACM. <\/span><\/span><\/p>\n

Worsley, M., & Blikstein, P. (2013). Towards the development of multimodal action based assessment. Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 94\u2013101). New York: ACM. <\/span><\/span><\/p>\n

Worsley, M., & Blikstein, P. (2014a). Deciphering the practices and affordances of different reasoning strategies through multimodal learning analytics. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 21\u201327). New York: ACM. <\/span><\/span><\/p>\n

Worsley, M., & Blikstein, P. (2014b). Using multimodal learning analytics to study learning mechanisms. In J. Stamper, Z. Pardos, M. Mavrikis, & B. M. McLaren (Eds.), Proceedings of the 7th International Conference on Educational Data Mining <\/i>(EDM2014), 4\u20137 July, London, UK (pp. 431\u2013432). International Educational Data Mining Society.<\/span><\/span><\/p>\n

Worsley, M., & Blikstein, P. (2015a). Leveraging multimodal learning analytics to differentiate student learning strategies. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 360\u2013367). New York: ACM. <\/span><\/span><\/p>\n

Worsley, M., & Blikstein, P. (2015b). Using learning analytics to study cognitive disequilibrium in a complex learning environment. Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 426\u2013427). New York: ACM. <\/span><\/span><\/p>\n

Zhang, Z. (2012). Microsoft Kinect sensor and its effect. IEEE MultiMedia, 19<\/i>(2), 4\u201310. <\/span><\/span><\/p>\n

Zhou, J., Hang, K., Oviatt, S., Yu, K., & Chen, F. (2014). Combining empirical and machine learning techniques to predict math expertise using pen signal features. Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge <\/i>(MLA\u201914), 12\u201316 November 2014, Istanbul, Turkey (pp. 29\u201336). New York: ACM.<\/span><\/span><\/p>\n

\n
\n

1<\/a> orj. modalities<\/span><\/span><\/p>\n<\/div>\n

\n

2<\/a> orj.capture<\/span><\/span><\/p>\n<\/div>\n

\n

3<\/a> orj.artefact<\/span><\/span><\/p>\n<\/div>\n

\n

4<\/a> orj. trainer<\/span><\/span><\/p>\n<\/div>\n

\n

5<\/a> \u00c7evirenin notu: Yazar George Orwel\u2019in 1984 isimli kitab\u0131na at\u0131fta bulunuyor.<\/span><\/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-64","chapter","type-chapter","status-publish","hentry","chapter-type-numberless"],"part":46,"_links":{"self":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/64","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\/64\/revisions"}],"part":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/parts\/46"}],"metadata":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/64\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/media?parent=64"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapter-type?post=64"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/contributor?post=64"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/license?post=64"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}