{"id":521,"date":"2023-07-29T15:29:02","date_gmt":"2023-07-29T15:29:02","guid":{"rendered":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/"},"modified":"2023-07-29T15:29:02","modified_gmt":"2023-07-29T15:29:02","slug":"rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi","status":"publish","type":"post","link":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/","title":{"rendered":"R&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r"},"content":{"rendered":"<p><\/p>\n<hr>\n<p><span style=\"color: #000000;\">\u0130statistiklerde genellikle iki nedenden dolay\u0131 modeller olu\u015ftururuz:<\/span><\/p>\n<ul>\n<li> Bir veya daha fazla yorday\u0131c\u0131 de\u011fi\u015fken ile <span style=\"color: #000000;\">bir yan\u0131t de\u011fi\u015fkeni<\/span> <span style=\"color: #000000;\">aras\u0131ndaki ili\u015fkiyi anlay\u0131n<\/span> .<\/li>\n<li> <span style=\"color: #000000;\">Gelecekteki g\u00f6zlemleri tahmin etmek i\u00e7in bir model kullan\u0131n.<\/span><\/li>\n<\/ul>\n<p> <span style=\"color: #000000;\"><strong>\u00c7apraz do\u011frulama,<\/strong> bir modelin gelecekteki g\u00f6zlemleri ne kadar iyi tahmin edebildi\u011fini tahmin etmek i\u00e7in kullan\u0131\u015fl\u0131d\u0131r.<\/span><\/p>\n<p> <span style=\"color: #000000;\">\u00d6rne\u011fin, \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fkenler olarak ya\u015f ve geliri, yan\u0131t de\u011fi\u015fkeni olarak da temerr\u00fct durumunu kullanan \u00e7oklu do\u011frusal regresyon<\/span> <span style=\"color: #000000;\"><em>modeli<\/em> <em>olu\u015fturabiliriz<\/em> <em>.<\/em><\/span> <span style=\"color: #000000;\">Bu durumda, modeli bir veri setine uydurmak isteyebiliriz ve daha sonra bu modeli,<\/span> <span style=\"color: #000000;\">yeni ba\u015fvuru sahibinin gelirine ve ya\u015f\u0131na ba\u011fl\u0131 olarak kredisini \u00f6deyememe olas\u0131l\u0131\u011f\u0131n\u0131 tahmin etmek i\u00e7in kullanabiliriz.<\/span><\/p>\n<p> Modelin g\u00fc\u00e7l\u00fc tahmin yetene\u011fine sahip olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in onu <span style=\"color: #000000;\">daha \u00f6nce hi\u00e7 g\u00f6rmedi\u011fi veriler<\/span> <span style=\"color: #000000;\">\u00fczerinde tahminlerde bulunmak \u00fczere kullanmam\u0131z gerekir<\/span> . <span style=\"color: #000000;\">Bu, modelin <strong>tahmin hatas\u0131n\u0131<\/strong> tahmin etmemizi sa\u011flayacakt\u0131r.<\/span><\/p>\n<h2> <strong><span style=\"color: #000000;\">Tahmin Hatas\u0131n\u0131 Tahmin Etmek \u0130\u00e7in \u00c7apraz Do\u011frulamay\u0131 Kullanma<\/span><\/strong><\/h2>\n<p> <span style=\"color: #000000;\"><strong>\u00c7apraz do\u011frulama,<\/strong> tahmin hatas\u0131n\u0131 tahmin edebilece\u011fimiz farkl\u0131 yollar\u0131 ifade eder.<\/span> <span style=\"color: #000000;\">\u00c7apraz do\u011frulamaya<\/span> <span style=\"color: #000000;\">genel yakla\u015f\u0131m<\/span> \u015fudur:<\/p>\n<p> <span style=\"color: #000000;\"><strong>1.<\/strong> Veri setinde belirli say\u0131da g\u00f6zlemi bir kenara koyun; genellikle t\u00fcm g\u00f6zlemlerin %15-25&#8217;i.<\/span><br \/> <span style=\"color: #000000;\"><strong>2.<\/strong> Modeli veri setinde tuttu\u011fumuz g\u00f6zlemlere g\u00f6re yerle\u015ftirin (veya &#8220;e\u011fitin&#8221;).<\/span><br \/> <span style=\"color: #000000;\"><strong>3.<\/strong> Modelin, modeli e\u011fitmek i\u00e7in kullanmad\u0131\u011f\u0131m\u0131z g\u00f6zlemler hakk\u0131nda ne kadar iyi tahminlerde bulunabildi\u011fini test edin.<\/span><\/p>\n<h2> <span style=\"color: #000000;\"><strong>Bir modelin kalitesini \u00f6l\u00e7me<\/strong><\/span><\/h2>\n<p> <span style=\"color: #000000;\">Yeni g\u00f6zlemler hakk\u0131nda tahminlerde bulunmak i\u00e7in uygun modeli kulland\u0131\u011f\u0131m\u0131zda, modelin kalitesini \u00f6l\u00e7mek i\u00e7in a\u015fa\u011f\u0131dakiler de dahil olmak \u00fczere birka\u00e7 farkl\u0131 \u00f6l\u00e7\u00fcm kullanabiliriz:<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>\u00c7oklu R-kare:<\/strong> Bu, yorday\u0131c\u0131 de\u011fi\u015fkenler ile yan\u0131t de\u011fi\u015fkeni aras\u0131ndaki do\u011frusal ili\u015fkinin g\u00fcc\u00fcn\u00fc \u00f6l\u00e7er<\/span> <span style=\"color: #000000;\">. 1&#8217;in R-kare kat\u0131 m\u00fckemmel bir do\u011frusal ili\u015fkiyi belirtirken,<\/span> <span style=\"color: #000000;\">0&#8217;\u0131n R-kare kat\u0131 do\u011frusal bir ili\u015fki olmad\u0131\u011f\u0131n\u0131 g\u00f6sterir. R-kare \u00e7arpan\u0131 ne kadar y\u00fcksek olursa, yorday\u0131c\u0131 de\u011fi\u015fkenlerin yan\u0131t de\u011fi\u015fkenini tahmin etme olas\u0131l\u0131\u011f\u0131 o kadar artar.<\/span><\/p>\n<p> K\u00f6k ortalama kare hatas\u0131 (RMSE): <span style=\"color: #000000;\">Yeni bir g\u00f6zlemin<\/span> <span style=\"color: #000000;\"><strong>de\u011ferini<\/strong> tahmin ederken modelin yapt\u0131\u011f\u0131 ortalama tahmin hatas\u0131n\u0131 \u00f6l\u00e7er<\/span> . <span style=\"color: #000000;\">Bu, bir g\u00f6zlemin ger\u00e7ek de\u011feri ile model taraf\u0131ndan tahmin edilen de\u011fer aras\u0131ndaki ortalama mesafedir.<\/span> RMSE i\u00e7in <span style=\"color: #000000;\">daha d\u00fc\u015f\u00fck<\/span> de\u011ferler <span style=\"color: #000000;\">daha iyi model uyumunu g\u00f6sterir.<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>Ortalama Mutlak Hata (MAE):<\/strong> Bir g\u00f6zlemin ger\u00e7ek de\u011feri ile model taraf\u0131ndan tahmin edilen de\u011fer aras\u0131ndaki ortalama mutlak farkt\u0131r.<\/span> <span style=\"color: #000000;\">Bu metrik genellikle ayk\u0131r\u0131 de\u011ferlere kar\u015f\u0131 RMSE&#8217;ye g\u00f6re daha az duyarl\u0131d\u0131r. MAE i\u00e7in daha d\u00fc\u015f\u00fck de\u011ferler daha iyi model uyumunu g\u00f6sterir.<\/span><\/p>\n<h2> <span style=\"color: #000000;\"><strong>R&#8217;de d\u00f6rt farkl\u0131 \u00e7apraz do\u011frulama tekni\u011finin uygulanmas\u0131<\/strong><\/span><\/h2>\n<p> <span style=\"color: #000000;\">Daha sonra a\u015fa\u011f\u0131daki \u00e7apraz do\u011frulama tekniklerinin R&#8217;de nas\u0131l uygulanaca\u011f\u0131n\u0131 a\u00e7\u0131klayaca\u011f\u0131z:<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>1.<\/strong> Do\u011frulama seti yakla\u015f\u0131m\u0131<\/span><br \/> <span style=\"color: #000000;\"><strong>2.<\/strong> k-katl\u0131 \u00e7apraz do\u011frulama<\/span><br \/> <span style=\"color: #000000;\"><strong>3.<\/strong> \u00c7apraz do\u011frulamay\u0131 bir kenara b\u0131rak\u0131n<\/span><br \/> <span style=\"color: #000000;\"><strong>4.<\/strong> Tekrarlanan k-katl\u0131 \u00e7apraz do\u011frulama<\/span><\/p>\n<p> <span style=\"color: #000000;\">Bu farkl\u0131 tekniklerin nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6stermek i\u00e7in <em>mtcars&#8217;\u0131n<\/em> yerle\u015fik R veri k\u00fcmesinin bir alt k\u00fcmesini kullanaca\u011f\u0131z:<\/span><\/p>\n<pre style=\"background-color: #e5e5e5; font-size: 15px;\"> <strong><span style=\"color: #008080;\">#define dataset\n<\/span>data &lt;- mtcars[, c(\"mpg\", \"disp\", \"hp\", \"drat\")]\n\n<span style=\"color: #008080;\">#view first six rows of new data\n<\/span>head(data)\n\n# mpg disp hp drat\n#Mazda RX4 21.0 160 110 3.90\n#Mazda RX4 Wag 21.0 160 110 3.90\n#Datsun 710 22.8 108 93 3.85\n#Hornet 4 Drive 21.4 258 110 3.08\n#Hornet Sportabout 18.7 360 175 3.15\n#Valiant 18.1 225 105 2.76\n<\/strong><\/pre>\n<p> Tahmin de\u011fi\u015fkenleri olarak disp, hp ve drat&#8217;\u0131 ve <span style=\"color: #000000;\">yan\u0131t de\u011fi\u015fkeni olarak<\/span> <span style=\"color: #000000;\"><em>mpg&#8217;yi<\/em> <em>kullanarak<\/em> <em>\u00e7oklu<\/em> <em>do\u011frusal<\/em> regresyon modeli olu\u015fturaca\u011f\u0131z<\/span> .<\/p>\n<h2> <strong><span style=\"color: #000000;\">Do\u011frulama seti yakla\u015f\u0131m\u0131<\/span><\/strong><\/h2>\n<p> <span style=\"color: #000000;\"><strong>Do\u011frulama seti yakla\u015f\u0131m\u0131<\/strong> \u015fu \u015fekilde \u00e7al\u0131\u015f\u0131r:<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>1.<\/strong> Verileri iki gruba b\u00f6l\u00fcn: bir grup modeli e\u011fitmek i\u00e7in kullan\u0131l\u0131r (yani model parametrelerini tahmin etmek),<\/span> <span style=\"color: #000000;\">di\u011fer grup ise modeli test etmek i\u00e7in kullan\u0131l\u0131r.<\/span> Genellikle <span style=\"color: #000000;\">verilerin %70-80&#8217;i<\/span> <span style=\"color: #000000;\">rastgele se\u00e7ilerek e\u011fitim seti olu\u015fturulur<\/span> ve geri kalan %20-30&#8217;luk k\u0131s\u0131m test seti olarak kullan\u0131l\u0131r.<\/p>\n<p> <span style=\"color: #000000;\"><strong>2.<\/strong> E\u011fitim veri k\u00fcmesini kullanarak modeli olu\u015fturun.<\/span><br \/> <span style=\"color: #000000;\"><strong>3.<\/strong> Test seti verileri hakk\u0131nda tahminlerde bulunmak i\u00e7in modeli kullan\u0131n.<\/span><br \/> <span style=\"color: #000000;\"><strong>4.<\/strong> R-kare, RMSE ve MAE gibi \u00f6l\u00e7\u00fcmleri kullanarak model kalitesini \u00f6l\u00e7\u00fcn.<\/span><\/p>\n<h3> <strong><span style=\"color: #000000;\">\u00d6rnek:<\/span><\/strong><\/h3>\n<p> <span style=\"color: #000000;\">A\u015fa\u011f\u0131daki \u00f6rnek yukar\u0131da tan\u0131mlad\u0131\u011f\u0131m\u0131z veri k\u00fcmesini kullanmaktad\u0131r. \u0130lk \u00f6nce verileri par\u00e7alara ay\u0131r\u0131yoruz.<\/span><br \/> <span style=\"color: #000000;\">Verilerin %80&#8217;ini e\u011fitim seti olarak ve verilerin geri kalan %20&#8217;sini test seti<\/span> <span style=\"color: #000000;\">olarak kullanan bir e\u011fitim seti ve bir test seti. Daha sonra e\u011fitim setini kullanarak modeli olu\u015fturuyoruz<\/span> <span style=\"color: #000000;\">. Daha sonra modeli test seti hakk\u0131nda tahminler yapmak i\u00e7in kullan\u0131r\u0131z. Son olarak modelin kalitesini<\/span> <span style=\"color: #000000;\">R-squared, RMSE ve MAE kullanarak \u00f6l\u00e7\u00fcyoruz.<\/span><\/p>\n<pre style=\"background-color: #e5e5e5; font-size: 15px;\"> <strong><span style=\"color: #008080;\">#load <em>dplyr<\/em> library used for data manipulation\n<\/span>library(dplyr)\n\n<span style=\"color: #008080;\">#load <em>caret<\/em> library used for partitioning data into training and test set\n<\/span>library(caret)\n\n<span style=\"color: #008080;\">#make this example reproducible\n<\/span>set.seed(0)\n\n<span style=\"color: #008080;\">#define the dataset\n<\/span>data &lt;- mtcars[, c(\"mpg\", \"disp\", \"hp\", \"drat\")]\n\n<span style=\"color: #008080;\">#split the dataset into a training set (80%) and test set (20%).\n<\/span>training_obs &lt;- data$mpg %&gt;% createDataPartition(p = 0.8, list = FALSE)\n\ntrain &lt;- data[training_obs, ]\ntest &lt;- data[-training_obs, ]\n\n<span style=\"color: #008080;\"># Build the linear regression model on the training set\n<\/span>model &lt;- lm(mpg ~ ., data = train)\n\n<span style=\"color: #008080;\"># Use the model to make predictions on the test set\n<\/span>predictions &lt;- model %&gt;% predict(test)\n\n<span style=\"color: #008080;\">#Examine R-squared, RMSE, and MAE of predictions\n<\/span>data.frame(R_squared = R2(predictions, test$mpg),\n           RMSE = RMSE(predictions, test$mpg),\n           MAE = MAE(predictions, test$mpg))\n\n#R_squared RMSE MAE\n#1 0.9213066 1.876038 1.66614\n<\/strong><\/pre>\n<p> <span style=\"color: #000000;\">Farkl\u0131 modelleri kar\u015f\u0131la\u015ft\u0131r\u0131rken test setinde en d\u00fc\u015f\u00fck RMSE&#8217;yi \u00fcreten model tercih edilir.<\/span><\/p>\n<h3> <span style=\"color: #000000;\"><strong>Bu yakla\u015f\u0131m\u0131n avantajlar\u0131 ve dezavantajlar\u0131<\/strong><\/span><\/h3>\n<p> <span style=\"color: #000000;\">Do\u011frulama seti yakla\u015f\u0131m\u0131n\u0131n avantaj\u0131 basit ve hesaplama a\u00e7\u0131s\u0131ndan verimli olmas\u0131d\u0131r. Dezavantaj\u0131<\/span> <span style=\"color: #000000;\">ise modelin toplam verinin yaln\u0131zca bir k\u0131sm\u0131 kullan\u0131larak olu\u015fturulmu\u015f olmas\u0131d\u0131r.<\/span> E\u011fitim seti d\u0131\u015f\u0131nda <span style=\"color: #000000;\">b\u0131rakt\u0131\u011f\u0131m\u0131z veriler<\/span> <span style=\"color: #000000;\">ilgin\u00e7 veya de\u011ferli bilgiler i\u00e7eriyorsa model bunu dikkate almayacakt\u0131r.<\/span><\/p>\n<h2> <span style=\"color: #000000;\"><strong>k-katl\u0131 \u00e7apraz do\u011frulama yakla\u015f\u0131m\u0131<\/strong><\/span><\/h2>\n<p> <span style=\"color: #000000;\"><strong>K-katl\u0131 \u00e7apraz do\u011frulama yakla\u015f\u0131m\u0131<\/strong> \u015fu \u015fekilde \u00e7al\u0131\u015f\u0131r:<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>1.<\/strong> Verileri rastgele olarak k &#8220;katlamaya&#8221; veya alt k\u00fcmeye (\u00f6rne\u011fin, 5 veya 10 alt k\u00fcmeye) b\u00f6l\u00fcn.<\/span><br \/> <span style=\"color: #000000;\"><strong>2.<\/strong> Modeli yaln\u0131zca bir alt k\u00fcmeyi d\u0131\u015far\u0131da b\u0131rakarak t\u00fcm veriler \u00fczerinde e\u011fitin.<\/span><br \/> <span style=\"color: #000000;\"><strong>3.<\/strong> D\u0131\u015flanan alt k\u00fcmedeki veriler hakk\u0131nda tahminlerde bulunmak i\u00e7in modeli kullan\u0131n.<\/span><br \/> <span style=\"color: #000000;\"><strong>4.<\/strong> k alt k\u00fcmenin her biri test k\u00fcmesi olarak kullan\u0131l\u0131ncaya kadar bu i\u015flemi tekrarlay\u0131n.<\/span><br \/> <span style=\"color: #000000;\"><strong>5<\/strong> . K test hatas\u0131n\u0131n ortalamas\u0131n\u0131 alarak modelin kalitesini \u00f6l\u00e7\u00fcn. Bu biliniyor<\/span><br \/> <span style=\"color: #000000;\">\u00e7apraz do\u011frulama hatas\u0131 olarak.<\/span><\/p>\n<h3> <strong><span style=\"color: #000000;\">\u00d6rnek<\/span><\/strong><\/h3>\n<p> <span style=\"color: #000000;\">Bu \u00f6rnekte \u00f6ncelikle verileri<\/span> <span style=\"color: #000000;\">5 alt k\u00fcmeye ay\u0131r\u0131yoruz. Daha sonra, verilerin bir alt k\u00fcmesi d\u0131\u015f\u0131ndaki t\u00fcm\u00fcn\u00fc kullanarak modeli uyduruyoruz.<\/span> Daha sonra modeli <span style=\"color: #000000;\">, d\u0131\u015far\u0131da b\u0131rak\u0131lan alt k\u00fcme hakk\u0131nda tahminler<\/span> <span style=\"color: #000000;\">yapmak i\u00e7in kullan\u0131r\u0131z<\/span> ve test hatas\u0131n\u0131 kaydederiz (R-kare, RMSE ve MAE kullanarak). <span style=\"color: #000000;\">Her alt k\u00fcme test k\u00fcmesi olarak kullan\u0131l\u0131ncaya kadar bu i\u015flemi tekrarl\u0131yoruz<\/span> <span style=\"color: #000000;\">.<\/span> <span style=\"color: #000000;\">Daha sonra basit\u00e7e 5 test hatas\u0131n\u0131n ortalamas\u0131n\u0131 hesapl\u0131yoruz<\/span> <span style=\"color: #000000;\">.<\/span><\/p>\n<pre style=\"background-color: #e5e5e5; font-size: 15px;\"> <strong><span style=\"color: #008080;\">#load <em>dplyr<\/em> library used for data manipulation\n<\/span>library(dplyr)\n\n<span style=\"color: #008080;\">#load <em>caret<\/em> library used for partitioning data into training and test set\n<\/span>library(caret)\n\n<span style=\"color: #008080;\">#make this example reproducible\n<\/span>set.seed(0)\n\n<span style=\"color: #008080;\">#define the dataset\n<\/span>data &lt;- mtcars[, c(\"mpg\", \"disp\", \"hp\", \"drat\")]\n\n<span style=\"color: #008080;\">#define the number of subsets (or \"folds\") to use\n<\/span>train_control &lt;- trainControl(method = \"cv\", number = 5)\n\n<span style=\"color: #008080;\">#train the model\n<\/span>model &lt;- train(mpg ~ ., data = data, method = \"lm\", trControl = train_control)\n\n<span style=\"color: #008080;\">#Summarize the results\n<\/span>print(model)\n\n#Linear Regression \n#\n#32 samples\n#3 predictor\n#\n#No pre-processing\n#Resampling: Cross-Validated (5 fold) \n#Summary of sample sizes: 26, 25, 26, 25, 26 \n#Resampling results:\n#\n# RMSE Rsquared MAE     \n#3.095501 0.7661981 2.467427\n#\n#Tuning parameter 'intercept' was held constant at a value of TRUE\n<\/strong><\/pre>\n<h3> <strong><span style=\"color: #000000;\">Bu yakla\u015f\u0131m\u0131n avantajlar\u0131 ve dezavantajlar\u0131<\/span><\/strong><\/h3>\n<p> <span style=\"color: #000000;\">K-katl\u0131 \u00e7apraz do\u011frulama yakla\u015f\u0131m\u0131n\u0131n do\u011frulama seti yakla\u015f\u0131m\u0131na g\u00f6re avantaj\u0131, modeli her seferinde farkl\u0131 veri par\u00e7alar\u0131 kullanarak birka\u00e7 farkl\u0131 kez olu\u015fturmas\u0131d\u0131r<\/span> , b\u00f6ylece <span style=\"color: #000000;\">modeli<\/span> <span style=\"color: #000000;\">olu\u015ftururken \u00f6nemli verileri atlama \u015fans\u0131m\u0131z olmaz<\/span> .<\/p>\n<p> Bu yakla\u015f\u0131m\u0131n \u00f6znel k\u0131sm\u0131, k i\u00e7in kullan\u0131lacak de\u011ferin, yani <span style=\"color: #000000;\">verinin<\/span> <span style=\"color: #000000;\">b\u00f6l\u00fcnece\u011fi alt k\u00fcmelerin say\u0131s\u0131n\u0131n se\u00e7ilmesidir<\/span> . <span style=\"color: #000000;\">Genel olarak, daha d\u00fc\u015f\u00fck k de\u011ferleri daha y\u00fcksek yanl\u0131l\u0131\u011fa ancak daha d\u00fc\u015f\u00fck de\u011fi\u015fkenli\u011fe yol a\u00e7arken, daha y\u00fcksek k de\u011ferleri<\/span> <span style=\"color: #000000;\">daha d\u00fc\u015f\u00fck yanl\u0131l\u0131\u011fa ancak daha y\u00fcksek de\u011fi\u015fkenli\u011fe yol a\u00e7ar.<\/span><\/p>\n<p> <span style=\"color: #000000;\">Uygulamada k genellikle 5 veya 10&#8217;a e\u015fit olacak \u015fekilde se\u00e7ilir, \u00e7\u00fcnk\u00fc bu say\u0131da<\/span> <span style=\"color: #000000;\">alt k\u00fcme ayn\u0131 anda \u00e7ok fazla \u00f6nyarg\u0131y\u0131 ve \u00e7ok fazla de\u011fi\u015fkenli\u011fi \u00f6nleme e\u011filimindedir.<\/span><\/p>\n<h2> <span style=\"color: #000000;\"><strong>Birini D\u0131\u015far\u0131da B\u0131rak \u00c7apraz Do\u011frulama (LOOCV) yakla\u015f\u0131m\u0131<\/strong><\/span><\/h2>\n<p> <span style=\"color: #000000;\"><strong>LOOCV yakla\u015f\u0131m\u0131<\/strong> \u015fu \u015fekilde \u00e7al\u0131\u015f\u0131r:<\/span><\/p>\n<p> <span style=\"color: #000000;\"><strong>1.<\/strong> Veri k\u00fcmesindeki g\u00f6zlemlerden biri hari\u00e7 t\u00fcm\u00fcn\u00fc kullanarak modeli olu\u015fturun.<\/span><br \/> <span style=\"color: #000000;\"><strong>2.<\/strong> Eksik g\u00f6zlemin de\u011ferini tahmin etmek i\u00e7in modeli kullan\u0131n. Bu tahmini test etme hatas\u0131n\u0131 kaydedin.<\/span><br \/> <span style=\"color: #000000;\"><strong>3.<\/strong> Veri setindeki her g\u00f6zlem i\u00e7in bu i\u015flemi tekrarlay\u0131n.<\/span><br \/> <span style=\"color: #000000;\"><strong>4.<\/strong> T\u00fcm tahmin hatalar\u0131n\u0131n ortalamas\u0131n\u0131 alarak modelin kalitesini \u00f6l\u00e7\u00fcn.<\/span><\/p>\n<h3> <strong><span style=\"color: #000000;\">\u00d6rnek<\/span><\/strong><\/h3>\n<p> <span style=\"color: #000000;\">A\u015fa\u011f\u0131daki \u00f6rnek, \u00f6nceki \u00f6rneklerde kullan\u0131lan ayn\u0131 veri k\u00fcmesi i\u00e7in LOOCV ger\u00e7ekle\u015ftirmenin nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n<pre style=\"background-color: #e5e5e5; font-size: 15px;\"> <strong><span style=\"color: #008080;\">#load <em>dplyr<\/em> library used for data manipulation\n<\/span>library(dplyr)\n\n<span style=\"color: #008080;\">#load <em>caret<\/em> library used for partitioning data into training and test set\n<\/span>library(caret)\n\n<span style=\"color: #008080;\">#make this example reproducible\n<\/span>set.seed(0)\n\n<span style=\"color: #008080;\">#define the dataset\n<\/span>data &lt;- mtcars[, c(\"mpg\", \"disp\", \"hp\", \"drat\")]\n\n<span style=\"color: #008080;\">#specify that we want to use LOOCV\n<\/span>train_control &lt;- trainControl( <span style=\"color: #800080;\">method = \"LOOCV\"<\/span> )\n\n<span style=\"color: #008080;\">#train the model\n<\/span>model &lt;- train(mpg ~ ., data = data, method = \"lm\", trControl = train_control)\n\n<span style=\"color: #008080;\">#summarize the results\n<\/span>print(model)\n\n#Linear Regression \n#\n#32 samples\n#3 predictor\n#\n#No pre-processing\n#Resampling: Leave-One-Out Cross-Validation \n#Summary of sample sizes: 31, 31, 31, 31, 31, 31, ... \n#Resampling results:\n#\n# RMSE Rsquared MAE     \n#3.168763 0.7170704 2.503544\n#\n#Tuning parameter 'intercept' was held constant at a value of TRUE\n<\/strong><\/pre>\n<h3> <span style=\"color: #000000;\"><strong>Bu yakla\u015f\u0131m\u0131n avantajlar\u0131 ve dezavantajlar\u0131<\/strong><\/span><\/h3>\n<p> <span style=\"color: #000000;\">LOOCV&#8217;nin avantaj\u0131 t\u00fcm veri noktalar\u0131n\u0131 kullanmam\u0131zd\u0131r, bu da genel olarak potansiyel yanl\u0131l\u0131\u011f\u0131 azalt\u0131r. Ancak<\/span> <span style=\"color: #000000;\">modeli her bir g\u00f6zlemin de\u011ferini tahmin etmek i\u00e7in kulland\u0131\u011f\u0131m\u0131z i\u00e7in bu,<\/span> <span style=\"color: #000000;\">tahmin hatas\u0131nda daha fazla de\u011fi\u015fkenli\u011fe yol a\u00e7abilir.<\/span><\/p>\n<p> <span style=\"color: #000000;\">Bu yakla\u015f\u0131m\u0131n di\u011fer bir dezavantaj\u0131, verimsiz ve hesaplama a\u00e7\u0131s\u0131ndan a\u011f\u0131r hale gelebilecek kadar \u00e7ok say\u0131da modele uymas\u0131 gerekti\u011fidir.<\/span><\/p>\n<h2> <strong><span style=\"color: #000000;\">Tekrarlanan k-katl\u0131 \u00e7apraz do\u011frulama yakla\u015f\u0131m\u0131<\/span><\/strong><\/h2>\n<p> <span style=\"color: #000000;\">Basit\u00e7e <strong>k-katl\u0131 \u00e7apraz do\u011frulamay\u0131 birden \u00e7ok kez ger\u00e7ekle\u015ftirerek tekrarlanan<\/strong> k-katl\u0131 \u00e7apraz do\u011frulamay\u0131 ger\u00e7ekle\u015ftirebiliriz. Son hata, tekrar say\u0131s\u0131n\u0131n ortalama hatas\u0131d\u0131r<\/span> <span style=\"color: #000000;\">.<\/span><\/p>\n<p> <span style=\"color: #000000;\">A\u015fa\u011f\u0131daki \u00f6rnek, 4 kez tekrarlanan 5 katl\u0131 bir \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirir:<\/span><\/p>\n<pre style=\"background-color: #e5e5e5; font-size: 15px;\"> <strong><span style=\"color: #008080;\">#load <em>dplyr<\/em> library used for data manipulation\n<\/span>library(dplyr)\n\n<span style=\"color: #008080;\">#load <em>caret<\/em> library used for partitioning data into training and test set\n<\/span>library(caret)\n\n<span style=\"color: #008080;\">#make this example reproducible\n<\/span>set.seed(0)\n\n<span style=\"color: #008080;\">#define the dataset\n<\/span>data &lt;- mtcars[, c(\"mpg\", \"disp\", \"hp\", \"drat\")]\n\n<span style=\"color: #008080;\">#define the number of subsets to use and number of times to repeat k-fold CV\n<\/span>train_control &lt;- trainControl(method = \"repeatedcv\", number = 5, <span style=\"color: #800080;\">repeats = 4<\/span> )\n\n<span style=\"color: #008080;\">#train the model\n<\/span>model &lt;- train(mpg ~ ., data = data, method = \"lm\", trControl = train_control)\n\n<span style=\"color: #008080;\">#summarize the results\n<\/span>print(model)\n\n#Linear Regression \n#\n#32 samples\n#3 predictor\n#\n#No pre-processing\n#Resampling: Cross-Validated (5 fold, repeated 4 times) \n#Summary of sample sizes: 26, 25, 26, 25, 26, 25, ... \n#Resampling results:\n#\n# RMSE Rsquared MAE     \n#3.176339 0.7909337 2.559131\n#\n#Tuning parameter 'intercept' was held constant at a value of TRUE\n<\/strong><\/pre>\n<h3> <span style=\"color: #000000;\"><strong>Bu yakla\u015f\u0131m\u0131n avantajlar\u0131 ve dezavantajlar\u0131<\/strong><\/span><\/h3>\n<p> <span style=\"color: #000000;\">Tekrarlanan k-katl\u0131 \u00e7apraz do\u011frulama yakla\u015f\u0131m\u0131n\u0131n avantaj\u0131, her tekrar i\u00e7in verilerin biraz farkl\u0131 alt k\u00fcmelere b\u00f6l\u00fcnmesidir; bu da modelin tahmin hatas\u0131na ili\u015fkin daha tarafs\u0131z bir tahmin sunmal\u0131d\u0131r. Bu yakla\u015f\u0131m\u0131n dezavantaj\u0131, model uydurma s\u00fcrecini birka\u00e7 kez tekrarlamam\u0131z gerekti\u011finden hesaplama a\u00e7\u0131s\u0131ndan yo\u011fun olabilmesidir.<\/span><\/p>\n<h2> <strong><span style=\"color: #000000;\">\u00c7apraz do\u011frulamada katlama say\u0131s\u0131 nas\u0131l se\u00e7ilir<\/span><\/strong><\/h2>\n<p> <span style=\"color: #000000;\">\u00c7apraz do\u011frulaman\u0131n en \u00f6znel k\u0131sm\u0131 ka\u00e7 katlaman\u0131n (yani alt k\u00fcmelerin) kullan\u0131laca\u011f\u0131na karar vermektir. Genel olarak, katlama say\u0131s\u0131 ne kadar k\u00fc\u00e7\u00fckse, hata tahminleri o kadar tarafl\u0131 olur ancak de\u011fi\u015fkenlikleri de o kadar az olur. Tersine, katlama say\u0131s\u0131 ne kadar y\u00fcksek olursa, hata tahminleri o kadar az tarafl\u0131 olur, ancak tahminler o kadar de\u011fi\u015fken olur.<\/span><\/p>\n<p> <span style=\"color: #000000;\">Hesaplama s\u00fcresini de ak\u0131lda tutmak \u00f6nemlidir. Her katlama i\u00e7in yeni bir desen e\u011fitmeniz gerekecektir ve bu yava\u015f bir s\u00fcre\u00e7 olsa da, \u00e7ok say\u0131da katlama se\u00e7erseniz uzun zaman alabilir.<\/span><\/p>\n<p> <span style=\"color: #000000;\">Pratikte \u00e7apraz do\u011frulama genellikle 5 veya 10 katla ger\u00e7ekle\u015ftirilir; \u00e7\u00fcnk\u00fc bu, de\u011fi\u015fkenlik ve \u00f6nyarg\u0131 aras\u0131nda iyi bir denge sa\u011flarken ayn\u0131 zamanda hesaplama a\u00e7\u0131s\u0131ndan da verimlidir.<\/span><\/p>\n<h2> <strong>\u00c7apraz do\u011frulama ger\u00e7ekle\u015ftirildikten sonra model nas\u0131l se\u00e7ilir?<\/strong><\/h2>\n<p> <span style=\"color: #000000;\">\u00c7apraz do\u011frulama, bir modelin tahmin hatas\u0131n\u0131 de\u011ferlendirmek i\u00e7in kullan\u0131l\u0131r. Bu, hangi modelin en d\u00fc\u015f\u00fck tahmin hatas\u0131na sahip oldu\u011funu (RMSE, R-kare vb. temel al\u0131narak) vurgulayarak iki veya daha fazla farkl\u0131 model aras\u0131nda se\u00e7im yapmam\u0131za yard\u0131mc\u0131 olabilir.<\/span><\/p>\n<p> <span style=\"color: #000000;\">En iyi modeli se\u00e7mek i\u00e7in \u00e7apraz do\u011frulamay\u0131 kulland\u0131ktan sonra, se\u00e7ilen modele uyacak \u015fekilde mevcut <em>t\u00fcm<\/em> verileri kullan\u0131r\u0131z. Son modelimiz i\u00e7in \u00e7apraz do\u011frulama s\u0131ras\u0131nda e\u011fitti\u011fimiz ger\u00e7ek model \u00f6rneklerini kullanm\u0131yoruz.<\/span><\/p>\n<p> <span style=\"color: #000000;\">\u00d6rne\u011fin, iki farkl\u0131 regresyon modeli aras\u0131nda hangi modelin kullan\u0131lmas\u0131n\u0131n daha iyi oldu\u011funu belirlemek i\u00e7in 5 katl\u0131 \u00e7apraz do\u011frulamay\u0131 kullanabiliriz. Ancak hangi modelin kullan\u0131lmas\u0131n\u0131n en iyi oldu\u011funu belirledikten sonra <em>t\u00fcm<\/em> verileri nihai modele uyacak \u015fekilde kullan\u0131r\u0131z. Yani son modeli olu\u015ftururken hi\u00e7bir k\u0131vr\u0131m\u0131 unutmuyoruz.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u0130statistiklerde genellikle iki nedenden dolay\u0131 modeller olu\u015ftururuz: Bir veya daha fazla yorday\u0131c\u0131 de\u011fi\u015fken ile bir yan\u0131t de\u011fi\u015fkeni aras\u0131ndaki ili\u015fkiyi anlay\u0131n . Gelecekteki g\u00f6zlemleri tahmin etmek i\u00e7in bir model kullan\u0131n. \u00c7apraz do\u011frulama, bir modelin gelecekteki g\u00f6zlemleri ne kadar iyi tahmin edebildi\u011fini tahmin etmek i\u00e7in kullan\u0131\u015fl\u0131d\u0131r. \u00d6rne\u011fin, \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fkenler olarak ya\u015f ve geliri, yan\u0131t de\u011fi\u015fkeni olarak da [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[],"class_list":["post-521","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v21.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>R - Statorials&#039;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r<\/title>\n<meta name=\"description\" content=\"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#039;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"R - Statorials&#039;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r\" \/>\n<meta property=\"og:description\" content=\"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#039;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/\" \/>\n<meta property=\"og:site_name\" content=\"Statorials\" \/>\n<meta property=\"article:published_time\" content=\"2023-07-29T15:29:02+00:00\" \/>\n<meta name=\"author\" content=\"Dr.benjamin anderson\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Dr.benjamin anderson\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"12 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/\",\"url\":\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/\",\"name\":\"R - Statorials&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r\",\"isPartOf\":{\"@id\":\"https:\/\/statorials.org\/tr\/#website\"},\"datePublished\":\"2023-07-29T15:29:02+00:00\",\"dateModified\":\"2023-07-29T15:29:02+00:00\",\"author\":{\"@id\":\"https:\/\/statorials.org\/tr\/#\/schema\/person\/365dc158a39a7c8ae256355451e3de48\"},\"description\":\"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#39;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.\",\"breadcrumb\":{\"@id\":\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Ev\",\"item\":\"https:\/\/statorials.org\/tr\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"R&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/statorials.org\/tr\/#website\",\"url\":\"https:\/\/statorials.org\/tr\/\",\"name\":\"Statorials\",\"description\":\"\u0130statistik okuryazarl\u0131\u011f\u0131 rehberiniz!\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/statorials.org\/tr\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"tr\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/statorials.org\/tr\/#\/schema\/person\/365dc158a39a7c8ae256355451e3de48\",\"name\":\"Dr.benjamin anderson\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/statorials.org\/tr\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/statorials.org\/tr\/wp-content\/uploads\/2023\/10\/Dr.-Benjamin-Anderson-96x96.jpg\",\"contentUrl\":\"https:\/\/statorials.org\/tr\/wp-content\/uploads\/2023\/10\/Dr.-Benjamin-Anderson-96x96.jpg\",\"caption\":\"Dr.benjamin anderson\"},\"description\":\"Merhaba, ben Benjamin, emekli bir istatistik profes\u00f6r\u00fc ve Statorials \u00f6\u011fretmenine d\u00f6n\u00fc\u015ft\u00fcm. \u0130statistik alan\u0131ndaki kapsaml\u0131 deneyimim ve uzmanl\u0131\u011f\u0131mla, \u00f6\u011frencilerimi Statorials arac\u0131l\u0131\u011f\u0131yla g\u00fc\u00e7lendirmek i\u00e7in bilgilerimi payla\u015fmaya can at\u0131yorum. Daha fazlas\u0131n\u0131 bil\",\"sameAs\":[\"https:\/\/statorials.org\/tr\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"R - Statorials&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r","description":"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#39;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/","og_locale":"tr_TR","og_type":"article","og_title":"R - Statorials&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r","og_description":"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#39;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.","og_url":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/","og_site_name":"Statorials","article_published_time":"2023-07-29T15:29:02+00:00","author":"Dr.benjamin anderson","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Dr.benjamin anderson","Tahmini okuma s\u00fcresi":"12 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/","url":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/","name":"R - Statorials&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r","isPartOf":{"@id":"https:\/\/statorials.org\/tr\/#website"},"datePublished":"2023-07-29T15:29:02+00:00","dateModified":"2023-07-29T15:29:02+00:00","author":{"@id":"https:\/\/statorials.org\/tr\/#\/schema\/person\/365dc158a39a7c8ae256355451e3de48"},"description":"Bu e\u011fitimde, model performans\u0131n\u0131 de\u011ferlendirmek i\u00e7in R&#39;de \u00e7apraz do\u011frulama ger\u00e7ekle\u015ftirmenin d\u00f6rt farkl\u0131 yolu a\u00e7\u0131klanmaktad\u0131r.","breadcrumb":{"@id":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/statorials.org\/tr\/rde-model-performansi-icin-capraz-dogrulamanin-nasil-yapilacagi\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Ev","item":"https:\/\/statorials.org\/tr\/"},{"@type":"ListItem","position":2,"name":"R&#39;de model performans\u0131 i\u00e7in \u00e7apraz do\u011frulama nas\u0131l yap\u0131l\u0131r"}]},{"@type":"WebSite","@id":"https:\/\/statorials.org\/tr\/#website","url":"https:\/\/statorials.org\/tr\/","name":"Statorials","description":"\u0130statistik okuryazarl\u0131\u011f\u0131 rehberiniz!","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/statorials.org\/tr\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"tr"},{"@type":"Person","@id":"https:\/\/statorials.org\/tr\/#\/schema\/person\/365dc158a39a7c8ae256355451e3de48","name":"Dr.benjamin anderson","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/statorials.org\/tr\/#\/schema\/person\/image\/","url":"https:\/\/statorials.org\/tr\/wp-content\/uploads\/2023\/10\/Dr.-Benjamin-Anderson-96x96.jpg","contentUrl":"https:\/\/statorials.org\/tr\/wp-content\/uploads\/2023\/10\/Dr.-Benjamin-Anderson-96x96.jpg","caption":"Dr.benjamin anderson"},"description":"Merhaba, ben Benjamin, emekli bir istatistik profes\u00f6r\u00fc ve Statorials \u00f6\u011fretmenine d\u00f6n\u00fc\u015ft\u00fcm. \u0130statistik alan\u0131ndaki kapsaml\u0131 deneyimim ve uzmanl\u0131\u011f\u0131mla, \u00f6\u011frencilerimi Statorials arac\u0131l\u0131\u011f\u0131yla g\u00fc\u00e7lendirmek i\u00e7in bilgilerimi payla\u015fmaya can at\u0131yorum. Daha fazlas\u0131n\u0131 bil","sameAs":["https:\/\/statorials.org\/tr"]}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"_links":{"self":[{"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/posts\/521","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/comments?post=521"}],"version-history":[{"count":0,"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/posts\/521\/revisions"}],"wp:attachment":[{"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/media?parent=521"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/categories?post=521"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/statorials.org\/tr\/wp-json\/wp\/v2\/tags?post=521"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}