Karar Ağaçları Kullanılarak Klinik Verilerle Covid-19 Enfeksiyonunun İncelenmesi

The coronavirus disease, namely Covid-19 infection, which was declared a worldwide epidemic by the World Health Organization (WHO) in 2020, first seen Wuhan, China last months of 2019 and has affected whole world. Early diagnosis this rapidly spreading is important to prevent disease. For reason, methods such as image processing, deep learning, machine learning have become detect early. In study, it been tried classify individuals who test positive negative for based on some laboratory results with several Decision Tree methods. Since original form data set an uneven distribution, balanced applying oversampling undersampling used sets pre-processing study. Balanced dataset using 5-Fold Cross Validation (CV), 10-Fold Leave-One-Out (LOO)-CV, Random Forest (RF), (RT), J48, ıt analyzed alternating decision tree (ADTree) Function Trees (FT) classifiers. As result examination, most successful shown RF classifier 87.5% success rates CV-5 set, 93.3% CV-10 LOO-CV method, 79% method. addition rates, sensitivity-specificity metrics, are patient healthy diagnosis, were examined terms each classification algorithm CV value.