Comparing Logistic Regression and Support Vector Machine in Breast Cancer Problem

There are several methods used for the classification problems. many different kinds of fields that can be used. Nowadays, Support Vector Machine (SVM) is a popular method has been proposed by researchers. Using same but distribution creating training and testing data in dataset yield varying results terms prediction accuracy, which crucial classification. In this paper, we compare accuracy between SVM Logistic Regression to determine better classify current condition patient after undergoing some treatment. Several treatments including feature selection, extraction, separating train using Holdout K-Fold CV. Stepwise selection done reduce features. Training obtained five stratified non-stratified holdout fold cross validation. The result shows best cancer validation with radial kernel. 81,816% variance as much 0,94%.