Comparison of classification models with statistical methods and machine learning techniques in healthcare datasets

Data analysis could be an expressive guide by analyzing and influencing decision-making processes in various fields as engineering, medicine, economics etc. For the purpose of obtaining objective and reliable results, the usage of appropriate methods contributes to the research processes. The aim of this study is for making statistical inferences with the clinical datasets of patients with cancer, COPD and heart failure. Firstly, logistic regression and regularization methods such as ridge regression, lasso regression and elastic-net regression were tried to analyze with diverse datasets. CART algorithm and C5.0 algorithm, which are one of the most frequently used methods due to their visuality and operational ease, were also included to this study. In the last step of the study, the results of analyses which performed in RStudio were compared along with various performance measures and inferences.