Efficient Ensemble Methods for Classification on Clear Cell Renal Cell Carcinoma Clinical Dataset

Abstract

Kidneys play an important role in human body. In essence, a kidney maintains homeostasis and removes harmful materials by making and ejecting a form of urine. Especially 2–3% of humans who have malignancies, also suffered a clear cell renal cell carcinoma (ccRCC) which is one kind of kidney diseases. When diagnosed early, this renal cell carcinoma can be easily treated with some incision surgical method. Nonetheless, some patients who cannot undergo incision surgery need a customized medical service. The ensemble method is usually used to improve the classification performance by combining classifier. For this reason, in this paper, we suggest an implementation of classification algorithm on clinical data to find important clinical factors for ccRCC using an ensemble method and compare the results with a recent work in the literature. The experimental results showed that classification with ensemble methods improved the classification result, especially bagging method.