A Comparative Analysis of Classification Performance Between Balanced and Imbalanced Rodent Tuber Datasets Using Support Vector Machine and Random Forest

The main discussion in this paper is focused on how the condition of a dataset (imbalanced or balanced) can affect classification results. The dataset used in this study is Rodent Tuber's imbalanced dataset, which contain chemical information that can be used for cancer identification. This dataset will be classified using Support Vector Machine with 4 kernels and Random Forest. Classification is performed on both an imbalanced and balanced dataset by Cluster Centroid Undersampling to observe the differences. The results show that a balanced dataset yields better outcomes with minimal running time. The best results from classification on the balanced dataset were obtained using the Random Forest algorithm.