An Efficiency Random Forest Algorithm for Classification of Patients with Kidney Dysfunction

In this paper, we presented a separate separation and comparison of data of people with renal impairment. By collecting information on CKD. The data was collected for selection in data mining using the CKD data set from UCI Machine Learn Repository to compare the classification of 400 CKD patients, comprising 25 attributes and dividing into two class, which one is for patients with CKD and those who do not suffer from CKD. In the experimental designing with 5-folds cross validation test, the result is separation by technique as Random Forest shows an accuracy of 100 %, BayesNet 98.75 %, Stochastic Gradient Descent (SGD) 98.25%, Sequential Minimal optimization (SMO) 95.75%, Multinomial Logistic Regression (MLR) 95.75% respectively.