An Empirical Study of Various Machine Learning Approaches in Prediction of Chronic Kidney Disease

In today’s era everybody is trying to be conscious about health. Although, due to workload and busy schedule, one gives attention to the health when any major symptoms occur. But Chronic Kidney Disease (CKD) is a disease which doesn’t shows symptoms it is hard to predict, detect and prevent such a disease and this can lead to permanently health damage, but some machine learning algorithms can come handy in this aspect for their efficient prediction and analysis. By using data of CKD, patients with 25 attributes and 400 records we are going to use various machine learning techniques like Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree etc. The purposes of our work is to virtuously predicting Chronic Kidney disease and have a comparative analysis among some of the popular machine learning based approaches based on some performance metrics. In our work, it is found that the Random Forest algorithm outperforming other machine learning based approaches we used in the experiment.