Comparative study of different machine learning algorithms for chronic kidney disease prediction

ABSTRACT

Chronic kidney disease (CKD) is a huge problem on the health-care system due to its growing occurrence, high risk of process to final phase kidney disease, and increased death rate. It is gradually increasing into a major public health problem. An early diagnostic method to measure renal function is necessary and highly needed in many respects. Machine Learning technique uses understanding of various key medical problems, including diabetes projection, heart observation, and coronavirus detection. Machine learning may be a better solution to achieve higher performance in the prediction of CKD and may be effective for other diseases as well. The main purpose of this study is to determine the optimal and appropriate machine learning techniques which may effectively identify and predict the CKD status. The dataset from Kaggle has been used for this study. Machine learning models such as Random Forest, Decision Tree and SVM have been used in this research to predict chronic kidney disease. Therefore, the results showed that Random Forest provides the highest accuracy in identifying chronic kidney disease. The overall accuracy of an algorithms used in this research is basically greater than that of previous research, intimating that this research is also more reliable than previous model.