Comparison Between Various Machine Learning Classification Techniques in Detecting Chronic Kidney Disease

Abstract

Application of Artificial Intelligence and related fields are increasing day by, rather than being confined to the core areas of computer science, they are spreading into various new domains. In recent times, Machine Learning i.e. a sub-domain of AI has been widely used in order to assist medical experts and doctors in the prediction, diagnosis, and prognosis of various diseases and other medical disorders. In this manuscript, the authors applied various machine learning algorithms to a problem in the medical diagnosis domain and analyzed their performance efficiency in predicting the results. The selected problem is the diagnosis of Chronic Kidney Disease. The used dataset consists of 400 instances and 24 attributes. The authors evaluated 10 classification algorithms by applying them to the Chronic Kidney Disease Dataset. The predictions by the candidates were compared with the actual medical results of the subject to calculate the candidates’ efficiency. The performance evaluation metrics are predictive accuracy, specificity, sensitivity, and precision. The results indicate that RF performed best with 100%, specificity of 1, sensitivity of 1, and precision of 1.