Performance Comparison of Some Classifier for Chronic Kidney Disease

Abstract: Nearly 10 percent of the world's population is affected by a major chronic kidney disease health issue. However, systematic and automatic methodologies are evidently used to predict chronic kidney disease. Machine training is one of the very kind methodologies. The classifier in the machine learning algorithms can provide known features and unknown class to the test samples with class labels. Existing works with machine learning algorithms do not provide the predictive accuracy to the extent required. To satisfy the gap, this initiative offers a new approach for the classification of acute renal illness with environmental variables from the medical dataset. The goal of this initiative is to forecast renal illness through the use of machine learning algorithm that supports vector machine (SVM). The primary goal of this study is to forecast renal disease using classification algorithms such as closest neighbour Optimal Fuzzy-K and Support-Vector-Machine. This study work concentrates primarily on discovering the finest classification algorithm based on the precise classification and output time variables. It is noted from the experimental outcomes that the SVM's output is a lot productive than the closest neighbouring machine Optimal Fuzzy-K. The precision is regarded to be the main measure for quality evaluation, and it is proven that the suggested technique offers greater precision in classification. Keywords: Data mining, machine learning, chronic kidney disease