Predict Chronic Kidney Disease using Data Mining Algorithms in WEKA Toolkit

This paper explains the chronic kidney disease prediction with the theory of various model evaluation mechanisms in data mining. From the given dataset, various statistical measures such as Accuracy, Error, Precision, Recall, F1 error, Elapsed Time used to analyze performances are described here. All of the above-mentioned parameters are compared in the tabular form for both the data mining classifiers, KNN and SVM are generated by using MATLAB itself. It is a two-step process, in first step KNN algorithm uses training dataset to build a classifier, and then in second step process, it uses this classifier to predict the class label of unlabeled data instance to perform the prediction of chronic kidney disease by accessing Hadoop in itself.