Chronic Kidney Disease (CKD) should be diagnosed earlier before kidneys fail to work. To help doctors or medical experts in prediction of CKD among patients easily, this paper has developed an expert system named Chronic Kidney Disease Prediction System (CKDPS) that can predict CKD among patients. The dataset used to develop CKDPS is taken from the Kaggle machine learning database. Before the implementation of CKDPS, different machine learning algorithms such as, k-Nearest Neighbors (KNN), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP) algorithm are applied on the dataset and their performances are compared to the matter of accuracy, precision and recall results. Finally, Random Forest algorithm is chosen to implement CKDPS as it gives 100% accuracy, precision and recall results. This paper also compares the accuracy results of different machine learning algorithms from different previous related works where same or different CKD dataset has been used.
CITATION STYLE
De, S., & Chakraborty, B. (2020). Development of Chronic Kidney Disease Prediction System (CKDPS) Using Machine Learning Technique. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 38, pp. 155–164). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-34080-3_18
Mendeley helps you to discover research relevant for your work.