An Empirical Evaluation of Machine Learning Techniques for Chronic Kidney Disease Prophecy

IEEE Access

Bilal Khan¹  Fazal Muhammad²  Rashid Naseem³  Ghulam Abbas⁴  Sunghwan Kim⁵ 

[1] Department of Computer Science, City University of Science and Information Technology, Peshawar, Pakistan;Department of Electrical Engineering, City University of Science and Information Technology, Peshawar, Pakistan;Department of IT and Computer Science, Pak-Austria Fachhochschule Institute of Applied Sciences and Technology, Haripur, Pakistan;Faculty of Computer Science and Engineering, GIK Institute of Engineering Sciences and Technology, Topi, Pakistan;School of Electrical Engineering, University of Ulsan, Ulsan, South Korea;

关键词: CKD;    ML techniques;    CHIRP;    MLP;    NB;    LR;   

DOI  :  10.1109/ACCESS.2020.2981689

来源: DOAJ

【 摘 要 】

Chronic Kidney Disease (CKD) implies that the human kidneys are harmed and unable to blood filter in the manner which they should. The disease is designated “chronic” in light of the fact that harm to human kidneys happen gradually over a significant time. This harm can make wastes to build up in your body. Many techniques and models have been developed to diagnos the CKD in early-stage. Among all techniques, Machine Learning (ML) techniques play a significant role in the early forecasting of different kinds ailments. ML techniques have been used for achieving analytical results which is one of the instruments utilize in medical analysis and prediction. In this paper, we employ experiential analysis of ML techniques for classifying the kidney patient dataset as CKD or NOTCKD. Seven ML techniques together with NBTree, J48, Support Vector Machine, Logistic Regression, Multi-layer Perceptron, Naïve Bayes, and Composite Hypercube on Iterated Random Projection (CHIRP) are utilized and assessed using distinctive evaluation measures such as mean absolute error (MAE), root mean squared error (RMSE), relative absolute error (RAE), root relative squared error (RRSE), recall, precision, F-measure and accuracy.The experimental outcomes accomplished of MAE are 0.0419 for NB, 0.035 for LR, 0.265 for MLP, 0.0229 for J48, 0.015 for SVM, 0.0158 for NBTree and 0.0025 for CHIRP. Moreover, experimental results using accuracy revealed 95.75% for NB, 96.50% for LR, 97.25% for MLP, 97.75% for J48, 98.25% for SVM, 98.75% for NBTree, and 99.75% for CHIRP. The overall outcomes show that CHIRP performs well in terms of diminishing error rates and improving accuracy.

【 授权许可】

Unknown