Chronic Kidney Disease Prediction based on Data Mining Method and Support Vector Machine

• Published in: 2022 IEEE 10th Conference on Systems, Process and Control, ICSPC 2022 - Proceedings
• Main Author: Bin Abdul Ghafar M.H.; Aleena Binti Abdullah N.; Abdul Razak A.H.; Syahirul Amin Bin Megat Ali M.; Mutalib Al-Junid S.A.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85146662869&doi=10.1109%2fICSPC55597.2022.10001806&partnerID=40&md5=48ea4aae8b5201c04b23ff2db6c35187
• DOI: 10.1109/ICSPC55597.2022.10001806

Chronic Kidney Disease (CKD) is when the kidneys are no longer working normally as they used to be, and filtering the blood was one of their obligations. The condition is classified as "chronic"since the kidney damage occurs gradually over time. This will cause waste to build up in the body. This study is aimed to predict the stages suffered by Chronic Kidney Disease patients, which might help in early detection and prevention. A Support Vector Machine (SVM) serves as the foundation for the prediction system developed by MathWorks and the missing data analysis will be done by using IBM SPSS Statistic 21. The work will show the feature selection and classification-based methods to enhance the performance accuracy of the algorithm in giving effective analysis and prediction of Chronic Kidney Disease. In conclusion, the accuracy achieved by using SVM with 50% holdout validation had the highest accuracy percentage of 93.5% out of other types of validation involved for the analysis of the datasets. © 2022 IEEE.