A machine learning-based early diagnosis model for chronic kidney disease using SPegasos

Chronic Kidney Disease is now one of the most severe illnesses that requires an immediate diagnosis. Previous research has shown that machine-learning techniques are reliable enough for medical care. Clinicians and medical staff can detect disease on time thanks to significant results obtained from machine learning classifier algorithms. Furthermore, by utilizing unbalanced and small datasets of chronic kidney disease, this study provides medical system developers with insights to aid in the chronic kidney disease early prediction, reducing the effects of late diagnosis, particularly in low-income and difficult-to-reach areas. In this paper, a new machine learning-based early diagnosis model is presented for chronic kidney disease. Furthermore, we used the SMOTE technique to remove all noisy data from the two proposed datasets during data pre-processing. Finally, the WEKA tool is used to evaluate the performance of the proposed prediction model using machine learning algorithms.