Abstract:

Chronic Kidney Disease (CKD) is a long-term condition that develops gradually over time. In its advanced stages, it becomes life-threatening and can only be managed through regular dialysis or a kidney transplant. Detecting CKD early is crucial, as timely treatment can help slow its progression or even prevent severe complications. This paper focuses on analyzing different classification techniques—Decision Tree, Random Forest, and Logistic Regression—to predict CKD. The study compares these algorithms using various evaluation metrics on a dataset sourced from the standard UCI repository.

Description:

.