Due to its increasing prevalence, chronic kidney disease (CKD) has a significant financial impact on the healthcare system, poor prognoses for death and health problems as well as a high probability of progressing to end-stage kidney disease. It is quickly developing into a global health emergency. This disease is largely caused by unhealthy eating habits and a lack of water consumption. The average lifespan of a person without kidney function is only 18 days, which requires dialysis and kidney transplantation. In order to correctly predict CKD in its early stages, reliable methodologies are essential. The ability to forecast CKD using Machine learning (ML) approaches is significant. Additionally, technology advancements typically have an advantageous effect on our everyday lives, particularly in the health. It is simpler to diagnose diseases presently than it was in the past because to improved technologies. In a analysis to identify The use of categorization algorithms for chronic kidney disease (CKD), this analysis’s main goal is to assess that attribute selection strategies affect the algorithms performance. The CKD data set utilised in the experiments was obtained from the machine learning repository at the University of California, Irvine (UCI). The process of designing classifier models involves a number of pre-processing, normalization, and attribute selection steps. Both the fuzzy rough set based attribute selection (FRSBAS) approach and the correlation based attribute selection (CBAS) method were utilized to identify the characteristics that have a significant effect on the classification outcomes. For performance analysis, the following metrics are used: accuracy, sensitivity, precision, and specificity.