A person might develop potentially fatal chronic kidney disease as a result of kidney cancer or impaired kidney function, and condition can persist throughout their lives. It is possible to prevent or delay the worsening of this chronic illness until dialysis or surgery are the only options left for patients to survive. This is more likely to occur if disease is detected early and treated appropriately. This study examined viability of employing several machine learning (ML) techniques to identify this disease in its first stages. A great deal of study has been carried out concerning this subject. However, we are using predictive modelling to strengthen our method. So, we look at the objective class traits and relationships across data components in our method. Because predictive modelling allows us to integrate better measurements of qualities. ML algorithms and data analytics may be employed to construct a repertoire of prediction models. Beginning with 25 features including target class characteristics, 13 unique ML classifiers have been assessed in supervised training environment. ML algorithm Logistic Regression gives the best performance with 99.29% accuracy, a precision of 99.09%, and a 99.08% recall of LR. Based on findings, it seems that combining predictive modelling with current advancements in ML might be a promising strategy for discovering novel methods to evaluate the predictive power of models used to diagnose CKD and other conditions.