Role of attributes selection in classification of Chronic Kidney Disease patients

In the present days the Chronic Kidney Disease (CKD) is a common problem to the public. CKD is generally considered as kidney damage and is usually measured with the GFR (Glomerular Filtration Rate). Several researchers from health care and academicians are working on the CKD problem to have an efficient model to predict and classify the CKD patient in the initial stage of CKD, so that the necessary treatment can be provided to prevent or cure CKD. In this work classification models have been built with different classification algorithms, Wrappersubset attribute evaluator and bestfirst search method to predict and classify the CKD and non CKD patients. These models have applied on recently collected CKD dataset downloaded from the UCI repository. The models have shown better performance in classifying CKD and non CKD cases. Results of different models are compared. From the comparison it has been observed that classifiers performed better on reduced dataset than the original dataset.