Risks of Chronic Kidney Disease Prediction using various Data Mining Algorithms

Abstract

Twenty million people have chronic kidney disease where patients experience a gradual deterioration of kidney function, the result of which is kidney failure. Early detection of chronic renal disease can help to slow its progression, avert complications, and reduce the risk of cardiovascular complications. Data mining has been broadly used in order to support medical professionals and physicians in the prediction and examination. Here, in this paper, multiple data mining algorithms are used to solve a problem in the field of medical diagnosis and examine how effective they were at predicting the consequences. The study's focus was on the diagnosis of chronic renal disease. This dataset used for this study consists 400 instances & 25 attributes. Preprocessing of the large amount of raw data is carried out to impute any missing data and determine which of the variables should be taken into account in the prediction models. The accuracy of the prediction is used to compare and contrast the various predictive analytic models

Keywords

• chronic kidney disease
• k-nearest neighbor
• classification
• predictive analytics
• decision tree
• data mining
• support vector machine
• random forest