Statistical and predictive analytics of chronic kidney disease

Abstract

Currently, health problems increasingly intrigue the curiosity of data scientists. In fact, data analytics as a rapidly evolving area can be the right solution to manage, detect and predict diseases which threaten human life and cause a high economic cost to health systems. This paper seeks to establish a statistical and predictive analysis of an available dataset related to chronic kidney disease (CKD) by employing the widely used software package called IBM SPSS. Indeed, we manage to create a 100% accurate model based on XGBoost linear machine learning algorithm for successful classification of patients into; affected by CKD or not affected.