Chronic Kidney Disease Prediction

Chronic kidney disease is a serious health problem. With the help of Machine learning Techniques, doctors can predict it earlier. The research also contributes to Sustainable goal 3 - Basic Health and well being. To predict the disease different Machine learning Algorithms like logistic regression, Decision tree, Random Forest, Support Vector Classifier are used and best suitable algorithm is analyzed. Data prepossessing is done depending on the requirement. Training is given depending on the model chosen. It is found that Random Forest model gives best accuracy - 99.1% with all features. Further, it is trained by choosing the best five features arrived using chi-square test but the accuracy is 93.5% for same Random Forest Classifier. Again, it is trained by choosing the best three features arrived using chi-square test but the accuracy is 85% for same Random Forest Classifier. Performance analysis of different algorithms and choosing the algorithm based on true negative value in confusion matrix.