To Detect Early Sepsis Disease using Clinical Data using MLP Classifier, AdaBoost Classifier, Gradient Boosting Classifier, GaussianNB, Linear Discriminant Analysis, Quadratic Discriminant Analysis.
Sepsis is caused by the body's immune system which works continuously to prevent the infection. The large amount of toxic compounds emitted into the blood creates general discomfort at this point in time. The practicality of identifying the onset of sepsis is an essential consideration for the individual to do this. The primarily goal is to develop and test Multiple classifiers with information available as electronic clinical health data and to predict outcome of class label as sepsis or no-sepsis for invisible health records. The secondary goal is to compare the accuracy of the multiple classifiers against each other models like MLP Classifier, AdaBoost Classifier, Gradient Boosting Classifier, GaussianNB, Linear Discriminant Analysis, Quadratic Discriminant Analysis using the metrics accuracy and log loss.