Prediction-of-Risk-of-Sepsis-using-ML-models-on-big-data

Determine the development of sepsis in hospital-administered patients earlier than the clinical predictions.
Data has been collected from over 1 million ICU patients with up to 40 clinical variables from 3 separate hospitals.
Various machine learning model was tried like Logistic Regression, SVM, Naïve Bayes, Random Forest.
Random Forest performs the best with an accuracy of 87% and f1 score of 91%.