Pneumonia Prediction System Using Classification and Regression Trees Algorithm

Pneumonia is an inflammation or chronic infection of lung tissue caused by various microorganisms, including parasites, viruses, and bacteria, as well as physical damage to the lungs and exposure to chemicals. The method used to calculate predictions uses the CART (Classification and Regression Trees) algorithm. The model is then implemented into a website-based prediction system. The purpose of this study was to determine the implementation of the CART algorithm to determine pneumonia and to determine the accuracy of the CART algorithm in predicting pneumonia. The average accuracy of the results of this study led to an accuracy value of 94%, r-square 87%, precision 95%, recall 94%, and f1 score 94% of the total dataset of 283. The results of this study get the best r-square on the 5th test with an accuracy of 85%.