Prediction of Hospital Admission Using Machine Learning

Abstract

The process of hospital admission is fraught with difficulties for patients. They will have to wait in line for admittance together for hours if it is a busy hospital. However, it is really poor in the ER. The Emergency Room is just for the most critical of conditions. In order to improve patient flow and avoid overcrowding, we must use more cutting-edge methods. Thus, using data mining methods, we may learn a nice way to foretell the ED. Admissions. In this article, we looked at the math behind three different types of prediction models—Naive Bayes, Random Forests, and Support Vector Machine—to determine which one is the most effective. Age, gender, systolic pressure, diastolic pressure, diabetes, prior records in the preceding month or year, and admission are all potential predictors of hospitalization. We also provide specifics on the algorithms we used. With the goal of better prediction, we categories the data using the Random Forests technique