Development of a Web-Based Application for Predicting Stroke Patient Emergency Levels Using the Naïve Bayes Algorithm

Abstract

Stroke is a serious disease that requires prompt and appropriate treatment; therefore, determining the level of patient Emenrgency Levels is critically important. This study aims to develop a web-based application for predicting the Emenrgency Levels level of stroke patients using the Naïve Bayes algorithm as a classification method. The research data were obtained from the medical records of stroke patients at RSUP Dr. M. Djamil Padang during March and April 2025, with a total of 222 data samples. The attributes used in this study include age, gender, address, length of stay, ward class, BPJS insurance membership status, and comorbidities, with Emenrgency Levels status as the class attribute classified into Emenrgency and non-Emenrgency. The application was developed as a web-based system to facilitate easy access for medical personnel in utilizing the prediction system. The experimental results indicate that the Naïve Bayes algorithm achieved an accuracy of 77.48% with an error rate of 22.52%. The findings of this study are expected to assist medical personnel in supporting faster and more objective decision-making regarding the Emenrgency Levels level of stroke patients.