Optimization of Tuberculosis Diagnosis Using the Support Vector Machine Method on Health Data of Central Sulawesi Province

Tuberculosis (TB) is a disease caused by infection with the bacterium Mycobacterium tuberculosis in the lungs. TB is a major global health problem, ranking as the second leading cause of death from infectious diseases worldwide. Given that TB is one of the infectious diseases that remains a public health problem, early and accurate detection is key in controlling its spread. This study uses historical data of TB patients from several hospitals in Central Sulawesi as research samples. One of the classification methods is the Support Vector Machine (SVM) method. This method was chosen because of its ability to classify non-linear data and its potential to produce high prediction accuracy with complex data. This study aims to improve the accuracy and efficiency of Tuberculosis (TB) disease diagnosis in Central Sulawesi Province through the application of the SVM method. The parameters used in this study were age, gender, body temperature, shortness of breath, chest pain, sputum examination, and final diagnosis. The results of this study show that handling the problem of imbalanced data with an approach at the data level using the Adaptive Synthetic Sampling (adasyn) method approach and SVM as a classification method for two classes in this study shows that this method can classify each research data used with an accuracy rate of more than 95%. The classification results show that the larger the data used for testing, the greater the classification accuracy. The best classification accuracy value is obtained from the scheme 4 division of training and testing data, namely 90%:10% with a value of 100%. This shows that the method used in this study can be used to help diagnose TB disease based on a patient's medical data. From this research, we introduce a web-based information system that can be used to conduct early detection for TB patients independently, where they are unwilling or embarrassed to come to the hospital.