Prediction of the Tuberculosis Patients Who Can Recover Normally Using a Support Vector Machine with Radial and Polynomial Kernels

Tuberculosis is a contagious disease that is generally transmitted through a sufferer’s cough and is deadly. Tuberculosis usually attacks the lungs but can also affect other parts of the body. Treatment of tuberculosis patients who do not recover with those who can recover is different, mishandling can cause death in patients. Therefore, we need a system that can predict whether the patient’s condition can recover normally, or the lungs cannot be recovered. Support vector machine is a learning system that uses a hypothetical linear function in a high dimensional space and is trained with an algorithm based on optimization theory by applying learning bias derived from statistical theory. In this study, the kernel function is used, namely the radial kernel and the polynomial. Based on the analysis and discussion that has been done, it can be concluded from this study that the performance of the radial and polynomial kernels is the same with an accuracy of 85% and a sensitivity value of 94%.