Author(s)

Arnold P. Dela Cruz^1,2, Gilbert M. Tumibay³

¹Nueva Ecija University of Science and Technology, Cabanatuan City, Philippines.
²Graduate School, Angeles University Foundation, Angeles City, Philippines.
³Angeles University Foundation, Angeles City, Philippines.

Tuberculosis remains an important problem in public health that threatens the world, including the Philippines. Treatment relapse continues to place a severe problem on patients and TB programs worldwide. A significant reason for the development of decline is poor compliance with medical treatments. The objectives of this research are to generate a predictive data mining model to classify the treatment relapse of TB patients and to identify the features influencing the category of treatment relapse. The TB patient dataset is applied and tested in decision tree J48 algorithm using WEKA. The J48 model identified the three (3) significant independent variables (DSSM Result, Age, and Sex) as predictors of category treatment relapse.

Data Mining, Decision Tree, J48, Tuberculosis, WEKA

Cruz, A. and Tumibay, G. (2019) Predicting Tuberculosis Treatment Relapse: A Decision Tree Analysis of J48 for Data Mining. Journal of Computer and Communications, 7, 243-251. doi: 10.4236/jcc.2019.77020.