Machine Learning-Based Framework for Predicting Smoke-Induced Noncommunicable Diseases

Still, the success rate of the prediction model for each patient smoking is yet lacking. In this connection, a good prediction model is important and necessary to predict smoking-related diseases. This study focuses on developing a prediction model for tuberculosis (TB) by using machine learning with the help of principal component analysis (PCA), a specific technique for factor analysis that generates linear combinations of variables to maximize the variance between the variables. To fulfill the inclusion criteria, the 259 data are acquired from the hill tracks area of Bangladesh through a field survey. Models with classification utilizing Support Vector Machine (SVM), Random Forest (RF), Naive Bayes (NB), and AdaBoost (AB) are developed to predict TB. A visual representation of the best prediction model’s result is provided graphically, along with the related scoring values.