Pošalji tekstualnu poruku: Learning naive Bayes regression models with missing data using mixtures of truncated exponentials