Due to efficiency considerations, we choose Naive Bayes algorithm, which makes a conditional independent assumption: given the class
label, the attribute values of any example are independent. This assumption is almost certainly violated, but Domingos and Pazzani (1996) showed that it may not affect the classification decision. Details of the Naive Bayes can be found in (Langley, Iba, & Thompson, 1992), and we will not repeat it here.