Naïve Bayes should come to mind when you are considering creating a model from a labeled dataset, for problems or application for which the features are conditionally independent. Its simplicity and robustness make Naïve Bayes one of the most widely applied supervised learning techniques.
This chapter illustrates the versatility of Naïve Bayes for text mining applications.
However, it should be noted that the requirement of feature independence cannot always be met. In the case of the classification of documents or news releases, Naïve Bayes incorrectly assumes that terms are semantically independent: the two entities age and date of birth are highly correlated. The discriminative classifiers described in the next few chapters address some of Naïve Bayes' limitations [5:14].
This chapter does not treat temporal dependencies, sequence of events, or conditional dependencies between observed and hidden features. These types of dependency necessitate a different approach to modeling, which...