In this section, we will see a step-by-step example using Naive Bayes (NB) algorithm. As already stated, NB is highly scalable, requiring a number of parameters linear in the number of variables (features/predictors) in a learning problem. This scalability has enabled the Spark community to make predictive analytics on large-scale datasets using this algorithm. The current implementation of NB in Spark MLlib supports both the multinomial NB and Bernoulli NB.
Bernoulli NB is useful if the feature vectors are binary. One application would be text classification with a bag of words (BOW) approach. On the other hand, multinomial NB is typically used for discrete counts. For example, if we have a text classification ...
