Naive Bayes vocabulary a set?

Question

I am reading this document to get a handle on naive bayes. It should link to page 35

https://web.stanford.edu/class/cs124/lec/naivebayes.pdf#page=35

given two documents "abb" and "cdd" would vocabulary be:

{a,b,b,c,d,d}, |Vocabulary| == 6

or: {a,b,c,d}, |Vocabulary| == 4

just need a sanity check, thanks

Answer 1

The vocabulary would be {a,b,c,d} . Vocabulary refers to the set of terms the classifier is working with, sometimes called the feature space of the classifier.