term - Calculating B and |V| in naive bayes text classification -
i found link multinomial naive bayes classifier
how calculate b'
or |v|
?
the page said number of terms in vocabulary. in example, how 6
b
? counting of term?
"chinese", "beijing", "shanghai", "meacao", "tokyo", "japan"
one more question, if new term appear in testing document? example, in doc 6 appears "bangkok" or new word never appear before. how count probability of new term ?
you right. it's total number of words in vocabulary, since there can 1 entry term in vocabulary.
Comments
Post a Comment