Text.Language.Classifier.NaiveBayesian (Text v0.2.0) View Source
A language detection model that uses n-gram frequencies.
It multiplies the frequencies of detected
n-grams. Since the frequencies are stored
as log(frequency) the addition of the
log(frequency) entries is the same as
frequency * frequency.
Link to this section Summary
Functions
Return the {language score} tuples
in the correct order for this classifier.
Sums the frequencies of each n-gram
Link to this section Functions
Return the {language score} tuples
in the correct order for this classifier.
Sums the frequencies of each n-gram
A strong negative weighting is applied if the n-gram is not contained in the given vocabulary.