"NaiveBayes algorithm"

clopes
clopes New Altair Community Member
edited November 5 in Community Q&A
Hello,

Right now I'm using the NaiveBayes classifier from Rapidminer 4.6 and I need to know which form of nb is implemented,
  • Multi-variate Bernoulli NB
  • Multinominal NB
  • Multi-variate Gauss NB
Reference: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.61.5542&rep=rep1&type=pdf
Tagged:

Answers

  • land
    land New Altair Community Member
    Hi,
    unfortunately I don't have the time to explain everything in detail, BUT:
    Each one is implemented in RapidMiner. First we have to distinguish between Flexible Bayes and the rest. Flexible Bayes is implemented by Naive Bayes (Kernel). The rest is performed by the standard Naive Bayes operator. Which of the algorithms below is implemented by the operator depends on the input data. If you give it in a boolean nominal form, I think it will collapse to MultiVariate Bernoulli. If in polynominal form the Multinominal NB will be used. If numerical attributes are given the Gauss NB will be used.

    Greetings,
      Sebastian

Categories