Operator Reference: Naive Bayes
RuhrMining
New Altair Community Member
Hey folks,
I have an issue with the documentation of the naive bayes classifier for which I seek clarification.
In particular "To complete the probability model, it is necessary to make some assumption about the conditional probability distributions for the individual Attributes, given the class. This Operator uses Gaussian probability densities to model the Attribute data. " (copied&pasted from the help file). That statement is incomplete as polynomial data is also supported - which is of course a great feature, but confusing if not mentioned in the reference.In addition, as far as my testing goes, laplace correction in Rapidminer only is applied to binomial/polynomial data and not with gaussians. For the latter point it would be great to know if this is how laplace is actually implemented.
I have an issue with the documentation of the naive bayes classifier for which I seek clarification.
In particular "To complete the probability model, it is necessary to make some assumption about the conditional probability distributions for the individual Attributes, given the class. This Operator uses Gaussian probability densities to model the Attribute data. " (copied&pasted from the help file). That statement is incomplete as polynomial data is also supported - which is of course a great feature, but confusing if not mentioned in the reference.In addition, as far as my testing goes, laplace correction in Rapidminer only is applied to binomial/polynomial data and not with gaussians. For the latter point it would be great to know if this is how laplace is actually implemented.
Tagged:
0
Answers
-
Glück Auf @RuhrMining !Maybe have a look at this video: https://academy.rapidminer.com/learn/article/naive-bayes-are-we-playing-golf-today It's not the most academic one, but this explains in detail what we do for polynominal values.Best,Martin
0
Altair Employee