Talk:Naive Bayes classifier/Archives/2017

Specify Probability Density Function : Gaussian Naive Bayes
In the Gaussian naive Bayes section, I think it should be made clear that whats given is just the Probability density function, and the actual probability should be something like $$p(x = x_0 | C_i) = \int_{x_0-\epsilon}^{x_0+\epsilon}f(x|\;\mu_i ,\sigma ^{2}_i)dx$$

right? Not so sure about the epsilon...

Where the probability density of the normal(gaussian) distribution is,

$${\displaystyle f(x\;|\;\mu ,\sigma ^{2})={\frac {1}{\sqrt {2\sigma ^{2}\pi }}}\;e^{-{\frac {(x-\mu )^{2}}{2\sigma ^{2}}}}}$$

Please note that the density of a continuous random variable is not a probability, but is the derivative of the cumulative probability function, in a similar way to instantaneous velocity not being a distance.

Aditya 05:47, 18 February 2017 (UTC) Thanks