Talk:Naive Bayes classifier/Archives/2016

This is an archive of past discussions about Naive Bayes classifier. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

What does this mean?

In the Gender Classification example it says:

"Note that a value [1.5789] greater than 1 is OK here – it is a probability density rather than a probability, because height is a continuous variable."

This sentence makes no sense... I haven't heard of a pdf value that is greater than 1, and in any case - this is not the real value of this pdf!

If I'm not mistaken, for continuous distribution you need to choose some distance (a,b) around the actual value, for example (5.855-0.5, 5.855+0.5) - and just make sure to have the same distance for all the calculations, (i.e. in the example, both for male and female) - this way the effect of the different distributions are constant. — Preceding unsigned comment added by 5.22.130.201 (talk) 19:30, 6 June 2016 (UTC)

Ad 'Constructing a classifier from the probability model'

Shouldn't be

{\hat {y}}={\underset {k\in \{1,\dots ,K\}}{\operatorname {argmax} }}\ p(C_{k})\displaystyle \prod _{i=1}^{n}p(x_{i}\vert C_{k}).

replaced by

k={\underset {k\in \{1,\dots ,K\}}{\operatorname {argmax} }}\ p(C_{k})\displaystyle \prod _{i=1}^{n}p(x_{i}\vert C_{k}).

? — Preceding unsigned comment added by 195.187.80.9 (talk) 07:30, 8 June 2016 (UTC)