negative values with density plot

msacs09msacs09 Member Posts: 55 Contributor II
when using auto model i got this density plot. 
I do not have negative values. How do i remove negative values and what operator is used to plot density plots

Best Answer

Answers

  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn
    Hi @msacs09,

    Can you share your dataset in order to understand ?

    Regards,

    Lionel
  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn
    Hi again @msacs09,

    After reflection, I think you have to keep these curves like that (although you have no negative values) : With Naive Bayes, when dealing with continuous data, a typical assumption is that the continuous values associated with each class are distributed according to a Gaussian distribution : 



    Regards,

    Lionel

  • msacs09msacs09 Member Posts: 55 Contributor II
    Thank you. Is there a way i can a do a density chart without negative values?? what operator should i use to get this graph with in Rapidminer?
  • msacs09msacs09 Member Posts: 55 Contributor II
    Waoow thank you. wanted to try this right away
  • msacs09msacs09 Member Posts: 55 Contributor II
    edited February 2019
    Also, I used "Kernel Naive Bayes" and still i get the same negative values.. further is there a way we can reduce the x axis values.. for better clarity
  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn
    What is being explained is that while there may not be negative values in your data (and negative values may in fact be impossible based on the quantity being measured), the NB algorthim models your numerical data as though they are continuous functions based on the Gaussian distribution, which means it will show the full possible range of values in the built-in charts.  This is a simplification, of course, of reality, but one that allows the NB algorithm to produce its estimates with minimal computational complexity.  If you want to customize the axis of the graphs to omit the negative range or cut off the long right tail, you'll have to do that manually (as Ingo showed above) by setting the min and max values for the X axis, which is pretty easy to do in the new charting interface in RapidMiner 9.2.

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
  • msacs09msacs09 Member Posts: 55 Contributor II
    Thank you all
Sign In or Register to comment.