0
$\begingroup$

I have a Super Market. I want to find if product A is out of stock which product should i replace with. I am not sure what should i do, someone suggested me K-means for that. If sppose my data looks like

enter image description here

I would like understand what should i handle these categorical values(red, yellow, sweet sour, veg etc). I can convert then into values but that would be random and scale them at the end so now we will get rid of dominant features. What if i start giving weights. to these categories of an attribute based on total sales by color, total sales by type veg or non veg. And if i give weights to values what would be its impact.

Improve this question
$\endgroup$
2
  • $\begingroup$ You could create dummy variables for each of the categorical variables or alternately, perform factor analysis and then use the resulting principal components as input to your kmeans $\endgroup$
    – SmitM
    Commented Aug 9, 2018 at 18:57
  • 2
    $\begingroup$ Stack Overflow isn't the right place to ask about choosing a statistical method since that's not really a specific programming question. This question has been asked and answered already on a more appropriate site: datascience.stackexchange.com/questions/22/… $\endgroup$
    – MrFlick
    Commented Aug 9, 2018 at 20:05

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.