0
$\begingroup$

I have a data set of 70 stores with a sales column (ranging from 50M to 70M) and 39 other features, like age group, income categories etc. I need to find the clusters based off of these metrics.

A sample dataset will have fields like this:

Store#:   0101  
Sales:    50M
Customers Between 20 and 40: Low
Customers Between 40 and 60: Low
Customers Between 60 and 80: High
Income<40: Low
40<Income<60: High 
60<Income<90: Low 

So you would read the dataset as follows: Store #0101 has 50M worth of sales, Low number of customers between 20 and 40, low number of customers between 40 and 60 etc.

Here are my questions:

  1. Does it make sense to convert Low, Medium, High to 0,0.5 and 1 respectively?
  2. Does sales need to get normalized between 0 and 1 ?
  3. Is there any other models than K-Means to address the clustering?
$\endgroup$

1 Answer 1

1
$\begingroup$

K-means is not appropriate for such data.

It is designed for continuous variables where the L2 error needs to be minimized. That does not seem to make sense for your data.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.