1
$\begingroup$

I have a dataset of 10,000 customers with ten features all of which are Likert type. Like this:

customer feature1 feature2 feature3
ID1 3 1 5
ID2 4 5 4
ID3 3 5 1
ID4 1 3 2
ID5 2 5 1
ID6 1 3 4

I want to use K_mean for clustering this database in python. I use:

from sklearn.preprocessing import MinMaxScaler
scaler=MinMaxScaler()
data_normal=scaler.fit_transform(data)
from sklearn.cluster import KMeans

k=4
model = KMeans(n_clusters=k)
pred=model.fit_predict(data_normal)
  1. What should I do after these steps for interpreting my results?
  2. And if I want to see the features in each cluster, what should I do?
$\endgroup$
2
  • $\begingroup$ Welcome to DataScienceSE. I don't know the precise way to do that with python but the main idea is that you should probably use the centroids* of the clusters as the representative point of the cluster. Also I don't think you need to use scaling with likert scores since they are all on a standard range of values. $\endgroup$
    – Erwan
    Nov 18 at 18:00
  • $\begingroup$ @Erwan, Thanks for your response. $\endgroup$
    – SA12
    Nov 19 at 16:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.