Share research findings or your experience on the effect of the parameter k on the performance of the k-means clustering technique using cluster distance performance?
Hi,
If you increase k, more clusters can be formed which implies fewer objects in one cluster. read more here at the link
https://towardsdatascience.com/k-means-clustering-from-a-to-z-f6242a314e9a
Dear Shakir Faris Tuleab and Lubaba Yemer - thank you for your comments!!
Dear followers,
Here is how one can use the elbow technique to learn the best k value in the cluster. One will operate k-means clustering on the dataset for a variety of k value more than 1 and for every value of k total of squared errors (SSE) plot a line graph. If the line graph seems like an arm, then the "elbow" on the arm is the value of k that is the best. The aim is to pick a small amount of k that still has a low SSE. However, the elbow method doesn't always work well; especially if the data is not very clustered.
Elbow method to find out the best k-value in the k-means algorithm!!
Dear قبل بداية و Co-Seakers و L Rivière
- thank you for your reply!!- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs