2
$\begingroup$

We have a great post to discuss the drawbacks of K-means.

Can DBSCAN overcome these drawbacks? and what are the drawbacks of DBSCAN?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ DBSCAN meets certain k-means limits. Among them, DBSCAN does not need to set the cluster number a priori, and it can identify irregular, not necessarily spherical, cluster shapes. Nevertheless, it has difficulties identifying nested clusters, which has led to other versions such as OPTICS and Hdbscan. $\endgroup$
    – curiosus
    Commented Sep 2, 2020 at 14:15

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.

Browse other questions tagged or ask your own question.