1
$\begingroup$

I'm working on a project where I use several clustering methods, mainly density based ones such as hdbscan, optics... I'm looking for a metric to evaluate clustering results that takes into account outliers and different forms of clusters. One of the evaluation metrics I found is DBCV, it hasn't received enough attention in the datascience community, so I'm not sure about its robustness. Also in runtime it is unsuitable when we have several thousand points, even in two dimensions.

DBCV source code: https://github.com/christopherjenness/DBCV

$\endgroup$
2
  • $\begingroup$ What exactly do you mean by robustness here? Also to say, many clustering criteria, even the majority perhaps, are meant for small/medium number of points. But you always can draw a random sample from the enterity of points and run your criterion on the sample. $\endgroup$
    – ttnphns
    Feb 12, 2020 at 10:13
  • $\begingroup$ @ttnphns Robustness could be defined through several characteristics: being able to 1) correctly evaluate clusters of arbitrary shapes and not necessarily spherical or almost spherical clusters, 2) evaluate nested clusters (one inside the other), 3) take into account in the quality score calculation the outliers. Taking a random sample of the final result to produce a score is not convincing if one wants to compare different algorithms in front of experts in this field, especially in research. $\endgroup$
    – curiosus
    Feb 13, 2020 at 9:58

1 Answer 1

6
$\begingroup$

I know it's a little bit late but I just wanted to say that I'm currently studying density based clustering algorithms, and I found out the most suitable metric was in fact DBCV:

1) it deals with noise (which is intrinsic to the definition of the density-based clustering, and it's not taken into account in indexes such as Silhouette or Davies Bouldin)

2) it allows you to capture the shape of each cluster creating an MST employing 'density', no distances (you can manage arbitrary-shaped clusters, which is not possible if you use metrics like the ones mentioned above)

Here, you can check out the DBCV paper for a better understanding.

I tried the same implementation than you did, but finally found a better one, used in the hdbscan implementation from the sci kit learn contribution repository (it's faster bc it has many functions coded in C).

Heard that another commonly used index is CDbw, because it let you choose how many representatives for each cluster you want to use. However, all clusters will have the same amount of representatives, and also the number must be specified by the user, which is not desirable due to the fact it is another parameter that must be tuned properly .... In the DBCV paper you will find a comparison between many metrics, including CDbw.

Cheers!

$\endgroup$

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.