You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for publishing this DBCV implementation. I believe, however, that there is an error in the logic. On page 842 of the paper, regarding the minimum spanning tree computations, the paper states:
Based on the MRDs, a Minimum Spanning Tree (MSTMRD ) is then built. This process is repeated for all the clusters in the partition, resulting in l minimum spanning trees, one for each cluster.
Thank you for publishing this DBCV implementation. I believe, however, that there is an error in the logic. On page 842 of the paper, regarding the minimum spanning tree computations, the paper states:
In this implementation, however, it appears that only one MST is being created for the entire data set: https://github.com/christopherjenness/DBCV/blob/master/DBCV/DBCV.py#L90
The text was updated successfully, but these errors were encountered: