Density-based clustering in scenarios with variable density

This repository contains two algorithms devoted to solve different problems regarding variable-density scenarios: K-DBSCAN and V-DBSCAN.

K-DBSCAN

Kernel-Density-Based Spatial Clustering of Applications with Noise (K-DBSCAN) aims at identifying arbitrarily-shaped groups of points within a significantly sparse sample-space, without previous knowledge of the amount of resulting clusters.

V-DBSCAN

Variable-Density-Based Spatial Clustering of Applications with Noise (V-DBSCAN) is a multi-scale variation of DBSCAN that takes into account the variations in density when moving away from the centroid of the data.

Code

https://github.com/plasavall/kdbscan_vdbscan

References

  • E. Pla-Sacristán, I. González-Díaz, T. Martínez-Cortés and F. Díaz-de-María, “Finding landmarks within settled areas using hierarchical density-based clustering and meta-data from publicly available images”, Expert Systems with Applications (ESWA). Vol. 123; pp. 315 – 327, 2019; Netherlands. DOI: https://doi.org/10.1016/j.eswa.2019.01.046

Comments are closed.