DBSCAN

DBSCAN is a density-based clustering algorithm introduced by Martin Ester and colleagues in 1996. Unlike partitioning methods such as k-means, DBSCAN defines clusters as contiguous regions of high point density separated by regions of lower density. The algorithm requires two parameters: epsilon (eps), the maximum distance between two points to be considered neighbors, and minPts, the minimum number of points required to form a dense region.

DBSCAN classifies each point as a core point, border point, or noise point. A core point has at least minPts neighbors within its eps-neighborhood. A border point falls within the eps-neighborhood of a core point but does not itself meet the minPts threshold. A noise point is neither core nor border. The algorithm iteratively expands clusters from core points by visiting their eps-neighbors and recursively including any newly discovered core points, forming arbitrarily shaped clusters.

In GISGISGeographic Information Systems (GIS) enable users to analyze and visualize spatial data to uncover patterns, relation..., DBSCAN is applied to identify geographic clusters in point datasets such as crime incidents, GPSGPSThe Global Positioning System (GPS) is a satellite-based navigation system operated by the U.S. Space Force that prov... trajectories, species sightings, and social media check-ins. Urban analysts use it to detect activity centers from mobile phone data, while environmental scientists cluster pollution measurements to delineate contaminated zones. Retail analysts group customer locations to identify trade area cores, and transportation planners detect congestion hotspots from vehicle probe data.

DBSCAN requires no prior knowledge of the number of clusters and naturally handles noise. It discovers clusters of arbitrary shapes, outperforming k-means on non-convex patterns. However, it struggles when clusters have widely varying densities, as a single eps value cannot accommodate all clusters simultaneously. Variants like HDBSCAN and OPTICS address this limitation by adapting density thresholds.