Cluster Definition in DBSCAN

In [[DBSCAN]], clusters are defined as follows - [Core Point: A point is a core point if at least minPts of points exist within a radius of epsilon centered at that point. - [Directly Reachable: If point p is a core point and point q is within distance ε of p, then q is considered directly reachable from p. - Reachable: If there is a sequence of consecutive directly reachable points from point p to point q, then q is considered reachable from p. - [Density-Connected: Points p and q are density-connected if they are reachable from each other via some point o.

Based on these concepts, DBSCAN defines clusters as follows

• Cluster: the largest set of points for which all points are density-connected to each other and from which any point in the cluster can be reached from any other point.
• Noise point: not belonging to any cluster.

Thus, DBSCAN defines clusters based on data density and identifies noise points.

---

This page is auto-translated from /nishio/DBSCANにおけるクラスタの定義 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.