DBSCAN is a density-based method for [[clustering]]. First define a radius $\epsilon$ and check how many objects are within the radius. If the number of points within radius $\epsilon$ exceeds a specified minimum number of points, the object is considered a core object. Identify all core objects that are **density-connected** in the neighborhood and any border objects that are **density-reachable** (within $\epsilon$ of a core object).