DBSCAN: Density-Based Spatial Clustering

Density-Based Spatial Clustering of Applications with Noise (DBSCAN) - Watch the algorithm examine each point, calculate neighbors within epsilon (ε) radius, classify points as Core, Border, or Noise, and progressively form clusters without needing to specify K.

Speed:

ε:

MinPts:

Dataset

How DBSCAN Works

1. Parameters

ε (epsilon): Maximum distance for neighbors
MinPts: Minimum points to form dense region

2. Point Classification

Core: Has ≥MinPts neighbors within ε
Border: <MinPts neighbors, reachable from core
Noise: Not reachable from any core point

3. Cluster Formation

For each unvisited point, find neighbors. If core, expand cluster by connecting all density-reachable points. Continue until all points processed.

💡 Tip: Toggle "Show ε-Neighborhoods" and "Show Connections" to better understand how clusters form based on density connectivity.

Related Algorithms

K-Means

Unsupervised clustering

GMM

Probabilistic soft clustering

Hierarchical Clustering

Tree-based clustering with linkage methods

Understanding DBSCAN