DBSCAN — Density-Based Spatial Clustering of Applications with Noise

Fact

An algorithm that groups crowded points and marks lonely points as noise.

In Plain Words

DBSCAN reads data like a school dance map. Snack-table crowds are groups. One kid doing the robot alone is noise.

It finds weird-shaped groups in messy data. It can also flag lonely points as outliers.

Related Concepts

• Clustering
• Unsupervised Learning
• k-Means Clustering
• Outlier Detection

Clustering
DBSCAN groups nearby points by density.

Unsupervised Learning
DBSCAN finds groups without human labels.

K-Means Clustering
DBSCAN handles weird-shaped groups better than K-Means.

Outlier Detection
DBSCAN marks lonely low-density points as noise or outliers.