Skip to main content

> ML_ALGORITHM // HIERARCHICAL-DENSITY-BASED-CLUSTERING_v1.0

HDBSCAN & DBSCAN Density-Based Clustering

Hierarchical density-based clustering algorithm that discovers clusters of varying densities and isolates noise without requiring a specified k.

Density-Based Clusteringclassical-unsupervisedhigh-intrinsicmedium (1k-100k)
Back to All Algorithms
Computational Complexity
Training Complexity:O(n * log n) with spatial index
Inference Complexity:O(log n) approximate prediction
Hardware Profile
CPU Friendly:Yes
Requires GPU:No
Memory Footprint:moderate
Interpretability & Data
Interpretability Tier:high-intrinsic
Training Data Needs:medium (1k-100k)

Interpretability Assessment

Explicitly outputs an outlier/noise class (-1) without forcing non-conforming points into spurious clusters.

Suitable Tasks & Supported Modalities

Suitable Tasks:
clusteringanomaly detection
Supported Modalities:
tabularspatial

Implementing Libraries

scikit-learnscikit-learn Consortium / Inria · v1.5.2
View Spec
hdbscan

Foundational Literature

A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with NoiseMartin Ester, Hans-Peter Kriegel (1996) · ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD)
Density-Based Clustering Based on Hierarchical Density EstimatesRicardo J. G. B. Campello, Davoud Moulavi (2013) · Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)
Common Pitfalls & Warnings
  • Curse of dimensionality causing mutual reachability distances to become uniform and grouping everything as noise