MachineLearning/dbscan library

🌊 DBSCAN (Density-Based Spatial Clustering of Applications with Noise)

A simple DBSCAN implementation with eps-neighborhood and minPts parameters. Marks noise points with label -1 and returns cluster labels for each point.

Contract:

  • Input: X (n x m), eps (distance threshold), minPts.
  • Output: List
  • Error: throws ArgumentError on invalid input.

Time Complexity: O(n^2) naive neighborhood search Space Complexity: O(n)

Classes

DBSCAN