notes

Personal notes
git clone git://git.laack.co/notes.git
Log | Files | Refs

DBSCAN.md (571B)


      1 # DBSCAN (Density based spatial clustering of applications with noise)
      2 
      3 ML D5
      4 
      5 **Definition:** DBSCAN is a clustering algorithm that groups clusters by continuous regions of high density.
      6 
      7 Steps to perform:
      8 1. For each instance count how many instances are in the neighborhood
      9 2. If it has at least min_samples instances in neighborhood it is a core instance (located in dense area)
     10 3. All instances in the neighborhood of a core instance belong to the same cluster
     11 4. All other instance that are not core instances and do not have one in the neighborhood are anomalies.