Skip to content
Quantoros

Why can a point be an outlier to LOF without being globally far from the dataset center?

Classic MLLevel: IntermediateLocal anomaly comparison

Why can a point be an outlier to LOF without being globally far from the dataset center?

Hints
  • The reference is a neighborhood, not a global center.
  • Density ratios can reveal local gaps.

Answer

LOF compares a point's local density with its neighbors' densities. A relatively sparse point near a dense cluster can be anomalous even when its global distance is unremarkable.

Source and licence: scikit-learn 1.7.2 — outlier detection · BSD-3-Clause · answer: Quantoros editorial

Related questions