dm.cs.tu-dortmund.de/en/mlbits/outlier-distance/
Distance-Based Approaches – Lecture Notes
and Zhou, Z.-H. 2012. Isolation-based anomaly detection. ACM Trans. Knowl. Discov. Data . 6, 1 (2012), 3:1–3:39. DOI: 10.1145/2133360.2133363
[RaRaSh00]
Ramaswamy, S., Rastogi, R. and Shim, K. 2000. Efficient [...] get:
\[\hat\varepsilon (o) = k\textup{-dist}(o) \quad \text{ with } k=\lceil \pi \cdot |D| \rceil -1\]
\(o\) is a \(DB(\varepsilon , \pi )\) -Outlier
iff \(\varepsilon \lt{} \hat\varepsilon (o) = k\te [...] NN-Weight-Outlier are those objects, with the largest distance sum to their \(k\) -nearest neighbors. 1
Both methods can be accelerated well with indexes to accelerate \(k\textup{NN}\) -queries.
There are …
