Robust Random Cut Forest Based Anomaly Detection on Streams

By Sudipto Guha, Nina Mishra, Gourav Roy, Okke Schrijvers. International Conference on Machine Learning, 2016

338 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

TensorCast: forecasting and mining with coupled tensors (2018, unknown): “For a cloud of points, an anomaly is a point that is ‘too far away’ from the rest [8]—this is exactly what outlier detection methods do (like the time-tested LOF [7], LOCI [27], and many more, up to the very recent Random-Cut-Forest-based anomaly detection [14]).”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “Examples of such methods include nearestneighbor-based methods [8], [9], [420]–[422], such as LOF [10] and partitioning tree-based methods [423], such as iForest [424], [425].”
ADBench: Anomaly Detection Benchmark (2022, background): “The presence of duplicated anomalies is also called the “anomaly masking” [55, 60, 100], posing challenges to many AD algorithms [25], e.g., the density-based KNN [11, 144].”
Continual Learning in Practice (2019, background): “A more efﬁcient approach based on a random cut forest [26] computes a sketch of the data upon which the anomaly detection decisions are made (this sketch can in fact be included as part of the sampler/sketcher component and re-purposed for other uses).”