stray

stray-package

<p>This package is a modification of <code><a rd-options="HDoutliers" href="/link/HDoutliers?package=stray&version=0.1.0&to=HDoutliers" data-mini-rdoc="HDoutliers::HDoutliers">HDoutliers</a></code> package. <code>HDoutliers</code> is a powerful algorithm for the
detection of anomalous observations in a dataset, which has (among other advantages) the ability to detect
clusters of outliers in multi-dimensional data without requiring a model of the typical behavior of the system.
However, it suffers from some limitations that affect its accuracy. In this package, we propose solutions to
the limitations of HDoutliers, and propose an extension of the algorithm to deal with data streams that exhibit
non-stationary behavior. The results show that our proposed algorithm improves the accuracy, and enables the
trade-off between false positives and negatives to be better balanced.</p>


This is a modification of 'HDoutliers' package. The 'HDoutliers' algorithm is a powerful
unsupervised algorithm for detecting anomalies in high-dimensional data, with a
strong theoretical foundation. However, it suffers from some limitations that
significantly hinder its performance level, under certain circumstances. This package
implements the algorithm proposed in Talagala, Hyndman and Smith-Miles (2019)
<arXiv:1908.04000>  for detecting anomalies in high-dimensional data
that addresses these limitations of 'HDoutliers' algorithm. We define an anomaly as an observation that deviates markedly from the majority
with a large distance gap. An approach based on extreme value theory is used
for the anomalous threshold calculation.

Priyanga Dilini Talagala

Anomaly Detection in High Dimensional and Temporal Data

Rob J Hyndman

Kate Smith-Miles

stray function

<p>This package is a modification of <code><a rd-options='HDoutliers' href='HDoutliers'>HDoutliers</a></code> package. <code>HDoutliers</code> is a powerful algorithm for the
detection of anomalous observations in a dataset, which has (among other advantages) the ability to detect
clusters of outliers in multi-dimensional data without requiring a model of the typical behavior of the system.
However, it suffers from some limitations that affect its accuracy. In this package, we propose solutions to
the limitations of HDoutliers, and propose an extension of the algorithm to deal with data streams that exhibit
non-stationary behavior. The results show that our proposed algorithm improves the accuracy, and enables the
trade-off between false positives and negatives to be better balanced.</p>

stray: stray: A package for robust anomaly detection in data streams with concept drift

Description

Arguments

References

See Also