In alaineiturria/otsad: Online Time Series Anomaly Detectors

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

otsad 0.2.0

• New Features
  • EORELM-AD R6 class that implements our ensemble-based online recurrent extreme learning machine anomaly detector (EORELM-AD).
  • DinamycNormalizer R6 class that implements the unsupervised dynamic z-score standardization proposed by Bollegala.
  • WindowNormalizer R6 class allows normalizing the data set in an online manner one-by-one or chunk-by-chunk.
  • AdaptiveNormalizer R6 class with our adaptation of the method proposed by Ogasawara et al. for one-pass online time-series adaptive normalization.
  • AdaptiveNormalizer2 R6 class that implements our adaptation to one-pass online processing of the adaptive normalization method proposed by Gupta and Hewett.
  • AnomalyLikelihoodScorer R6 class that implements the anomaly likelihood introduced by Ahmad et al.
  • DynamicThresholdScorer R6 class that implements the online anomaly scoring method (based on the prediction errors) proposed by Buda et al.
  • SigmaScorer R6 class that allows computing the anomaly score online based on historical prediction errors and 3-sigma control limits.
  • DynamicSigmaScorer R6 class that allows computing the anomaly score online based on the prediction errors and 3-sigma control limits.

alaineiturria/otsad documentation built on Jan. 12, 2023, 12:26 p.m.