Is it worth it? An experimental comparison of six deep- and classical machine learning methods for unsupervised anomaly detection in time series.
Ferdinand RewickiJoachim DenzlerJulia NieblingPublished in: CoRR (2022)
Keyphrases
- unsupervised anomaly detection
- anomaly detection
- semi supervised
- subsequence matching
- weather forecasting
- learning environment
- multivariate time series
- dynamic time warping
- non stationary
- principal component analysis
- deep learning
- sequential data
- data sets
- probabilistic model
- long term
- pattern recognition
- data structure
- bayesian networks
- neural network