A support vector machine based technique for online detection of outliers in transient time series.
Hugo MartinsLuís Brito PalmaAlberto CardosoPaulo GilPublished in: ASCC (2015)
Keyphrases
- false positives
- detection accuracy
- automatic detection
- detection algorithm
- real time
- steady state
- object detection
- online learning
- detection rate
- missing data
- dynamic time warping
- online communities
- detection method
- non stationary
- multivariate time series
- outlier detection
- autoregressive
- novelty detection
- detecting outliers
- stock market
- change detection
- hidden markov models
- learning algorithm
- data sets