Outliers detection in non-stationary time-series: Support vector machine versus principal component analysis.
Paulo GilHugo MartinsAlberto CardosoLuís Brito PalmaPublished in: ICCA (2016)
Keyphrases
- principal component analysis
- support vector machine
- dimension reduction methods
- detection accuracy
- false alarms
- multi class
- detection method
- detection algorithm
- svm classifier
- multi class support vector machines
- detecting outliers
- automatic detection
- detection rate
- outlier detection
- feature space
- machine learning
- training procedure
- object detection
- event detection
- k nearest neighbor
- feature extraction
- support vector
- principal components
- artificial neural networks
- independent component analysis
- low dimensional
- covariance matrix
- anomaly detection
- support vector machine svm
- dimensionality reduction
- neural network
- support vector regression
- support vector machine classifier
- small sample
- kernel function
- training set
- singular value decomposition
- knn
- noisy data
- dimension reduction
- data points
- linear discriminant analysis