Semi-supervised and unsupervised novelty detection using nested support vector machines.
Frank de MorsierMaurice BorgeaudChristoph KüchlerVolker GassJean-Philippe ThiranPublished in: IGARSS (2012)
Keyphrases
- novelty detection
- semi supervised
- support vector
- large margin classifiers
- learning machines
- semi supervised learning
- text filtering
- supervised learning
- svm classifier
- labeled data
- unlabeled data
- anomaly detection
- pairwise
- unsupervised learning
- active learning
- classification accuracy
- kernel function
- weakly supervised
- sentence level
- support vector data description
- training examples
- loss function
- knn
- neural network
- text categorization
- support vectors
- support vector machine
- maximum margin
- query processing
- image analysis
- supervised methods
- training data
- image processing
- minimum volume