Recursive feature selection based on non-parallel SVMs and its application to hyperspectral image classification.
Gülsen Taskin KayaYucel TorunCaglar KucukPublished in: IGARSS (2014)
Keyphrases
- feature selection
- hyperspectral image classification
- support vector
- microarray classification
- selected features
- active learning
- multi class
- text categorization
- mutual information
- hyperspectral images
- feature selection algorithms
- machine learning
- feature ranking
- feature set
- feature space
- knn
- classification accuracy
- high dimensionality
- support vector machine
- feature extraction
- hyperspectral
- feature subset
- multi task
- text classification
- dimensionality reduction
- data sets
- hyperplane
- kernel function
- image data
- model selection
- feature vectors
- hyperspectral data