SVM parameter selection based on harmony search with an application to hyperspectral image classification.
Oguzhan CeylanGülsen Taskin KayaPublished in: SIU (2016)
Keyphrases
- parameter selection
- harmony search
- hyperspectral image classification
- harmony search algorithm
- ls svm
- support vector machine svm
- support vector
- metaheuristic
- differential evolution
- hyperspectral
- active learning
- genetic algorithm
- bp neural network
- hyperspectral images
- simulated annealing algorithm
- optimization algorithm
- hill climbing
- generalization ability
- knn
- support vector machine
- optimization method
- svm classification
- svm classifier
- feature selection
- model selection
- training set
- variable selection
- machine learning
- feature extraction
- optimal parameters
- search space
- hyperspectral data
- feature vectors
- simulated annealing
- optimization problems
- rbf neural network
- cross validation
- artificial neural networks
- feature space
- pattern recognition
- similarity measure
- radial basis function
- neural network
- remote sensing