Improving classification performance of Support Vector Machine by genetically optimising kernel shape and hyper-parameters.
Laura DiosanAlexandrina RogozanJean-Pierre PécuchetPublished in: Appl. Intell. (2012)
Keyphrases
- support vector
- support vector machine
- hyperparameters
- cross validation
- kernel function
- svm classifier
- kernel methods
- feature space
- gaussian process
- model selection
- support vector machine svm
- decision function
- gaussian processes
- histogram intersection kernel
- feature vectors
- generalization ability
- machine learning
- prior information
- bayesian methods
- bayesian framework
- feature selection
- incremental learning
- random sampling
- genetic programming
- linear classifiers
- feature extraction
- training data
- bayesian inference
- maximum likelihood
- knn
- bayesian networks
- ls svm
- parameter settings
- sample size
- higher order
- closed form