A Novel Hyper-parameters Selection Approach for Support Vector Machines to Predict Time Series.
Yanhua YuMeina SongJunde SongPublished in: J. Comput. (2012)
Keyphrases
- hyperparameters
- support vector
- learning machines
- cross validation
- model selection
- gaussian process
- closed form
- variational bayes
- bayesian inference
- prior information
- posterior distribution
- support vector machine
- bayesian framework
- incremental learning
- random sampling
- ls svm
- maximum likelihood
- em algorithm
- noise level
- bayesian methods
- sample size
- support vector regression
- feature selection
- conjugate priors
- maximum a posteriori
- missing values
- pairwise
- similarity measure
- image processing