A New Feature Selection Method Based on Stability Theory - Exploring Parameters Space to Evaluate Classification Accuracy in Neuroimaging Data.
Jane M. RondinaJohn Shawe-TaylorJanaina Mourão MirandaPublished in: MLINI (2011)
Keyphrases
- classification accuracy
- feature selection
- input data
- feature subset
- synthetic data
- training data
- test data
- data sets
- data structure
- feature set
- statistical methods
- training samples
- clustering method
- pairwise
- parameter space
- irrelevant features
- missing values
- missing data
- dimensional data
- estimation process
- method for feature selection
- naive bayes
- parameter estimation
- text classification
- data mining techniques
- support vector machine
- probabilistic model
- prior knowledge
- bayesian methods
- machine learning
- parameter estimates
- learning algorithm
- similarity measure
- measured data
- support vector
- correlation analysis
- preprocessing
- training set
- noisy data
- high dimensional data
- model selection