Data-Specific Feature Selection Method Identification for Most Reproducible Connectomic Feature Discovery Fingerprinting Brain States.
Nicolas GeorgesIslem RekikPublished in: CNI@MICCAI (2018)
Keyphrases
- synthetic data
- input data
- irrelevant features
- feature subset
- data sets
- feature set
- feature selection
- noisy data
- preprocessing
- test data
- prior knowledge
- missing values
- statistical methods
- data analysis
- redundant features
- training data
- data structure
- knowledge discovery
- discriminative features
- feature extraction
- similarity measure
- classification method
- training samples
- text classification
- data sources
- missing data
- clustering method
- feature vectors
- classification accuracy
- mutual information
- human brain
- feature weighting
- support vector machine
- correlation analysis
- feature ranking
- face recognition
- machine learning