Feature selection method based on mutual information and class separability for dimension reduction in multidimensional time series for clinical data.
Liying FangHan ZhaoPu WangMingwei YuJianzhuo YanWenshuai ChengPeiyu ChenPublished in: Biomed. Signal Process. Control. (2015)
Keyphrases
- dimension reduction
- mutual information
- feature selection
- class separability
- clinical data
- similarity measure
- unsupervised learning
- statistical analysis
- high dimensional
- preprocessing
- feature space
- image registration
- feature extraction
- learning algorithm
- real world
- databases
- least squares
- principal component analysis
- feature set
- image data
- feature vectors
- pairwise
- support vector machine svm
- clustering method
- high dimensionality