Hellinger distance-based stable sparse feature selection for high-dimensional class-imbalanced data.
Guang-Hui FuYuan-Jiao WuMin-Jie ZongJianxin PanPublished in: BMC Bioinform. (2020)
Keyphrases
- feature selection
- class imbalanced data
- high dimensional
- high dimension
- feature space
- sparse data
- high dimensionality
- dimensionality reduction
- low dimensional
- microarray data
- data points
- support vector machine
- gene expression data
- dimension reduction
- mutual information
- generalized linear models
- sparse pca
- variable selection
- distance measure
- text categorization
- classification accuracy
- support vector
- feature selection algorithms
- high dimensional data
- similarity search
- imbalanced data
- feature extraction
- metric space
- information gain
- text classification
- kullback leibler
- machine learning
- sparse coding
- nearest neighbor
- feature set
- multi dimensional
- outlier detection
- additive models
- model selection
- euclidean distance
- input space
- unsupervised learning
- selected features
- feature weighting
- classification models
- irrelevant features
- random projections
- cross validation
- feature selection and classification
- multi class
- knn
- feature subset
- data sets