A novel autoencoder approach to feature extraction with linear separability for high-dimensional data.
Jian ZhengHongchun QuZhaoni LiLin LiXiaoming TangFei GuoPublished in: PeerJ Comput. Sci. (2022)
Keyphrases
- high dimensional data
- linear separability
- feature extraction
- dimensionality reduction
- feature space
- linearly separable
- dimension reduction
- input space
- data points
- linear discriminant analysis
- hyperplane
- low dimensional
- manifold learning
- high dimensional
- high dimensionality
- nearest neighbor
- subspace clustering
- data sets
- high dimensional feature space
- data analysis
- principal component analysis
- similarity search
- data distribution
- pattern recognition
- feature vectors
- sparse representation
- lower dimensional
- original data
- input data
- data mining
- image classification
- face recognition
- low rank
- pattern classification
- support vector machine svm
- neural network
- random projections
- soft margin
- feature selection
- kernel function
- linear classifiers
- euclidean distance
- image processing