A Metric Multidimensional Scaling-Based Nonlinear Manifold Learning Approach for Unsupervised Data Reduction.
Matthieu BrucherChristian HeinrichFabrice HeitzJean-Paul ArmspachPublished in: EURASIP J. Adv. Signal Process. (2008)
Keyphrases
- data reduction
- multidimensional scaling
- nonlinear manifold learning
- low dimensional
- dimensionality reduction
- manifold learning
- unsupervised learning
- geodesic distance
- data compression
- euclidean distance
- metric learning
- high dimensional
- high dimensionality
- high dimensional data
- semi supervised
- principal component analysis
- preprocessing
- data analysis
- singular value decomposition
- feature selection
- data points
- euclidean space
- vector space
- classification accuracy
- metric space
- dimension reduction
- distance measure
- knowledge discovery
- feature space
- cluster analysis
- data mining
- classification rules
- model selection
- rough set theory
- feature extraction
- pattern recognition
- kernel pca
- supervised learning
- image processing
- distance metric
- data sets
- machine learning